Arjun

introDUction there is a lot of new process technology around. there can be few, if any, operations that have not been affected by the advances in process technology. and all indications are that the pace of technological development is not slowing down. this has important implications for operations managers because all operations use some kind of process technology, whether it is a simple internet link or the most complex and sophisticated of automated factories. But whatever the technology, all operations managers need to understand what emerging technologies can do, in broad terms how they do it, what advantages the technology can give and what constraints it might impose on the operation. Figure 8.1 shows where the issues covered in this chapter relate to the overall model of operations management activities.

process technology

Key questions

❯ What is process technology?

❯ What do operations managers need to know about process technology?

❯ how are process technologies evaluated?

❯ how are process technologies implemented?

8

Operations management

Direct

Design Develop

Deliver

Design

Layout and flow

Process design

Process technology

People in operations

Topic covered in this chapter

Figure 8.1 this chapter examines process technology

M08_SLAC8678_08_SE_C08.indd 246 06/02/16 4:57 PM

CHAPTER 8 PROCESS TECHNOLOGY 247

WHAT IS PROCESS TECHNOLOGY?

How operations managers deal with process technology is now one of the most important decisions that shape the capabilities of operations. This was not always the case, at least not for all operations. There used to be a simple division between those operations that used a lot of process technology, usually manufacturing operations, and those that used little or no process technology, usually service operations. But this is no longer true, and arguably has not been true for decades. High-volume services have for years understood the value of pro- cess technology. Online transactions for retail and other services are vital for their success. Yet even professional services such as legal and medical services can benefit from new and value-adding technologies (see the section on telemedicine later in this chapter).

So what do operations managers need to know about process technology? It must be important to them because they are continually involved in the choice, installation and man- agement of process technology. But operations managers are not (or need not be) technolo- gists as such. They do not need to be experts in engineering, computing, biology, electronics or whatever constitutes the core science of the technology. Yet they should be able to do three things. First, they need to understand the technology to the extent that they are able to articu- late what it should be able to do. Second, they should be able to evaluate alternative technol- ogies and share in the decisions of which technology to choose. Third, they must implement the technology so that it can reach its full potential in contributing to the performance of the operation as a whole. These are the three issues which this chapter deals with. This is illus- trated in Figure 8.2 and forms the structure of the chapter.

Process technology defined First, let us define what is meant by process technology. It is ‘the machines, equipment, and devices that create and/or deliver products and services’. Process technologies range from milk- ing machines to marking software, from body scanners to bread ovens, from mobile phones to milling machines. Disney World uses flight simulation technologies to create the thrill of space travel on its rides – just one in a long history of Disney Corporation and its ‘imagineers’ using technology to engineer the experience for their customers. In fact process technology is pervasive in all types of operations. Without it many of the products and services we all pur- chase would be less reliable, take longer to arrive and arrive unexpectedly, only be available in a limited variety, and be more expensive. Process technology has a very significant effect on quality, speed, dependability, flexibility and cost. That is why it is so important to operations managers, and that is why we devote a whole chapter to it. Even when technology seems peripheral to the actual creation of goods and services, it can play a key role in facilitating the

Question–What do operations managers need to know about process technology?

Stage 3 Implement the process

technology

Question–How does the process technology a�ect the operation?

Question–How can operations managers introduce new process technology smoothly?

Stage 1 Understand the

process technology

Stage 2 Evaluate the process

technology

Figure 8.2 The three stages of process technology management

M08_SLAC8678_08_SE_C08.indd 247 6/2/16 1:24 PM

248 PART TWO DESIGNING THE OPERATION

OPERATIONS IN PRACTICE

Back in 1920, a Czech playwright, Karel Capek, first coined the name ‘robot ’ (it comes from the Slavonic word for ‘work’). Since then, robots have moved from the stuff of science fiction to become a common, if not ubiquitous, element of mass production operations. There are more than a million industrial robots doing routine jobs on production lines. Robots do not take meal breaks, fall ill, complain or leave for better pay. They perform repetitive tasks cheaper than humans, give greater accuracy and repeatability, and can also be used where conditions are hazardous or uncomfortable for humans. Anyone who has seen the way that robots weld together automobile bodies, assemble complex prod- ucts, or load and unload work pieces onto a machine cannot fail to recognize the impact that robotics has had on manufacturing operations since robots were first introduced in the 1960s.

But like most new process technologies, the effect of robotics on operations management practice can be both positive and negative, depending on one’s perspective. (Film critics, who voted on Hollywood’s 50 greatest good guys and 50 greatest baddies, included a robot – the Terminator – on both lists.) Certainly they can save humans from exposure to danger. Robots were used during the clear-up operation among the rubble of the Twin Towers in New York. ‘ Enough people have died here ,’ said a spokesperson for the emergency services. ‘ We don’t want to risk any one .’ Bomb disposal squads use specialized robots which can take at least some of the risk from what remains a hazardous job. Nuclear power stations are decommissioned using robots to move, dis- mantle and manipulate hazardous radioactive material. They are also becoming both cheaper and more ver- satile in their production role. For example, Canon has announced its plans to move towards fully automating its digital camera production. Decades ago, Canon, like other manufacturers, began using cell production with teams or a single worker assembling a major part of the product, rather than repeating a simple task (see Chapter 6 ) . And over the years robots have been rou- tinely used as part of production cells. Canon calls it a ‘man–machine cell’, and says that ‘ human involvement will be phased out in making some products ’.

Only by substituting robots for people will produc- tion be kept in Japan, according to Canon, reversing the trend of Japanese manufacturers moving production to

China, India and the rest of Asia, where labour costs are cheaper. ‘ When machines become more sophisticated, human beings can be transferred to do new kinds of work ’, Jun Misumi, a Canon spokesperson, said. But it is the nature of the interface between people and robots that is concerning some experts. Akihito Sano, a professor at Nagoya Institute of Technology, has stressed the need for some way in which workers can communicate effec- tively so that robotic technology can be fine-tuned to become more practical. He also says, reassuringly, that there will always be room for human intelligence and skill. ‘ Human beings are needed to come up with inno- vations on how to use robots. Going [totally] to a no-man operation at that level is still the world of science fiction. ’ Yet people have always been nervous that new process technologies will take away their jobs. (Capek’s original play that gave robots their name described how, at first, they brought many benefits but eventually led to mass unemployment and unhappiness.) But there are some examples of a smooth introduction of robotics. Audi is said to have been successful in introducing industrial robots, partly because it asked its workers to suggest potential applications of robotics where they could both improve performance and then gave the same work- ers jobs supervising, maintaining and programming the robots. It may even be that robots can help defend manufacturing jobs in the rich world. For example, it has been pointed out that one reason why Germany has lost fewer such jobs than the UK is that it has five times as many robots for every 10,000 workers.

I, Robot 1

So u

rc e:

S h

u tt

er st

o ck

.c o

m : J

im m

i

M08_SLAC8678_08_SE_C08.indd 248 6/2/16 1:24 PM

CHAPTER 8 PROCESS TECHNOLOGY 249

direct transformation of inputs to an operation. For example, the computer systems which run planning and control activities, accounting systems and stock control systems can be used to help managers and operators control and improve the processes. This type of technology is called indirect process technology. It is becoming increasingly important. Many businesses spend more on the computer systems which control their processes than they do on the direct process technology which acts on its material, information or customers.

Process technology and transformed resources One common method of distinguishing between different types of process technology is by what the technology actually processes – materials, information or customers. We used this distinction in Chapter 1 when we discussed inputs to operations and processes.

Material-processing technologies These include any technology that shapes, transports, stores, or in any way changes physical objects. It obviously includes the machines and equipment found in manufacturing opera- tions (such as the robots described in the ‘Operations in practice’ case at the start of this chap- ter), but also includes trucks, conveyors, packing machines, warehousing systems and even retail display units. In manufacturing operations, technological advances have meant that the ways in which metals, plastics, fabric and other materials are processed have improved over time. Generally it is the initial forming and shaping of materials at the start, and the han- dling and movement through the supply network, that have been most affected by technology advances. Assembling parts to make products, although far more automated than it was once, presents more challenges.

Information-processing technology Information-processing technology, or just information technology (IT), is the most common single type of technology within operations, and includes any device which collects, manip- ulates, stores or distributes information. Arguably, it is the use of Internet-based technology (generally known as e-business) that has had the most obvious impact on operations – espe- cially those that are concerned with buying and selling activity (e-commerce). Its advantage was that it increased both reach (the number of customers who could be reached and the number of items they could be presented with) and richness (the amount of detail which could be provided concerning both the items on sale and customers’ behaviour in buying them). Traditionally, selling involved a trade-off between reach and richness. The widespread adoption of Internet-based technologies effectively overcame this trade-off. Also, the Internet had equally powerful implications on many other operations management tasks.

Customer-processing technology Although customer-processing operations were once seen as ‘low technology’, now process technology is very much in evidence in many services. In any airline flight, for example e-ticket reservation technology, check-in technology, the aircraft and its in-flight entertainment, all play vital parts in service delivery. Increasingly the human element of service is being reduced, with customer-processing technology used to give an acceptable level of service while signif- icantly reducing costs. There are three types of customer-processing technologies. The first category includes active interaction technology such as automobiles, telephones, Internet bookings and purchases, fitness equipment and cash machines (ATMs). In all of these, cus- tomers themselves are using the technology to create the service. By contrast, aircraft, mass transport systems, moving walkways and lifts, cinemas and theme parks are passive inter- active technology; they ‘processes’ and control customers by constraining their actions in some way. Some technology is ‘aware’ of customers but not the other way round: for example, security monitoring technologies in shopping malls or at national frontier customs areas. The

M08_SLAC8678_08_SE_C08.indd 249 6/2/16 1:24 PM

250 PART TWO DESIGNING THE OPERATION

objective of these ‘hidden technologies’ is to track customers’ movements or transactions in an unobtrusive way.

Integrating technologies Of course, some technologies process more than one type of resource. Many newer technolo- gies process combinations of materials, people and customers. These technologies are called integrating technologies. Electronic point-of-sale (EPOS) technology, for example, processes shoppers, products and information.

OPERATIONS IN PRACTICE

In his book, The Power of Habit , Charles Duhigg relates a story to demonstrate that human beings are more predictable than we sometimes like to think. A man walked into a supermarket to complain to the manager. The supermarket had been sending direct mail to the man’s daughter containing discount vouchers for baby clothes and equipment. ‘ She is only in high school ’, the father protested. The manager apologised profusely. It was the fault of a new program that predicted pregnancy based on the buying behaviour of their customers, he said. It was obviously a mistake and he was very sorry. A few days later, the man again visited the supermarket and said that it was his turn to apologise. His daughter was indeed pregnant and due to give birth due in a few months’ time. The point of the story is that technology is increasing in sophistication to the extent that it is now capable of performing tasks that previously required skilled people making judgements based on insight and experience. Moreover, technology can often do those tasks better. A piece of software has replaced the mar- keting team trying to guess who to sell baby clothes to. So technology is not only replacing people, but also ‘climbing the skills ladder all the time’.

Of course, technological advances have always had an impact on the type of jobs that are in demand by businesses, and, by extension, the type of jobs that are eliminated. So, much of the highly routine work of some mass manufacturing, or the type of standardized accounting processes that pay invoices, have been over- taken by the ‘the robot and the spreadsheet’. Yet the type of work that is more difficult to break down into a set of standardized elements is less prone to being dis- placed by technology. The obvious examples of work that is difficult to automate are the types of manage- ment tasks that involve decision making based on judge- ment and insight, teaching small children, diagnosing complex medical conditions, and so on. However, the future may hold a less certain future for such jobs. As the convenience of data collection and analysis becomes more sophisticated, and process knowledge increases,

it becomes easier to break more types of work down into routine constituents, which allows them to be automated. Carl Benedikt Frey and Michael Osborne, of the University of Oxford, maintain that the range of jobs that are likely to be automated is far higher than many assume, especially traditionally white-collar jobs such as accountancy, legal work, technical writing and (even) teaching. It is not simply that technology is get- ting cleverer; in addition it can exploit the capability to access far more data. Medical samples can be ana- lysed cheaper and faster by image-processing software than by laboratory technicians, case precedents can be sourced by ‘text-mining ’ programs more extensively than by para-legals, computers can even turn out news stories based on sports results or financial data. Frey and Osborne go so far as to estimate the probability that technology will mean job losses for certain jobs in the next two decades (bravely, because such forecasting is notoriously difficult). Among jobs most at risk are tele- marketers (0.99, where 1.0 = certainty), accountants and auditors (0.94), retail salespersons (0.92), technical writ- ers (0.89) and retail estate agents (0.86). Those jobs least likely to be replaced include actors (0.37), firefighters (0.17), editors (0.06), chemical engineers (0.02), athletic trainers (0.007) and dentists (0.004).

Technology or people? The future of jobs 2

So u

rc e:

S h

u tt

er st

o ck

.c o

m : D

ig it

al S

to rm

M08_SLAC8678_08_SE_C08.indd 250 6/2/16 1:24 PM

CHAPTER 8 PROCESS TECHNOLOGY 251

WHAT DO OPERATIONS MANAGERS NEED TO KNOW ABOUT PROCESS TECHNOLOGY?

Understanding process technology does not (necessarily) mean knowing the details of the science and engineering embedded in the technology. But it does mean knowing enough about the principles behind the technology to be comfortable in evaluating some technical information, capable of dealing with experts in the technology, and confident enough to ask relevant questions.

The four key questions In particular the following four key questions can help operations managers to grasp the essentials of the technology:

● What does the technology do which is different from other similar technologies?

● How does it do it? That is, what particular characteristics of the technology are used to perform its function?

● What benefits does using the technology give to the operation? ● What constraints or risks does using the technology place on the operation?

For example, return to the ‘Operations in practice’ case that discussed some developments in robotics. Now think through the four key questions.

● What does the technology do? Primarily used for handling materials, for example load- ing and unloading work pieces onto a machine, for processing where a tool is gripped by the robot, and for assembly where the robot places parts together. Some robots have some limited sensory feedback through vision control and touch control.

● How does it do it? Through a programmable and computer-controlled (sometimes multi- jointed) arm with an effector end piece which will depend on the task being performed.

● What benefits does it give? Can be used where conditions are hazardous or uncomforta- ble for humans, or where tasks are highly repetitive. Performs repetitive tasks at lower cost than using humans and gives greater accuracy and repeatability. Some robots are starting to mimic human abilities.

● What constraints or risks does it impose? Although the sophistication of robotic move- ment is increasing, robots’ abilities are still more limited than popular images of robot- driven factories suggest. Not always good at performing tasks which require delicate sensory feedback or sophisticated judgement. The human–robot interface needs managing carefully, especially where robotics could replace human jobs.

✽ ✽ ✽ Operations principle Operations principle Operations principle

Worked example

QB House speeds up the cut 3

It was back in 1996 when Kuniyoshi Konishi became so frustrated by having to wait to get his hair cut, and then pay over 3,000 yen for the privilege, that he decided there must be a better way to offer this kind of service. ‘ Why not ’, he said, ‘ create a no-frills barbers shop where the cus- tomer could get a haircut in ten minutes at a cost of 1,000 yen [€7] ? ’ He realized that a combi- nation of technology and process design could eliminate all non-essential elements from the basic task of cutting hair. How is this done? Well, first, QB House’s barbers never handle cash. Each shop has a ticket vending machine that accepts 1,000 yen bills (and gives no change!) and issues a ticket that the customer gives the barber in exchange for the haircut. Second, QB House does not take reservations. The shops do not even have telephones. Therefore, no

▼

M08_SLAC8678_08_SE_C08.indd 251 6/2/16 1:24 PM

252 PART TWO DESIGNING THE OPERATION

Emerging technologies – assessing their implications The four questions are universal, in the sense that they can help to understand the implica- tions for operations management of any new or emerging technology. By ‘implications’, we mean the natural consequence for the operation of adopting the technology. In other words, what would (or could) be the effects on the operation if the technology were included in the operation’s transforming resources.

In the rest of this section we look at three technologies that, at the time of writing, were new(ish). One processes materials (3D printing), one processes informa- tion (the Internet of Things) and one processes customers (telemedicine). The intention is not to provide a comprehensive survey of technologies – that could be expanded into a whole book – nor is it to delve into technical details. Rather it is to demonstrate how operations managers have to look beyond the technology in order to start to understand their implications.

receptionist is needed, or anyone to sched- ule appointments. Third, QB House devel- oped a lighting system to indicate how long customers will have to wait. Electronic sensors under each seat in the waiting area and in each barber’s chair track how many customers are waiting in the shop and dif- ferent coloured lights are displayed outside the shop. Green lights indicate that there is no waiting, yellow lights indicate a wait of about 5 minutes, and red lights indicate that the wait may be around 15 minutes. This system can also keep track of how long it takes for each customer to be served. Fourth, QB has done away with the tradi- tional Japanese practice of shampooing customers’ hair after the haircut to remove any loose hairs. Instead, the barbers use QB House’s own ‘air wash’ system where a vacuum cleaner hose is pulled down from the ceiling and used to vacuum the cus- tomer ’s hair clean. The QB House system has proved so popular that its shops (now over 200) can be found not only in Japan, but also in many other South-East Asian countries such as Singapore, Malaysia and Thailand. Each year almost 4,000,000 customers experience QB House’s 10-minute haircuts.

Analysis

● What does the technology do? Signals availability of servers, so managing customers’ expectations. It avoids hairdressers having to handle cash. Speeds service by substituting ‘air wash’ for traditional shampoo.

● How does it do it? Uses simple sensors in seats, ticket dispenser and air wash blowers. ● What benefits does it give? Faster service with predictable wait time (dependable ser-

vice) and lower costs, therefore less expensive prices. ● What constraints or risks does it impose? Risks of customer perception of quality of

service. It is not an ‘indulgent’ service. It is a basic, but value, service that customers need to know what to expect and how to use.

✽ ✽ ✽ Operations principle Operations principle Operations principle

So u

rc e:

P h

o to

gr ap

h er

s D

ir ec

t/ A

n d

y M

al u

ch e

M08_SLAC8678_08_SE_C08.indd 252 6/2/16 1:24 PM

CHAPTER 8 PROCESS TECHNOLOGY 253

3D printing (additive manufacturing) For decades and, in some industries, for centuries, producing physical products has been dominated by the principles of mass production. Standardized designs, repetitive pro- cesses and rigid, but productive process technology help to produce most of the items we use every day at (relatively) low cost. The downside of mass production was that vari- ety and customization are difficult to achieve at the same time as economies of scale. However, a process technology called 3D printing (also known as ‘additive manufactur- ing’) could have the potential to change fundamentally the economics of manufacturing, and in doing so challenge the dominance of mass production. But 3D printing is not a new technology as such. Since the 1990s designers have been using the technology to make prototype products or parts quickly and cheaply prior to committing to the expense of equipping a factory to produce the real thing. Yet the technology has advanced to the point where it is used, not just to make prototypes, but to produce finished products for real customers.

A 3D printer produces a 3D object by laying down layer upon layer of material until the final form is obtained. This is why it is also known as ‘additive manufacturing’, because, starting from nothing, successive layers are built up. This contrasts with ‘subtractive man- ufacturing’ that starts with more material than an item requires and reduces it through cutting, drilling, squeezing and other wise removing material until the finished form is reached. The process starts with a computer-based design which is ‘digitally decon- structed’ by software that takes a series of virtual digital slices through the design, details of which are sent to the 3D printer. Different materials can be used to build up the object from plastic to metals (and even food) and in various sizes limited only by the capacity of the printer.

Implications The obvious implication of 3D printing is the effect it has on the economics of production, especially the economics of making small quantities of novel and/or complicated items eco- nomically. The technology’s more enthusiastic proponents claim that, at last, the trade-off between speed and efficiency on the one hand, and flexibility and variety on the other, has been overcome. Most conventional process technology is at its most efficient when standard- ized products are made in large batches. But with 3D printing the cost of changing from one product to another is effectively zero. Also, because the technology is ‘additive’ it reduces waste significantly. Sometimes as much as 90 per cent of material is wasted in machining some aerospace parts, for example. It also enables a single ‘experimental’ item to be made quickly and cheaply, followed by another one after the design has been refined, as Ian Harris, from the Additive Manufacturing Consortium says: ‘It adds up to a new industry which reduces immensely the gap between design and production. Manufacturers will be able to say to their cus- tomers, “Tell us what you want” and then they will be able to make specific products for them.’ Some commentators even believe that 3D printing will challenge the advantage of low-cost, low-wage countries. As labour costs become less important, it is argued, manufacturers will return to make items close to their market.

The Internet of Things4

Back in 1973 the Universal Product Code or bar code was developed to enable a part or prod- uct type to be identified when read by a bar-code scanner. Now bar codes are used to speed up checkout operations in most large supermarkets. However, they also have a role to play in many of the stages in the supply chain that delivers products to retail outlets. During man- ufacture and in warehouses bar codes are used to keep track of products passing through processes. But bar codes do have disadvantages. It is sometimes difficult to align the item so that the bar code can be read conveniently, items can only be scanned one by one and, most significantly, the bar code only identifies the type of item not a specific item itself. That is, the

M08_SLAC8678_08_SE_C08.indd 253 6/2/16 1:24 PM

254 PART TWO DESIGNING THE OPERATION

code identifies that an item is, say, a can of one type of drink rather than one specific can. Yet these drawbacks can be overcome through the use of automatic identification technologies such as radio frequency identification (RFID). Here an electronic product code (ePC) that is a unique number 96 bits long is embedded in a memory chip or smart tag. These tags are put on individual items so that each item has its own unique identifying code. At various points dur- ing its manufacture, distribution, storage and sale each smart tag can be scanned by a wireless radio frequency ‘reader’. This can transmit the item’s embedded identify code to a network such as the Internet. See Figure 8.3.

Over the last several years the full potential of RFID technology has risen to a more revo- lutionary level, and one which has some important implications for operations management. Embedding physical objects with sensors and actuators (from vehicles to pharmaceuticals), and connecting them using wireless networks and the protocol that connects the Internet, allows information networks and physical networks to merge to form what has become known as ‘the Internet of Things’ (IoT). SAP, the developer of enterprise resource systems, describes the Internet of Things as follows: ‘A world where physical objects are seamlessly inte- grated into the information network, and where the physical objects can become active partici- pants in business processes. Services are available to interact with these “smart objects” over the Internet, query and change their state and any information associated with them, taking into account security and privacy issues.’5

Network analyses data to be used for monitoring and

process control Sensors ‘read’ item and transmit unique

code to network

RFID chip has a unique code number 96 bits long

Products have an RFID chip that transmits its

unique code

Figure 8.3 The Internet of Things (IoT) is a combination of RFID chips, sensors and Internet protocols that allows information on the location and state of physical objects to be networked

M08_SLAC8678_08_SE_C08.indd 254 6/2/16 1:26 PM

CHAPTER 8 PROCESS TECHNOLOGY 255

Implications According to some authorities the IoT promises to create new ways of doing business, the potential to improve processes, and more possibilities to reduce costs and risks. Putting sensors on ‘things’ gives information networks the ability to generate huge volumes of current data that can both sense the environment and communicate between the ‘things’. Operations managers can track and analyse the data to understand what is happening, even in complex systems, and respond quickly if necessary. This helps operations save significant amounts of money in lost, stolen or wasted products by helping manufacturers, distribution companies and retailers to pinpoint exactly the position and state of every item in the supply chain. So, for example, if a product had to be recalled because of a health-risk scare, the exact location of every potentially dangerous product could be immediately identified. Shoppers could easily scan a product to learn more about its characteristics and features while they are in the store, waiting at check- out counters could be eliminated because items will be scanned automatically by readers, the bill could even be automatically debited from your personal account as you leave the store. There are also potential benefits in tracking products after they leave the store. Data on how customers use products can be collected automatically and accurate recycling of waste mate- rials could be made considerably easier. McKinsey, the consultants, see six distinct types of emerging applications with implications for operations managers. These implications fall into two broad categories: first, information and analysis and, second, automation and control.

Information and analysis Because IoT networks link data from products, equipment, pro- cesses and the operating environment, they will produce enhanced information and more sophisticated analysis, which can augment operations management decisions. In particular three aspects of information and analysis could be affected:

● Knowing where things are – tracking will be easier because the movements of products and their interactions with processes will be monitored in real time. For example, some insur- ance companies will install location sensors in customers’ cars, allowing the insurer to base its fees on how a car is driven as well as where it travels.

● Knowing what is happening – the data from a large numbers of sensors, located in such infrastructural resources as roads and buildings, can report on conditions so that managers have an instantaneous awareness of events. For example, security systems can use sensor information from a combination of video, audio and vibration sensors to detect unauthor- ized entry to restricted areas.

● Knowing what to do – the IoT’s storage and computing power, when combined with advanced decision support systems, could significantly enhance decision making. For example, in retailing, shoppers can be monitored as they move through stores. Sensors record how long customers loiter at individual displays and record what they ultimately buy. The resulting data can help to optimize retail layouts.

Automation and control Controlling any operation or process involves monitoring what is actually happening within the operation or process, comparing what is actually happening with what should be happening, then making any necessary interventions to correct any devi- ations from what should be happening. So monitoring and data collection are at the heart of the control activity, and monitoring and data are what the IoT is particularly good at. When information is fed back through a network to some kind of automation that can intervene and modify process behaviour, control can be exercised (theoretically at least) without human intervention. Again, three aspects could be affected:

● Process optimization – processes that can be controlled can be more easily optimized. For example, in some semi-continuous processes in pulp and paper manufacturing, the requirement for the temperature of lime kilns to be continually adjusted limits their pro- ductivity. Yet by embedding temperature sensors in the process the kiln’s flame can be automatically adjusted to reduce temperature variance (and therefore increase quality) to near zero without frequent operator intervention.

M08_SLAC8678_08_SE_C08.indd 255 6/2/16 1:26 PM

256 PART TWO DESIGNING THE OPERATION

● Optimized resource usage – knowing exactly how much resource is being used can help in reducing costs. For example, some energy companies are providing customers with ‘smart’ meters that give visual displays showing energy usage and the real-time costs of provid- ing it. This allows domestic commercial customers to do things such as moving the use of energy-intensive processes away from peak energy demand periods to off-peak periods.

● Fast reactions – the most demanding use of the IoT involves rapid, real-time sensing of unpredictable circumstances and immediate responses governed by automated systems. The idea is for the IoT to imitate human decision makers’ reactions, but at a faster and more accurate level. For example, it could be possible for a group of robots to clean up toxic waste spills when detected.

However, the IoT does pose problems. There are technical challenges in integrating RFID chips into physical objects in such a way that makes sure that information is accurately trans- mitted. And although, as volume has increased, the cost of such chips and sensors has fallen, cost is still a factor in adopting the technology. But perhaps the most contested issues are those relating to customer privacy in extending data capture from products beyond the checkout. It is this issue that particularly scares some civil liberties activists. Keeping track of items within a supply chain is a relatively uncontentious issue. Keeping track of items when those items are identified with particular individuals going about their everyday lives is far more problematic. So, beyond the checkout, for every arguably beneficial application there is also potential for misuse. For example, smart tags could drastically reduce theft because items could automati- cally report when they are stolen, their tags serving as a homing device to pinpoint their exact location. But similar technology could be used to trace any citizen, honest or not.

Telemedicine6

The technological breakthroughs in medical care reported in the press often focus on those dramatic ‘miracle cures’ which have undoubtedly improved the quality of medical care. Yet a whole collection of changes in medical process technology has also had a huge impact on the way healthcare operations manage themselves. In particular, telemedicine has challenged one of the most fundamental assumptions of medical treatment – that medical staff need to be physically present to examine and diagnose a patient. No longer; web-connected devices are now able to monitor an individual’s health-related data and communicate the information to healthcare professionals located anywhere in the world. Doing this allows medical staff to be alerted to changing conditions as they occur, providing a status report of a person’s health so that the appropriate care can be administered. Telemedicine generally refers to the use of information and communications technologies for the delivery of clinical care. Formally, telemedicine is the ability to provide interactive healthcare utilizing modern technology and telecommunications. It allows patients to virtually ‘visit’ physicians – sometimes live, maybe using video links; sometimes automatically in the case of an emergency; sometimes where patient data is stored and sent to physicians for diagnosis and follow-up treatment at a later time. Telemedicine may be as simple as two health professionals discussing a case over the telephone, or as complex as using diagnostic algorithms and video-conferencing equipment to conduct a real-time consultation between medical specialists in different countries. The first interactive telemedicine system was developed and marketed in the USA by MedPhone Corporation in 1989. It operated over standard telephone landlines and was used for remotely diagnosing and treating patients requiring cardiac resuscitation. A year later the company introduced a mobile cellular version.

Broadly, there are three types of telemedicine: store-and-forward, remote monitoring and interactive services.

● Store-and-forward telemedicine – involves acquiring medical data such as medical images, blood test results, dermatological data, biosigns, etc., and then transmitting this data to a (remote) medical specialist at a convenient time for assessment offline. Because this does not require the presence of both parties at the same time, there is no actual physical

M08_SLAC8678_08_SE_C08.indd 256 6/2/16 1:26 PM

CHAPTER 8 PROCESS TECHNOLOGY 257

examination and sometimes no opportunity to collect a medical history. The store-and- forward process requires the clinician to rely on a medical record report and maybe audio/ video information as a substitute for a physical examination.

● Remote monitoring – allows medical professionals to monitor a patient remotely using various technological devices. This method is primarily used for managing chronic (long- lasting) diseases or specific conditions, such as heart disease. Because monitoring can be almost continuous, remote monitoring services can provide better, or at least comparable, health outcomes to traditional physician–patient interactions. In addition, they could be more convenient for both patient and doctor.

● Interactive telemedicine – involves real-time interactions between patient and provider. These could include online communication, telephone conversations and facilitated home visits by a non-specialist. This type of telemedicine is similar to traditional face-to-face vis- its by a physician, and normal activities such as history review, physical examination, psy- chiatric evaluations, etc., can be performed, at least partially.

Implications For communities in remote or isolated areas telemedicine can be particularly beneficial. Where previously no, or only a partial (or delayed), service was possible, it allows medi- cal services to be delivered. This is particularly important in developing countries. Known as ‘Primary Remote Diagnostic Visits’, a doctor uses devices to remotely examine and treat a patient. Telemedicine can also be useful in facilitating communication between a general practitioner and a specialist. All doctors need to seek advice. The easier, faster and cheaper it is to get this advice, the more likely they are to do it. The approach can also make use of deci- sion support diagnostic systems, which give accurate and consistent diagnoses. The quality of medical care in terms of accuracy of diagnosis and appropriateness of treatment is therefore enhanced by ‘virtually’ bringing specialist expertise to patients. New knowledge, improved medical practice, novel pharmaceuticals, the latest guidelines, and so on, can all be commu- nicated more effectively. Monitoring patients at home using standard equipment like blood pressure monitors and transmitting the information to a carer provides the basis for a faster emergency service. This is certainly true for situations where a physician is needed but no physician is present, such as on a passenger aircraft. For example, telemedicine kits are regu- larly used by pilots, cabin crew and other attending staff – non-medical experts who may have to deal with possible medical emergencies. They can use the kits to collect and transmit the data that would normally be collected in a hospital emergency room. This enables doctors, at a remote advice service, to help manage the medical emergency, make sure the right deci- sions are made and determine what treatment can be carried out and whether a diversion or medical evacuation is necessary.

Just as important in a world where some healthcare costs are likely to increase substan- tially, telemedicine has the potential to bring substantial cost savings. Requiring patients to visit physicians at their surgeries or hospitals is costly for the patient. Requiring doctors to visit patients at home can be even more expensive. Connecting through telemedicine reduces these costs dramatically. Patients having convenient access to medical advice may make fewer visits to the hospital. It is also family centred in the sense that the patient’s family life and work are less disrupted. More significantly, nurses can see up to 15 patients in four hours, whereas, visiting them in their home, they can see only 5 or 6 patients a day. Even when the costs of the technology are taken into account, telemedicine can represent a significant cost saving. Similarly, telemedicine can make the outsourcing of medical services easier. Primary-care physicians routinely outsource some services. For example, they take blood samples but send them to a specialist laboratory for analysis. With the more extensive use of telemedicine the data required for diagnostic decisions (for example, X-ray images) can be processed by a large- scale (therefore less expensive) specialist facility, possibly in a less expensive part of the world.

But there are issues with the adoption of telemedicine technology. One study7 found that there were three major barriers to the adoption of telemedicine in emergency and critical-care

M08_SLAC8678_08_SE_C08.indd 257 6/2/16 1:26 PM

258 PART TWO DESIGNING THE OPERATION

units. The first of these is the regulatory environment in some regions. Medicine must be (of course) a regulated activity, but the difficulty and cost of obtaining permission and/or licences, especially when multiple states and multiple facilities are involved, can be prohib- itive. Second, there can be a lack of acceptance by whoever pays for medical care, whether this is government or commercial insurance companies. This creates a major financial bar- rier because it puts the payment responsibility upon the hospital or healthcare system. Third, there may be cultural barriers, with some physicians unable or unwilling to adapt clinical procedures for telemedicine applications.

HOW ARE PROCESS TECHNOLOGIES EVALUATED?

The most common technology-related decision in which operations managers will be involved is the choice between alternative technolo- gies. It is an important decision because process technology can have a significant effect on the operation’s long-term strategic capability; no one wants to change expensive technologies too frequently. This means that the characteristics of alternative technologies need to be evaluated so that they can be compared. Here we use three sets of cri- teria for evaluation:

● Does the technology fit the processing task for which it is intended? ● How does the technology improve the operation’s performance? ● Does the technology give an acceptable financial return?

✽ ✽ ✽ Operations principle Operations principle Operations principle Operations principle Operations principle Operations principle

OPERATIONS IN PRACTICE

For those readers who live in regions of the world where Marmite is not a big seller, Marmite is ‘a nutritious savoury spread that contains B vitamins, enjoyable in a sand- wich, on toast , bread or even as a cook- ing ingredient ’. It is not to everyone’s taste, which is why it is advertised with the line ‘ you’ll either love it or hate it ’. But behind the clever advertising, Marmite, which is part of Unilever, the large food company, is a pioneer in recycling the leftovers from its production process to energy at the factory where it is made. The factory is in Burton upon Trent in the UK and every year around 18,000 tonnes of solidified Marmite deposit is left adhering to the surfaces of the machines and handling equipment that are used to produce the product . For years this residue was cleaned off and then either flushed into the sewerage system or sent to landfill sites. Then Unilever installed an anaerobic digester. This is a composter that uses the waste by-product where it is digested by microbes that feed on the waste. As they do, they release methane which is burned in a boiler connected to a generator

that produces power. The system also captures the waste heat that comes through the exhaust and helps heat the factory ’s water system. See Figure 8.4 . But the Marmite example is just one part of Unilever ’s ‘Sustainable Living Plan’, first published in 2010. Since then it has published an update every year on the progress it is making globally and nationally towards meeting its Sustainable Living Plan targets.

Love it or hate it, Marmite’s energy recycling technology 8

▼

So u

rc e:

A la

m y

Im ag

es : L

u ci

a La

n p

u r

M08_SLAC8678_08_SE_C08.indd 258 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 259

Does the process technology fit the processing task? Different process technologies will be appropriate for different types of operations, not just because they process different transformed resources, but also because they do so at dif- ferent levels of volume and variety. High-variety–low-volume processes generally require process technology that is general purpose , because it can perform the wide range of pro- cessing activities that high variety demands. High-volume–low-variety processes can use technology that is more dedicated to its narrower range of processing requirements. Within

Unilever publishes its performance against its Sustainable Living Plan targets as falling into three cat- egories. The first is ‘ areas where we are making genu- inely good progress ’. These included sustainable sourcing, nutrition and eco-efficiency (including the Marmite pro- ject). The second category is ‘ areas where we have had to consider carefully how to reach our targets but are now ready to scale up ’. For instance, a programme to increase the recycling rates of aerosols, encouraging more local councils to collect aerosols kerbside. ‘ However ’, the report admitted, ‘ we have more to do, working in part- nership with industry, Government and NGOs to help to increase recycling and recovery rates .’ The third category is ‘ areas where we are finding it difficult to make progress and will need to work with others to find solutions ’. This

included targets that require a change in consumer behaviour, such as encouraging people to eat foods with lower salt levels or reducing the use of heated water in showering and washing clothes.

Amanda Sourry, Unilever UK and Ireland Chairman, said: ‘ The old view of growth at any cost is unaccept- able; today the only responsible way to do business is through sustainable growth. It’s for this reason that the Unilever Sustainable Living Plan is not just a bolt-on strategy, it’s our blue-print for the future. Today’s progress update shows that we’ve made some fantastic steps for- ward, particularly in the areas of sustainable sourcing, health and nutrition and reducing greenhouse gases. Just one year into the decade-long plan, we are proud of our achievements so far but there’s still much more to do .’

The major material used in the process is waste material produced during the manufacture of Marmite paste. A large proportion of this waste is substances 'driven o�' during the evaporation stage.

This waste is a mixture of materials generated during the manufacture of Marmite paste.

The methane in 'bio-gas' is supplied to the site boiler house where it is burnt to produce steam

Steam, produced by burning bio-gas,

provides power for the factory. It heats the product stream and lowers evaporator

pressure.

Figure 8.4 Waste product recycling at Marmite

M08_SLAC8678_08_SE_C08.indd 259 6/2/16 1:27 PM

260 PART TWO DESIGNING THE OPERATION

the spectrum from general-purpose to dedicated process technologies three dimensions in particular tend to vary with volume and variety. Figure 8.5 illustrates these three dimen- sions of process technology:

● Its degree of ‘automation’. ● The capacity of the technology to process work, that is its ‘scale’ or ‘scalability’. ● The extent to which it is integrated with other technologies; that is, its degree of ‘coupling’

or ‘connectivity’.

The degree of automation of the technology To some extent, all technology needs human intervention. It may be minimal, for example the periodic maintenance interventions in a petrochemical refinery. Conversely, the person who operates the technology may be the entire ‘brains’ of the process, for example the sur- geon using keyhole surgery techniques. The ratio of technological to human effort it employs is sometimes called the capital intensity of the process technology. Generally processes that have high variety and low volume will employ process technology with lower degrees of auto- mation than those with higher volume and lower variety. For example, investment banks trade in highly complex and sophisticated financial ‘derivatives’, often customized to the needs of individual clients, and each may be worth millions of dollars. The back office of the bank has to process these deals to make sure that payments are made on time, documents are exchanged, and so on. Much of this processing will be done using relatively general-purpose technology such as spreadsheets. Skilled back-office staff are making the decisions rather than the tech- nology. Contrast this with higher volume, lower variety products, such as straightforward equity (stock) trades. Most of these products are simple and straightforward and are pro- cessed in very high volume of several thousand per day by ‘automated’ technology.

The scale/scalability of the technology There is usually some discretion as to the scale of individual units of technology. For example, the duplicating department of a large office complex may decide to invest in a single, very large, fast copier, or alternatively in several smaller, slower copiers distributed around the

Figure 8.5 Different process technologies are important for different volume–variety combinations

M08_SLAC8678_08_SE_C08.indd 260 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 261

operation’s various processes. An airline may purchase one or two wide-bodied aircraft or a larger number of smaller aircraft. The advantage of large-scale technologies is that they can usually process items cheaper than small-scale technologies, but usually need high volume and can cope only with low variety. By contrast, the virtues of smaller scale technology are often the nimbleness and flexibility that are suited to high-variety, lower volume process- ing. For example, four small machines can between them produce four different products simultaneously (albeit slowly), whereas a single large machine with four times the output can produce only one product at a time (albeit faster). Small-scale technologies are also more robust. Suppose the choice is between three small machines and one larger one. In the first case, if one machine breaks down, a third of the capacity is lost, but in the second, capacity is reduced to zero. The advantages of large-scale technologies are similar to those of large- capacity increments discussed in Chapter 4 .

The equivalent to scale for some types of information-processing technology is scalability . By scalability we mean the ability to shift to a different level of useful capacity quickly, and cost-effectively. Scalability is similar to absolute scale in as much as it is influenced by the same volume–variety characteristics. IT scalability relies on consistent IT platform architec- ture and the high process standardization that is usually associated with high-volume and low-variety operations.

The coupling/connectivity of the technology Coupling means the linking together of separate activities within a single piece of process technology to form an interconnected processing system. Tight coupling usually gives fast process throughput. For example, in an automated manufacturing system products f low quickly without delays between stages, and inventory will be lower – it cannot accumulate when there are no ‘gaps’ between activities. Tight coupling also means that flow is simple and predictable, making it easier to keep track of parts when they pass through fewer stages, or information when it is automatically distributed to all parts of an information network. However, closely coupled technology can be both expensive (each connection may require capital costs) and vul- nerable (a failure in one part of an interconnected system can affect the whole system). The fully integrated manufacturing system con- strains parts to flow in a predetermined manner, making it difficult to accommodate products with very different processing requirements. So, coupling is generally more suited to rela- tively low variety and high volume. Higher variety processing generally requires a more open and unconstrained level of coupling because different products and services will require a wider range of processing activities.

How does the technology improve the operation’s performance? In Chapters 2 and 3 , we identified the five operations performance objectives . So a sensible approach to evaluating the impact of any process technology on an operation is to assess how it affects the quality, speed, dependability, flexibility and cost performance of the operation. For example, consider a warehouse that stores spare parts which it packs and distributes to its customers. It is considering investing in a new ‘retrieval and packing’ system which converts sales orders into ‘retrieval lists’ and uses materials-handling equipment automatically to pick up the goods from its shelves and bring them to the packing area. The market requirements evaluation for this warehouse might be as follows:

● Quality – The impact on quality could be the fact that the computerized system is not prone to human error, which may previously have resulted in the wrong part being picked off the shelves.

● Speed – The new system may be able to retrieve items from the shelves faster than human operators can do safely.

✽✽✽ Operations principle Operations principle Operations principle Operations principle Operations principle Operations principle

M08_SLAC8678_08_SE_C08.indd 261 6/2/16 1:27 PM

262 PART TWO DESIGNING THE OPERATION

● Dependability – This will depend on how reliable the new system is. If it is less likely to break down than the operators in the old system were likely to be absent (through illness etc.), then the new system may improve dependability of service.

● Flexibility – New service f lexibility is not likely to be as good as the previous manual system. For example, there will be a physical limit to the size of products able to be retrieved by the automatic system, whereas people are capable of adapting to doing new things in new ways. Mix f lexibility will also be poorer than was previously the case, for the same reason. Volume (and perhaps delivery) f lexibility, however, could be better. The new system can work for longer hours when demand is higher than expected or deadlines are changed.

● Cost – The new system is certain to require fewer direct operatives to staff the warehouse, but will need extra engineering and maintenance support. Overall, however, lower labour costs are likely.

Does the technology give an acceptable financial return? Assessing the financial value of investing in process technology is in itself a specialized sub- ject. And while it is not the purpose of this book to delve into the details of financial analysis, it is important to highlight one important issue that is central to financial evaluation: while the benefits of investing in new technology can be spread over many years into the future, the costs associated with investing in the technology usually occur up front. So we have to consider the time value of money. Simply, this means that receiving €1,000 now is better than receiving €1,000 in a year’s time. Receiving €1,000 now enables us to invest the money so that it will be worth more than the €1,000 we receive in a year’s time. Alternatively, reversing the logic, we can ask ourselves how much would have to be invested now to receive €1,000 in one year’s time? This amount (lower than €1,000) is called the net present value of receiving €1,000 in one year’s time.

For example, suppose current interest rates are 10 per cent per annum; then the amount we would have to invest to receive €1,000 in one year’s time is:

€1,000 *

1 = €909.10

(1.10)

So the present value of €1,000 in one year’s time, discounted for the fact that we do not have it immediately , is €909.10. In two years’ time, the amount we would have to invest to receive €1,000 is:

€1,000 *

1 *

1 = €1,000 *

1 = €826.50

(1.10) (1.10) (1.10)2

The rate of interest assumed (10 per cent in our case) is known as the discount rate. More generally, the present value of € x in n years’ time, at a discount rate of r per cent, is:

€

x (1 + r/100)n

Worked example

The warehouse which we have been using as an example has been subjected to a costing and cost savings exercise. The capital cost of purchasing and installing the new technology can be spread over three years, and from the first year of its effective operation, overall operations cost savings will be made. Combining the cash that the company will have to spend and the savings that it will make, the cash flow year by year is shown in Table 8.1 .

▼

M08_SLAC8678_08_SE_C08.indd 262 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 263

However, these cash flows have to be discounted in order to assess their ‘present value’. Here the company is using a discount rate of 10 per cent. This is also shown in Table 8.1 The effective life of this technology is assumed to be six years:

Total cash fl ow (sum of all the cash fl ows) = €1.38 million

However:

Net present value (NPV) = €816,500

This is considered to be acceptable by the company. Calculating discount rates, although perfectly possible, can be cumbersome. As an alter-

native, tables are usually used such as the one in Table 8.2 . So now the net present value is:

P = DF * FV

where: DF = the discount factor from Table 8.2 FV = future value

To use the table, find the vertical column and locate the appropriate discount rate (as a percentage). Then find the horizontal row corresponding to the number of years it will take to receive the payment. Where the column and the row intersect is the present value of €1. You can multiply this value by the expected future value, in order to find its present value.

▼ Years 3.0% 4.0% 5.0% 6.0% 7.0% 8.0% 9.0% 10.0%

1 €0.970 €0.962 €0.952 €0.943 €0.935 €0.926 €0.918 €0.909

2 €0.942 €0.925 €0.907 €0.890 €0.873 €0.857 €0.842 €0.827

3 €0.915 €0.889 €0.864 €0.840 €0.816 €0.794 €0.772 €0.751

4 €0.888 €0.855 €0.823 €0.792 €0.763 €0.735 €0.708 €0.683

5 €0.862 €0.822 €0.784 €0.747 €0.713 €0.681 €0.650 €0.621

6 €0.837 €0.790 €0.746 €0.705 €0.666 €0.630 €0.596 €0.565

7 €0.813 €0.760 €0.711 €0.665 €0.623 €0.584 €0.547 €0.513

8 €0.789 €0.731 €0.677 €0.627 €0.582 €0.540 €0.502 €0.467

9 €0.766 €0.703 €0.645 €0.592 €0.544 €0.500 €0.460 €0.424

10 €0.744 €0.676 €0.614 €0.558 €0.508 €0.463 €0.422 €0.386

11 €0.722 €0.650 €0.585 €0.527 €0.475 €0.429 €0.388 €0.351

12 €0.701 €0.626 €0.557 €0.497 €0.444 €0.397 €0.356 €0.319

13 €0.681 €0.601 €0.530 €0.469 €0.415 €0.368 €0.326 €0.290

14 €0.661 €0.578 €0.505 €0.442 €0.388 €0.341 €0.299 €0.263

Table 8.2 Present value of €1 to be paid in future

Year 0 1 2 3 4 5 6 7

Cash fl ow (€000s) −300 30 50 400 400 400 400 0

Present value (discounted at 10%)

−300 27.27 41.3 300.53 273.21 248.37 225.79 0

Table 8.1 Cash flows for the warehouse process technology

M08_SLAC8678_08_SE_C08.indd 263 6/2/16 1:27 PM

264 PART TWO DESIGNING THE OPERATION

Worked example

A healthcare clinic is considering purchasing a new analysis system. The net cash flows from the new analysis system are as follows:

Year 1: −€10,000 (outflow of cash) Year 2: €3,000 Year 3: €3,500 Year 4: €3,500 Year 5: €3,000

Assuming that the real discount rate for the clinic is 9 per cent, using the net present value table ( Table 8.2 ), demonstrate whether the new system would at least cover its costs. Table 8.3 shows the calculations. It shows that, because the net present value of the cash flow is positive, purchasing the new system would cover its costs, and will be ( just) profit- able for the clinic.

HOW ARE PROCESS TECHNOLOGIES IMPLEMENTED?

Implementing process technology means organizing all the activities involved in making the technology work as intended. No matter how potentially beneficial and sophisticated the technology, it remains only a prospective benefit until it has been implemented successfully. So implementation is an important part of process technology management. Yet it is not always straightforward to make general points about the implementation process because it is very context dependent. That is, the way one implements any technology will very much depend on its specific nature, the changes implied by the technology and the organizational

Years 3.0% 4.0% 5.0% 6.0% 7.0% 8.0% 9.0% 10.0%

15 €0.642 €0.555 €0.481 €0.417 €0.362 €0.315 €0.275 €0.239

16 €0.623 €0.534 €0.458 €0.394 €0.339 €0.292 €0.252 €0.218

17 €0.605 €0.513 €0.436 €0.371 €0.317 €0.270 €0.231 €0.198

18 €0.587 €0.494 €0.416 €0.350 €0.296 €0.250 €0.212 €0.180

19 €0.570 €0.475 €0.396 €0.331 €0.277 €0.232 €0.195 €0.164

20 €0.554 €0.456 €0.377 €0.312 €0.258 €0.215 €0.179 €0.149

Year Cash fl ow Table factor Present value

1 (€10,000) * 1.000 = (€10,000.00)

2 €3,000 * 0.917 = €2,752.29

3 €3,500 * 0.842 = €2,945.88

4 €3,500 * 0.772 = €2,702.64

5 €3,000 * 0.708 = €2,125.28

Net present value = €526.09

Table 8.3 Present value calculations for the clinic.

M08_SLAC8678_08_SE_C08.indd 264 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 265

conditions that apply during its implementation. In the remainder of this chapter we look at four particularly important issues that affect technology implementation: the way technology is planned over the long term, the idea of resource and process ‘distance’, the need to consider customer acceptability, and the idea that if anything can go wrong, it will.

Technology planning in the long-term – technology roadmapping However operations managers are involved with the development of process technologies, it is likely to be in consultation and collaboration with other parts of the firm. It is also likely to be in the context of some kind of formal planning process such as technology roadmap- ping. A technology roadmap (TRM) is an approach that provides a structure that attempts to assure the alignment of developments (and investments) in technology, possible future market needs, and the new development of associated operations capabilities. Motorola orig- inally developed the approach in the 1970s so that it could support the development of its products and its supporting technologies. Bob Galvin, then Motorola’s CEO, defined a TRM as: ‘an extended look at the future of a chosen field of inquiry composed from the collective knowl- edge and imagination of the brightest drivers of change in that field’. A TRM is essentially a pro- cess that supports technology development by facilitating collaboration between the various activities that contribute to technology strategy. It allows technology managers to define their firm’s technological evolution in advance by planning the timing and relationships between the various elements that are involved in technology planning. For example, these ‘elements’ could include the business goals of the company, market developments or specific events, the component products and services that constitute related offerings, product/service and process technologies, the underlying capabilities that these technologies represent, and so on. Figure 8.6 shows the generic form of technology roadmaps, while Figure 8.7 shows an example of a TRM for the development of products/services, technologies and processes for a facilities management service.

The benefits of TRMs are mainly associated with the way they bring together the significant stakeholders involved in technology strategy and various (and often differing) perspectives

Time

Elements of technology

planning

Market developments

Products/services

Technologies

Capabilities

Business goals

Projects

Process developments

Knowledge enablers

Intellectual resources

Decision points

External events, e.g. competitor activity

Etc.

For example Timing of, and relationship between, the elements of

technology planning

Figure 8.6 The generic form of a technology roadmap (TRM)

M08_SLAC8678_08_SE_C08.indd 265 6/2/16 1:27 PM

266 PART TWO DESIGNING THE OPERATION

Year 1 Year 2 Year 3 Year 4 etc.

Meet budget limits Integrate with divisions

Transfer knowledge to divisions

Market launch prep

Control software Interface integration

Website prototype test

Client interface portal

Establish client response centre

Develop service teams

ERP integration

Time

Digital adaptive agents

Resource planning algorithms

CRM implementation

Elements of technology planning

Strategic business goals

Product/ service development

Development of underlying technologies

Process developments

Develop resource planning model

Figure 8.7 Simplified example of a TRM for the development of products/services, technologies, and processes for a facilities management service

they have. The approach forms a basis for communication, and possibly consensus. After all, it does tackle some fundamental questions that concern any technology strategy. Why do we need to develop our technology? Where do we want to go with our technological capabilities? How far away are we from that objective? How can we get to where we want to be? In what order should we do things? By when should development goals be reached? Yet TRMs do not offer any solutions to any firm’s technological strategic options; in fact they need not offer options or alternative technology trajectories. They are essentially a narrative description of how a set of interrelated developments should (rather than will) progress. Because of this they have been criticized as encouraging over-optimistic projections of the future. Nevertheless, they do provide, at the very least, a plan against which technology strategy can be assessed.

Resource and process ‘distance’ The degree of difficulty in the implementation of process technology will depend on the degree of novelty of the new technology resources and the changes required in the operation’s processes. The less that the new technology resources are understood (influenced perhaps by the degree of innovation), the greater their ‘distance’ from the current technology resource base of the operation. Similarly, the extent to which an implementation requires an operation to modify its existing processes, the greater the ‘process distance’. The greater the resource and process distance, the more difficult any implementation is likely to be. This is because

such distance makes it difficult to adopt a systematic approach to ana- lysing change and learning from mistakes. Those implementations which involve relatively little process or resource ‘distance’ provide an ideal opportunity for organizational learning. As in any classic scien- tific experiment, the more variables that are held constant, the more confidence you have in determining cause and effect. Conversely, in an implementation where the resource and process ‘distance’ mean

✽ ✽ ✽ Operations principle Operations principle Operations principle Operations principle Operations principle Operations principle

M08_SLAC8678_08_SE_C08.indd 266 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 267

that nearly everything is ‘up for grabs’, it becomes difficult to know what has worked and what has not. More importantly, it becomes difficult to know why something has or has not worked. 9 This idea is illustrated in Figure 8.8 .

Figure 8.8 Learning potential depends on both technological resource and process ‘distance’

OPERATIONS IN PRACTICE

Not enough people choose ‘Choose and Book’ It was a technology project that was 10 years in the mak- ing. The ‘Choose and Book’ system should have trans- formed the way in which patients and their ‘General Practitioner ’ (GP) physicians could select an outpatient hospital appointment at a convenient date and time in the UK’s National Health Service (NHS). The aim was to speed up the process and cut out the need for costly paperwork. Yet in 2014 it was quietly dropped despite costing £356m during the 10 years that it had been struggling to establish itself. It was taken as another example of the difficulties of introducing new technol- ogy systems into such a huge and complex organiza- tion. An investigation by the UK’s House of Commons’ Public Accounts Committee was told by NHS staff that, although some GPs liked the ‘Choose and Book’ system, many did not. Moreover, not all outpatient appointment slots were available on the system, which limited its use- fulness. Many patients and doctors found ‘Choose and

Book’ complicated and time consuming. One GP, Sarah Wollaston, said, ‘ the system suits patients who were good with technology but not those who were less so. Doctors often did not have time to log on to it during appoint- ments with their patients .’ A Member of Parliament

Two technology failures 10

▼

So u

rc e:

S h

u tt

er st

o ck

.c o

m : P

at h

d o

c

M08_SLAC8678_08_SE_C08.indd 267 6/2/16 1:27 PM

268 PART TWO DESIGNING THE OPERATION

said: ‘ It’s another NHS cock up. A system designed for use by GPs but only used by half of them…has been quietly dropped, so quietly that even most of the NHS seems una- ware. In the middle of all of this are patients. Choose and Book was supposed to speed things up but the evidence we heard in committee showed this was not so in most cases .’ Despite the failure of ‘Choose and Book’ (or only partial success) the government department that over- sees the NHS decided to replace it with a potentially even more expensive e-referral scheme, saying that the new e-referral system would use different technology and have additional features as well as being available on mobile apps. A spokesperson said, ‘ we are aiming to have 100% electronic referrals within the next five years – sooner than that if we can make it. That will cut out a lot of these errors. ’ It was also reported that the idea of making it compulsory for GPs to use the replacement system when it comes on-stream, with an inbuilt incen- tive and penalty scheme for doctors and hospitals, was being considered.

The BBC’s Digital Media Initiative The BBC is one of the best-known broadcasters in the world, with an unrivalled reputation for the quality of some of its programmes. Sadly, its reputation for intro- ducing new technology is less exemplary. Among its more spectacular failures was its Digital Media Initiative (DMI). The DMI was an endeavour by the BBC to dis- pense with videotapes and create a kind of ‘internal YouTube’ of archive content that staff could access, upload, edit and then air from their computers. When the project was originally envisaged, creating a sin- gle TV programme could involve 70 individual video- handling processes. DMI was meant to halve that. The project cost almost £100 million and lasted five years before it was scrapped. The flaws in the technology were exposed during the BBC’s coverage of the state funeral of Margaret Thatcher, a well-known ex-Prime Minister. The DMI was supposed to create a production system linked to the BBC’s huge broadcasting archive, but instead of

streamlining access to old video footage, video edi- tors were unable to access archive footage to use in news reports from their computers in Central London. Instead they had to transport videotapes there using taxis and the underground network from the archive storage facility in north-west London. Admitting that to continue with the project would be ‘ throwing good money after bad ’, the BBC suspended its chief technol- ogy officer. One BBC manager called the DMI project ‘ the axis of awful ’, while another said, ‘ The scale of the project was just too big, and it got out of hand .’ Anthony Fry, a member of the BBC’s governing body, said that the project had ‘ generated little or no assets for the corpora- tion. This is because much of the software and hardware which has been developed could only be used by the BBC if the project were completed, which, due to technologi- cal difficulties and changes to business needs … [was not possible]. Tony Hall, the BBC’s Director General, said that off-the-shelf tools ‘ that simply didn’t exist five years ago ’ had now become available and they could do the same job as some elements of the DMI. Professors Elizabeth Daniel of the Open University Business School and John Ward of Cranfield School of Management, commenting on the BBC DMI case, said, ‘ it is not the biggest or the worst IT project failure in the public or private sectors and, without organizations’ implementing measures to guard against them, it will almost certainly not be the last ’. While at first glance, they say, it seems the BBC’s Digital Media Initiative project suffered from the challenges encoun- tered in many other large IT projects, there are some aspects of the BBC operation and culture that may have exacerbated them. The organization appears to have reacted slowly to concerns raised at senior level, there was an inability to identify that things were going wrong and then to act impartially. The failure of the DMI was regarded as an IT failure, not of the BBC, and, most wor- rying, there was a culture which apparently did not allow staff involved to be given a voice, so, unable to feed their concerns about projects into review processes, they were instead reduced to privately voicing them.

Customer acceptability When an operation’s customers interact with its process technology it is essential to consider the customer interaction when evaluating it. If customers are to have direct contact with tech- nology, they must have some idea of how to operate it. Where customers have an active inter- action with technology, the limitations of their understanding of the technology can be the main constraint on its use. For example, even some domestic technology such as smart TVs cannot be used to their full potential by most owners. Other customer-driven technologies can face the same problem, with the important addition that if customers cannot use technologies such as Internet banking, there are serious commercial consequences for a bank’s customer service. Staff in manufacturing operations may require several years of training before they

M08_SLAC8678_08_SE_C08.indd 268 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 269

are given control of the technology they operate. Service operations may not have the same opportunity for customer training. Walley and Amin 11 suggest that the ability of the opera- tion to train its customers in the use of its technology depends on three factors: complexity, repetition, and the variety of tasks performed by the customer. If services are complex, higher levels of ‘training’ may be needed; for example, the technologies in theme parks and fast food outlets rely on customers copying the behaviour of others. Frequency of use is important because the payback for the ‘investment’ in training will be greater if the customer uses the technology frequently. Also, customers may, over time, forget how to use the technology, but regular repetition will reinforce the training. Finally, training will be easier if the customer is presented with a low variety of tasks. For example, vending machines tend to concentrate on one category of product, so that the sequence of tasks required to operate the technology remains consistent.

In other cases the technology may not be trusted by customers because it is technology and not a person. Sometimes we prefer to put ourselves in the care of a person, even if their performance is inferior to a technology. For example, the use of robot technologies in surgery has distinct advantages over conventional surgery, but in spite of the fact that the surgeon is in control, it is viewed with suspicion by some patients and physicians. When robot surgeons operate without any direct human control, rather than simply mirroring the movement of human surgeons, resistance is likely to be even greater. Similarly the idea of pilotless aircraft is difficult to ‘sell’ to customers; see the ‘Who’s in the cockpit?’ case.

OPERATIONS IN PRACTICE

Modern aircraft fly on automatic pilot for most of their time, certainly more than most passengers realize. ‘ Most people are blissfully unaware that when an aircraft lands in mist or fog, it is a computer that is landing it ’, says Paul Jackson of Jane’s All The World’s Aircraft . ‘ It is the only sen- sible thing to do ’, agrees Ken Higgins of Boeing. ‘ When auto pilots can do something better than a human pilot, we obviously use auto pilots .’ Generally this means using auto pilots to do two jobs. First, they can take control of the aircraft during the long and (for the pilot) monotonous part of the flight between take-off and landing. Automatic pilots are not prone to the tedium or weariness which can affect humans and which can cause pilot error. The second job is to make landings, especially when visibility is poor because of fog or light conditions. The auto pilot communicates with automatic equipment on the ground which allows the aircraft to be landed, if necessary, under conditions of zero visibility. In fact, automatic landings when visibility is poor are safer than when the pilot is in control. Even in the unlikely event of one of the aircraft’s two engines failing, an auto pilot can land it safely. This means that, on some flights, the auto pilot is switched on within seconds of the aircraft wheels leaving the ground and then remains in charge throughout the flight and the landing. One of the few reasons not to use the auto pilot is if the pilot is training or needs to log up the required number of landings to keep licensed.

As yet , commercial flights do not take off auto- matically, mainly because it would require airports and airlines to invest in extra guidance equipment which would be expensive to develop and install. Also take-off is technically more complex than landing. More things could go wrong and some situations (for example, an engine failure during take-off ) require split-second decision making from the pilot. Industry analysts agree that it would be technically feasible to develop automatic take-off technology that met required safety standards, but it could be prohibitively expensive.

Who’s in the cockpit? 12

▼

So u

rc e:

T h

e K

o b

al C

o lle

ct io

n : P

ar am

o u

n t

P ic

tu re

s

M08_SLAC8678_08_SE_C08.indd 269 6/2/16 1:27 PM

270 PART TWO DESIGNING THE OPERATION

Anticipating implementation problems The implementation of any process technology will need to account for the ‘adjustment’ issues that almost always occur when making any organizational change. By adjustment issues we mean the losses that could be incurred before the improvement is functioning as intended. But estimating the nature and extent of any implementation issues is notori- ously difficult. This is particularly true because, more often than not, Murphy’s law seems to prevail. This law is usually stated as: ‘if anything can go wrong, it will’. This effect has been identified empirically in a range of operations, especially when new types of process technology are involved. Specifically discussing technology-related change (although the ideas apply to almost any implementation), Bruce Chew of the Massachusetts Institute of Technology 13 argues that adjustment ‘costs’ stem from unforeseen mismatches between the new technology’s capabilities and needs and the existing operation. New technology rarely behaves as planned, and as changes are made their impact ripples throughout the organization. Figure 8.9 is an example of what Chew calls a Murphy curve. It shows a typ- ical pattern of performance reduction (in this case, quality) as a new process technology

Yet some in the airline industry believe that tech- nology could be developed to the point where com- mercial flights can do without a pilot on the aircraft entirely. This is not as far-fetched as it seems. In April 2001 the Northrop Grumman Global Hawk, an ‘unmanned aerial vehicle’ (UAV), completed the first entirely unmanned flight of the Pacific when it took off from California and landed nearly 24 hours later in South Australia . The Global Hawk made the jour- ney without any human intervention whatsoever. ‘ We made a historic flight with two clicks of the mouse ’, said Bob Mitchell of Northrop Grumman. The first mouse click told the aircraft to take off; the second, made after landing, told it to switch of its engine. UAVs are used for military reconnaissance purposes but enthusiasts point out that most aircraft breakthroughs, such as the

jet engine and radar, were developed for military use before they found civilian applications. However, even the enthusiasts admit that there are some significant problems to overcome before pilotless aircraft could become commonplace. The entire commercial flight infrastructure from air traffic control through to air- port control would need to be restructured, a wholly automatic pilotless aircraft would have to be shown to be safe, and, perhaps most important, passengers would have to be persuaded to fly in them. If all these objections could be overcome, the rewards are sub- stantial. Airlines’ largest single cost is the wages of its staff (far more than fuel costs or maintenance costs etc.) and, of all staff, pilots are by far the most costly. Automated flights would cut costs significantly, but no one is taking bets on its happening soon!

Figure 8.9 The reduction in performance during and after the implementation of a new process reflects ‘adjustments costs’

M08_SLAC8678_08_SE_C08.indd 270 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 271

● Process technologies are the machines, equipment or devices that help operations to cre- ate or deliver products and services. Indirect process technology helps to facilitate the direct creation of products and services.

❯ What is process technology?

SUMMARY ANSWERS TO KEY QUESTIONS

● Operations managers do not need to know the technical details of all technologies, but they do need to know the answers to four key questions: What does it do? How does it do it? What advantages does it give? What constraints does it impose?

● Process technologies can also be classifi ed according to the transformed resources that they process, namely material-processing technologies, information-processing technolo- gies and customer-processing technologies. In addition some technologies process more than one type of resource; they are called integrating technologies.

● An important element in understanding process technologies is to understand the implica- tions they hold for the operations where they will be used.

❯ What do operations managers need to know about process technology?

● All technologies should be appropriate for the activities that they have to undertake. In practice this means making sure that the degree of automation of the technology, the scale or scalability of the technology, and the degree of coupling or connectivity of the technol- ogy fi t the volume and variety characteristics of the operation.

● All technologies should be evaluated by assessing the impact that the process technology will have on the operation’s performance objectives (quality, speed, dependability, fl exibil- ity and cost).

● All technologies should be evaluated fi nancially. This usually involves the use of some of the more common evaluation approaches, such as net present value (NPV).

❯ How are process technologies evaluated?

is introduced. It is recognized that implementation may take some time; therefore allow- ances are made for the length and cost of a ‘ramp-up’ period. However, as the operation prepares for the implementation, the distraction causes performance actually to deterio- rate. Even after the start of the implementation this downward trend continues and it is only weeks, indeed maybe months, later that the old performance level is reached. The area of the dip indicates the magnitude of the adjustment costs, and therefore the level of vulnerability faced by the operation.

● Implementing process technology means organizing all the activities involved in making the technology work as intended.

● A technology roadmap (TRM) is an approach that provides a structure that attempts to assure the alignment of developments (and investments) in technology, possible future market needs, and the new development of associated operations capabilities.

❯ How are process technologies implemented?

M08_SLAC8678_08_SE_C08.indd 271 6/2/16 1:27 PM

272 PART TWO DESIGNING THE OPERATION

Dr Rhodes was losing his temper. ‘ It should be a simple enough decision. There are only two alternatives. You are only being asked to choose a machine! ’

The Management Committee looked abashed. Rochem Ltd was one of the largest independent companies sup- plying the food-processing industry. Its initial success had come with a food preservative used mainly for meat-based products and marketed under the name of ‘Lerentyl’. Other products were subsequently developed in the food colour- ing and food container coating fields, so that now Lerentyl accounted for only 25 per cent of total company sales, which were now slightly over £10 million.

The decision The problem over which there was such controversy related to the replacement of one of the process units used to manufacture Lerentyl. Only two such units were used; both were ‘Chemling ’ machines. It was the older of the two Chemling units which was giving trouble. High breakdown figures, with erratic quality levels, meant that output-level requirements were only just being reached. The problem was: should the company replace the ageing

Chemling with a new Chemling, or should it buy the only other plant on the market capable of the required process, the ‘AFU’ unit? The Chief Chemist’s staff had drawn up a comparison of the two units, shown in Table 8.4 .

The body considering the problem was the newly formed Management Committee. The committee consisted of the

CASE STUDY Rochem Ltd

● The resource and process ‘distance’ implied by the technology implementation will indicate the degree of diffi culty.

● Customer acceptability may be a barrier to implementation in customer-processing technologies.

● It is necessary to allow for the adjustment costs of implementation.

Chemling AFU

Capital cost £590,000 £880,000

Processing costs Fixed: £15,000/month Fixed: £40,000/month

Variable: £750/kg Variable: £600/kg

Design capacity 105 kg/month 140 kg/month

98 ± 0.7% purity 99.5 ± 0.2% purity

Quality Manual testing Automatic testing

Maintenance Adequate but needs servicing Not known – probably good

After-sales services Very good Not known – unlikely to be good

Delivery Three months Immediate

Table 8.4 A comparison of the two alternative machines So

u rc

e: P

re ss

A ss

o ci

at io

n Im

ag es

: A P

/ E

ck eh

ar d

S ch

u lz

M08_SLAC8678_08_SE_C08.indd 272 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 273

four senior managers in the firm: the Chief Chemist and the Marketing Manager, who had been with the firm since its beginning, together with the Production Manager and the Accountant, both of whom had joined the company only six months earlier.

What follows is a condensed version of the information presented by each manager to the committee, together with their attitudes to the decision.

The Marketing Manager The current market for this type of preservative had reached a size of some £5 million, of which Rochem Ltd supplied approximately 48 per cent. There had, of late, been significant changes in the market – in particular, many of the users of preservatives were now able to buy products similar to Lerentyl. The result had been the evo- lution of a much more price-sensitive market than had previously been the case. Further market projections were somewhat uncertain. It was clear that the total mar- ket would not shrink (in volume terms) and best estimates suggested a market of perhaps £6 million within the next three or four years (at current prices). However, there were some people in the industry who believed that the present market only represented the tip of the iceberg.

Although the food preservative market had advanced by a series of technical innovations, ‘real’ changes in the basic product were now few and far between. Lerentyl was sold in either solid powder or liquid form, depending on the par- ticular needs of the customer. Prices tended to be related to the weight of chemical used, however. Thus, for exam- ple, the current average market price was approximately £1,050 per kg. There were, of course, wide variations depending on order size etc.

‘At the moment I am mainly interested in getting the right quantity and quality of Lerentyl each month and although Production has never let me down yet, I’m worried that unless we get a reliable new unit quickly, it soon will. The AFU machine could be on line in a few weeks, giving better quality too. Furthermore, if demand does increase (but I’m not saying it will), the AFU will give us the extra capacity. I will admit that we are not trying to increase our share of the preservative market as yet. We see our priority as establishing our other products first. When that’s achieved, we will go back to con- centrating on the preservative side of things.’

The Chief Chemist The Chief Chemist was an old friend of Dr Rhodes and together they had been largely responsible for every prod- uct innovation. At the moment, the major part of his budget was devoted to modifying basic Lerentyl so that it could be used for more acidic food products such as fruit. This was not proving easy and as yet nothing had come of the research, although the Chief Chemist remained optimistic.

‘If we succeed in modifying Lerentyl the market opportu- nities will be doubled overnight and we will need the extra

capacity. I know we would be taking a risk by going for the AFU machine, but our company has grown by gambling on our research findings, and we must continue to show faith. Also the AFU technology is the way all similar technologies will be in the future. We have to start learning how to exploit it sooner or later.’

The Production Manager The Lerentyl Department was virtually self-contained as a production unit. In fact, it was physically separate, located in a building a few yards detached from the rest of the plant . Production requirements for Lerentyl were cur- rently at a steady rate of 190 kg per month. The six techni- cians who staffed the machines were the only technicians in Rochem who did all their own minor repairs and full quality control. The reason for this was largely historical since, when the firm started, the product was experimen- tal and qualified technicians were needed to operate the plant. Four of the six had been with the firm almost from its beginning.

‘It’s all right for Dave and Eric [Marketing Manager and Chief Chemist] to talk about a big expansion of Lerentyl sales; they don’t have to cope with all the problems if it doesn’t hap- pen. The fixed costs of the AFU unit are nearly three times those of the Chemling. Just think what that will do to my budget at low volumes of output. As I understand it, there is absolutely no evidence to show a large upswing in Lerentyl. No, the whole idea [of the AFU plant] is just too risky. Not only is there the risk. I don’t think it is generally understood what the consequences of the AFU would mean. We would need twice the variety of spares for a start. But what really worries me is the staff ’s reaction. As fully qualified technicians they regard themselves as the elite of the firm; so they should, they are paid practically the same as I am! If we get the AFU plant, all their most interesting work, like the testing and the maintenance, will disappear or be greatly reduced. They will finish up as highly paid process workers.’

The Accountant The company had financed nearly all its recent capital investment from its own retained profits, but would be taking out short-term loans the following year for the first time for several years.

‘At the moment, I don’t think it wise to invest extra cap- ital we can’t afford in an attempt to give us extra capac- ity we don’t need. This year will be an expensive one for the company. We are already committed to considerably increased expenditure on promotion of our other prod- ucts and capital investment in other parts of the firm, and Dr Rhodes is not in favour of excessive funding from outside the firm. I accept that there might eventually be an upsurge in Lerentyl demand but, if it does come, it probably won’t be this year and it will be far bigger than the AFU can cope with anyway, so we might as well have three Chemling plants at that time.’

M08_SLAC8678_08_SE_C08.indd 273 6/2/16 1:27 PM

274 PART TWO DESIGNING THE OPERATION

QUESTIONS 1 How do the two alternative process technologies

(Chemling and AFU) differ in terms of their scale and automation? What are the implications of this for Rochem?

2 Remind yourself of the distinction between feasibility, acceptability and vulnerability discussed in Chapter 4 . Evaluate both technologies using these criteria.

3 What would you recommend the company should do?

PROBLEMS AND APPLICATIONS

1 In the early part of this chapter, three technologies are described: 3D printing, the Internet of Things, and Telemedicine. Try to describe the technologies by answering the ‘four key ques- tions’ that are also described.

2 A new machine requires an investment of €500,000 and will generate profits of €100,000 for 10 years. Will the investment have a positive net present value assuming that a realistic inter- est is 6 per cent?

3 A local government housing office is considering investing in a new computer system for managing the maintenance of its properties. The system is forecast to generate savings of around £100,000 per year and will cost £400,000. It is expected to have a life of seven years. The local authority expects its departments to use a discount rate of 0.3 to calculate the finan- cial return on its investments. Is this investment financially worthwhile?

4 In the problem above, the local government’s finance officers have realized that their dis- count rate has been historically too low. They now believe that the discount rate should be doubled. Is the investment in the new computer system still worthwhile?

5 A new optical reader for scanning documents is being considered by a retail bank. The new system has a fixed cost of €30,000 per year and a variable cost of €2.5 per batch. The cost of the new scanner is €100,000. The bank charges €10 per batch for scanning documents and it believes that the demand for its scanning services will be 2,000 batches in year 1, 5,000 batches in year 2, 10,000 batches in year 3, and then 12,000 batches per year from year 4 onwards. If the realistic discount rate for the bank is 6 per cent, calculate the net present value of the investment over a five-year period.

SELECTED FURTHER READING

Arthur, W.B. (2010) The Nature of Technology: What It Is and How It Evolves, Penguin, Harmondsworth.

Popular science in a way, but very interesting on how technologies evolve.

Brain, M. (2001) How Stuff Works , Wiley, New York.

Exactly what it says. A lot of the ‘stuff ’ is product technology, but the book also explains many pro- cess technologies in a clear and concise manner without sacrificing relevant detail.

Brynjolfsson, E. and McAfee, A. (2014) The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies , W. W. Norton, New York.

This is one of the most influential recent books on how technology will change our lives.

M08_SLAC8678_08_SE_C08.indd 274 6/2/16 1:27 PM

CHAPTER 8 PROCESS TECHNOLOGY 275

Carr, N.G. (2000) Hypermediation: ‘Commerce and Clickstream’, Harvard Business Review, January–February.

Written at the height of the Internet boom, it gives a flavour of how Internet technologies were seen.

Chew, W.B., Leonard-Barton, D. and Bohn, R.E. (1991) Beating Murphy ’s Law, Sloan Management Review, vol. 5, Spring.

One of the few articles that treats the issue of why everything seems to go wrong when any new technology is introduced. Insightful.

Cobham, D. and Curtis, G. (2004) Business Information Systems: Analysis, Design and Practice, Financial Times Prentice Hall, Harlow.

A good solid text on the subject.

Evans, P. and Wurster, T. (1999) Blown to Bits: How the new economics of information transforms strategy, Harvard Business School Press, Boston, MA.

Interesting exposition of how Internet-based technologies can change the rules of the game in business.

M08_SLAC8678_08_SE_C08.indd 275 6/2/16 1:27 PM

Operations management

Direct

Design Develop

Deliver

Design

Layout and flow

Process design

Process technology

People in operations

Topic covered in this chapter

figure 9.1 this chapter examines people in operations

❯ why are people so important in operations management?

❯ how do operations managers contribute to human resource strategy?

❯ how can the operations function be organized?

❯ how do we go about designing jobs?

❯ how are work times allocated?

key questions introduction operations management is often presented as a subject the main focus of which is on technology, systems, procedures and facilities – in other words, the non-human parts of the organization. this is not true of course. on the contrary, the manner in which an organization’s human resources are managed has a profound impact on the effectiveness of its operations function. in this chapter we look especially at the elements of human resource management which are traditionally seen as being directly within the sphere of operations management. these are how operations managers contribute to human resource strategy, organization design, job design, and the allocation of ‘work times’ to operations activities. the more detailed (and traditional) aspects of these last two elements are discussed further in the supplement on work study at the end of this chapter. Figure 9.1 shows how this chapter fits into the overall model of operations activities.

people in operations 9

M09_SLAC8678_08_SE_C09.indd 276 06/02/16 5:00 PM

CHAPTER 9 PEOPLE IN OPERATIONS 277

WHY ARE PEOPLE SO IMPORTANT IN OPERATIONS MANAGEMENT?

To say that an organization’s human resources are its greatest asset is something of a cliché. Yet it is worth reminding ourselves of the importance of the abilities, attitudes and culture of the people who make up the operations function. It is, after all, where most ‘human resources’ are to be found. It follows that it is operations managers who are most involved in the leadership, development and organization of human resources. In this chapter we examine some of the issues that most directly affect, or are affected by, operations management; these are illus- trated in Figure 9.2 . But the influence of operations management on the organization’s staff is not limited to the topics covered in this chap- ter. Almost everything discussed in this book has a ‘people’ dimension. Yet, in some chapters, the human perspective is particularly impor- tant. In addition to this chapter, Chapters 16 and 17 , for example, are concerned largely with how the contribution of the operation’s staff can be harnessed. In essence the issues covered in this chapter define how people go about their working lives. It positions their expecta- tions of what is required of them, and it influences their perceptions of how they contribute to the organization. It defines their activities in relation to their work colleagues and it channels the flows of communication between different parts of the operation. But, of most importance, it helps to develop the culture of the organization – its shared values, beliefs and assumptions.

Operations culture What do we mean by culture in the context of the operations function? There is a wealth of academic or popular literature that treats the concept of organizational culture, but no single authoritative definition has emerged. Nevertheless most of us know roughly what is meant by ‘culture’ in an organization. It is what it feels like to be part of it – what is assumed about how things get done rather than is necessarily formally articulated. It is, in the words of one well- known writer on the subject, ‘ the way we do things around here ’ or ‘ the organisation’s climate ’. 1 But the idea of ‘organizational’ culture can also apply to a single function like the operations function. In fact there is considerable interest among researchers and practitioners in over- coming the cultural differences between different functions that can sometimes lead to what has been called ‘cultural fragmentation’. Even though there may be elements of an organiza- tion’s culture that are shared across all parts of the enterprise, different functions are very likely to have their own subcultures.

People in operations

Understand how operations can be

organized

Design the working environment

Contribute to human resource

strategy

Allocate work times

Design individuals’ and groups’ jobs

Figure 9.2 People in operations

✽ ✽ ✽ Operations principle Operations principle Operations principle

M09_SLAC8678_08_SE_C09.indd 277 6/2/16 1:32 PM

278 PART TWO DESIGNING THE OPERATION

OPERATIONS IN PRACTICE

Most famous for its high-performance fabrics such as Gore-Tex, W.L. Gore also has an enviable reputation as being one of the best companies to work for wherever it operates. In a recent ‘Best Companies to work for’ list its associates (the company does not use the term ‘employ- ees’) gave it the very top marks for ‘feeling you can make a difference’. More than half of its staff have been with the firm for at least a decade, a consequence of its phi- losophy (‘to make money and have fun’) and its unique organizational culture and job design practices. Few in the company have any formal job titles, or job descrip- tions. There are no managers, only leaders and associates, people are paid ‘according to their contribution’ and staff help to determine each other's pay – ideas which seem revolutionary yet are based on the company's founding principles from over 50 years ago. Started by Bill and Vieve Gore in the basement of their home in Delaware, it has now become a global business with facilities in more than 45 locations around the world. Its skilled staff develop, manufacture and sell a range of innovative prod- ucts, virtually all of which are based on just one material (expanded polytetrafluoroethylene) which was discov- ered by Bob Gore (the founders' son) in 1969. It now has approximately 8,000 associates in its four main divisions (textiles, electronic, medical and industrial products) and annual revenues of over $2 billion.

Gore's approach to how it works with its staff is at the heart of the company's success. On almost every level Gore is different to other global companies. Associates are hired for general work areas rather than specific jobs, and with the guidance of their ‘spon- sors’ (not bosses), and as they develop experience, they commit to projects that match their skills. Teams organize around opportunities as they arise, with associates committing to the projects that they have chosen to work on, rather than having tasks delegated to them. Project teams are small, focused, multi-dis- ciplined, and foster strong relationships between team members. Personal initiative is encouraged, as is ‘hands-on’ innovation, which involves those clos- est to a project in its decision making. There are, says Gore, no traditional organizational charts, no chains of command, no predetermined channels of com- munication. Instead, team members communicate directly with each other and are accountable to the other members of their team. Groups are led by who- ever is the most appropriate person at each stage of a project. Leaders are not appointed by senior man- agement ; they ‘emerge’ naturally by demonstrating special knowledge, skill or experience that advances

a business objective. Everyone's performance is assessed using a peer-level rating system. Even the group's CEO (one of the few people with a title), Terri Kelly, ‘emerged’ in this way. When the previous CEO retired, no shortlist of preferred candidates was inter- viewed; instead, along with board discussions, a wide range of associates were invited to nominate people they would be willing to follow. ‘ We weren't given a list of names – we were free to choose anyone in the com- pany ’, she says. ‘ To my surprise, it was me. ’

The explicit aim of the company's culture is to ‘com- bine freedom with co-operation and autonomy with synergy ’. Everyone can earn the credibility to define and drive projects. Sponsors help associates chart a course in the organization that will offer personal fulfilment while maximizing their contribution to the enterprise. Associates adhere to four basic guiding principles, origi- nally expressed by Bill Gore:

● Fairness to each other and everyone with whom we come in contact.

● Freedom to encourage, help, and allow other asso- ciates to grow in knowledge, skill, and scope of responsibility.

● The ability to make one's own commitments and keep them.

● Consultation with other associates before undertak- ing actions that could impact the reputation of the company.

This degree of personal commitment and con- trol by associates would not sit happily with a large ‘corporate’-style organization. It is no surprise, then, that Gore has unusual notions of economies of scale. Bill Gore believed in the need ‘to divide so that you can multi- ply ’. So when units grow to around 200 people, they are

W.L. Gore 2

So u

rc e:

S h

u tt

er st

o ck

.c o

m : M

ar q

u es

M09_SLAC8678_08_SE_C09.indd 278 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 279

Believe, know and behave Culture is difficult to explain. As was said of one organization with a particularly strong cul- ture (a university as it happens), ‘ From the outside looking in, you can’t understand it. From the inside looking out, you can’t explain it. ’ 3 As far as the operations function is concerned, it is best summed up by what the operations team believes , what it knows and how it behaves . It is these three elements of operations culture – belief, knowledge and behaviour – which provide the foundations for how it contributes to the business and how capable it is to improve over time:

● What operations believe – By ‘operations belief’, we mean what the people within the operations function accept as self-evident. For example, do operations believe that they have a responsibility to understand fully all other functions’ strategies and their implica- tions for operations; do they develop capabilities within their operations resources and processes that offer a unique and long-lasting strategic advantage?

● What operations should know – What should the operations team know? Obviously it should understand the underlying principles that govern how operations and processes work. Only with a thorough understanding of the objectives, concepts, tools and tech- niques of operations management will the operations function ever contribute fully to the success of any business.

● How operations should behave – The way operations managers should behave is not fundamentally different from any effective manager. The popular and academic literature have for decades been full of ‘key behaviours’ for effective leadership, and they do not seem to have changed much for years: ‘Don’t micromanage your team, empower them while still being available for advice.’ ‘Be a coach to your team.’ ‘Be clear and results-oriented, but help the team to see how they can achieve them.’ ‘Have a clear vision and strategy.’ ‘Always communicate; both ways – and listen to your team.’ ‘Support open discussion and listen to the team’s concerns.’ All of which are may be obvious, but make good managerial sense.

HOW DO OPERATIONS MANAGERS CONTRIBUTE TO HUMAN RESOURCE STRATEGY?

Human resource strategy is the overall long-term approach to ensuring that an organization’s human resources provide a strategic advantage. It involves two interrelated activities. First, identifying the number and type of people that are needed to manage, run and develop the organization so that it meets its strategic business objectives. Second, putting in place the programmes and initiatives that attract, develop and retain appropriate staff. It is an essential

usually split up, with these small facilities organized in clusters or campuses. Ideally a dozen or so sites are close enough to permit good communication and knowl- edge exchange, but can still be intimate yet separate enough to promote a feeling of ownership. Bill Gore also believed that people come to work to be innovative and had a desire to invent great products. This, he said, ‘ would be the glue holding the company together ’, rather than the official procedures other companies rely on. And at Gore's Livingstone plant in Scotland the story of ‘the breathable bagpipes’ is used to illustrate this type of crea- tive innovation generated from the company's culture of trust that allows people to follow their passion. The story goes that an associate who worked in Gore's filter bags

department at Livingstone was also a keen exponent of his national instrument – the bagpipes. By day he would be working on filter systems, in the evening he would play his bagpipes. It occurred to him that the physical properties of the product he was putting together dur- ing the day could make a synthetic bag for the pipes he played in the evening. Traditionally, bagpipes have a bag made from sheepskin or cow leather which fills up with moisture and becomes a smelly health hazard. He recog- nized that if you added Gore-Tex, it would be breathable and it would be dry. He put a prototype together, tried it, and it worked. So he decided to spend time developing it, created a team to develop it further, and now almost all Scottish bagpipes have a Gore-Tex bag in them.

M09_SLAC8678_08_SE_C09.indd 279 6/2/16 1:32 PM

280 PART TWO DESIGNING THE OPERATION

activity. Here is what Accenture, one of the top consultancies in the world, has to say about it 4 : ‘ Attention to people is more critical than ever…a company’s workforce has become increasingly important to business success – so much so that most senior executives now view people and work- force issues as a critical competitive differentiator and one of their top agenda items…A superior workforce – supported by highly effective, flexible and business-oriented HR and learning organ- izations – will be essential to achieving these objectives and taking greater strides toward high performance. ’

Developing the specific details of an HR strategy is outside the scope of this book. Yet one set of issues is directly relevant: that is, how can operations managers make sure that they are well served by, and contribute to, the strategy?

An influential contribution to the strategic role of HR comes from Dave Ulrich, 5 at the University of Michigan. His assumption is that traditional HR departments are often inadequate at fulfilling a mean- ingful strategic role. He proposes four elements to the HR activity:

being a ‘strategic partner’ to the business, administering HR procedures and processes, being an ‘employee champion’, and being a ‘change agent’. Table 9.1 explains each role and suggests how operations managers can be associated with each role.

It is important to recognize the interdependence of all the activities in Table 9.1 . Managers may focus only on whatever of these activities currently demands attention. But, just as in the operations function generally, people issues are inter-reliant. There is little point in attempt- ing, for example, to develop a more egalitarian team-based structure and then fail to change the organization’s training or reward procedures. This is why a strategic perspective aimed at identifying the relationship between all four roles is necessary, and why the first step in developing an HR strategy is to understand the organization’s overall strategy. In particular, key questions are: What are the implications of the strategy for HR? And how can the people in the organization contribute to successfully achieving the strategy?

Work-related stress The idea that there is a link between HR strategy and the incidence of stress at work is not new. Even some of the early ‘scientific management’ pioneers accepted that working arrangements should not result in conditions that promoted stress. Now it is generally accepted that stress

✽ ✽ ✽ Operations principle Operations principle Operations principle

Table 9.1 Ulrich's HR roles and their relevance to operations managers

HR role What it involves Relevance to operations management (OM)

Strategic partner

Aligning HR and business strategy: ‘organizational diagnosis’, staff planning, environmental monitoring, etc.

OM integrates operations strategy with HR strategy. OM both specifi es its long-term skills requirements and relies on HR to supply/develop them informed by labour market forecasts, succession planning, etc.

Administrative expert

Running the organization's HR processes and ‘shared services’: payroll, appraisal, selection and recruitment, communication, etc.

OM is largely an ‘internal customer’ for HR's processes. OM must be clear in its requirements with agreed service levels mutually negotiated. Note that OM should also be able to advise HR on how to design and manage its processes effi ciently and eff ectively

Employee champion

Listening and responding to employees: ‘providing resources to employees’, conciliation, career advice, grievance procedures, etc.

OM and HR must develop a good working relationship and clear procedures to deal with any ‘emergency ’ issues that arise. Also OM must be sensitive to feedback from HR on how it manages day-to-day operations

Change agent Managing transformation and change: ‘ensuring capacity for change’, management development, performance appraisal, organization development, etc.

OM and HR are jointly responsible for operations improvement activities. HR has a vital role in all the cultural, developmental and evaluation activities associated with improvement

M09_SLAC8678_08_SE_C09.indd 280 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 281

can seriously undermine the quality of people’s working lives and, in turn, their effectiveness in the workplace. Here stress is defined as ‘the adverse reaction people have to excessive pres- sures or other types of demand placed on them’.6 In addition to the obvious ethical reasons for avoiding work-related stress, there are also business-related benefits, such as the following:

● Staff feel happier at work, their quality of working life is improved and they perform better. ● Introducing improvements is easier when ‘stress’ is managed effectively. ● Employment relations: problems can be resolved more easily. ● Attendance levels increase and sickness absence reduces.

Table 9.2 illustrates some of the causes of stress at work and what operations managers can do about it.

HOW CAN THE OPERATIONS FUNCTION BE ORGANIZED?

There are two issues here. The first is ‘how should the total set of processes and resources that produce products and services be organized?’ The second is ‘how should operations manag- ers, who make the decisions about operations, position themselves relative to the rest of the operations function’? We will look at the first issue by examining some common forms of organizational structures, and the second by examining the role of operations ‘decision mak- ing’. First, though, it is worth looking at how ‘organizations’ can be described.

Perspectives on organizations7

How we illustrate organizations says much about our underlying assumptions of what an ‘organization’ is and how it is supposed to work. For example, the illustration of an organi- zation as a conventional ‘organogram’ implies that organizations are neat and controllable with unambiguous lines of accountability. But this is rarely the case. In fact taking such a

Table 9.2 Causes of stress at work and what could be done about it

Causes of stress What can be done about it

Staff can become overloaded if they cannot cope with the amount of work or type of work they are asked to do

Change the way the job is designed, training needs and whether it is possible for employees to work more flexible hours

Staff can feel disaffected and perform poorly if they have no control or say over how and when they do their work

Actively involve staff in decision making, the contribution made by teams, and how reviewing performance can help identify strengths and weaknesses

Staff feel unsupported: levels of sick absence often rise if employees feel they cannot talk to managers about issues that are troubling them

Give staff the opportunity to talk about the issues causing stress, be sympathetic and keep them informed

A failure to build relationships based on good behaviour and trust can lead to problems related to discipline, grievances and bullying

Check the organization's policies for handling grievances, unsatisfactory performance, poor attendance and misconduct, and for tackling bullying and harassment

Staff will feel anxious about their work and the organization if they do not know their role and what is expected of them

Review the induction process, work out an accurate job description and maintain a close link between individual targets and organizational goals

Change can lead to huge uncertainty and insecurity

Plan ahead so change is not unexpected. Consult with employees so they have a real input, and work together to solve problems

M09_SLAC8678_08_SE_C09.indd 281 6/2/16 1:32 PM

282 PART TWO DESIGNING THE OPERATION

mechanistic view may be neither appropriate, nor desirable. Seeing an organization as though it was unambiguously machine-like is just one of several metaphors commonly used to understand organiza- tions. One well-known analysis by Gareth Morgan proposes a number of ‘images’ or ‘metaphors’ which can be used to understand organiza- tions, as follows:

● Organizations are machines – The resources within organizations can be seen as ‘compo- nents’ in a mechanism whose purpose is clearly understood. Relations within the organiza- tion are clearly defined and orderly, processes and procedures that should occur usually do occur, and the flow of information through the organization is predictable. Such mechani- cal metaphors appear to impose clarity on what is actually messy organizational behaviour. But, where it is important to impose clarity (as in much operations analysis) such a meta- phor can be useful, and is the basis of the ‘process approach’ used in this and similar books.

● Organizations are organisms – Organizations are living entities. Their behaviour is dic- tated by the behaviour of the individual humans within them. Individuals, and their organ- izations, adapt to circumstances just as different species adapt to the environment. This is a particularly useful way of looking at organizations if parts of the environment (such as the needs of the market) change radically. The survival of the organization depends on its ability to exhibit enough flexibility to respond to its environment.

● Organizations are brains – Like brains, organizations process information and make deci- sions. They balance conflicting criteria, weigh up risks and decide when an outcome is acceptable. They are also capable of learning, changing their model of the world in the light of experience. This emphasis on decision making, accumulating experience and learning from that experience is important in understanding organizations. They consist of conflicting groups where power and control are key issues.

● Organizations are cultures – An organization’s culture is usually taken to mean its shared values, ideology, pattern of thinking and day-to-day ritual. Different organizations will have different cultures stemming from their circumstances and their history. A major strength of seeing organizations as cultures is that it draws attention to their shared ‘enact- ment of reality’. Looking for the symbols and shared realities within an organization allows us to see beyond what the organization says about itself.

● Organizations are political systems – Organizations, like communities, are governed. The system of government is rarely democratic, but nor is it usually a dictatorship. Within the mechanisms of government in an organization are usually ways of understanding alter- native philosophies, ways of seeking consensus (or at least reconciliation) and sometimes ways of legitimizing opposition. Individuals and groups seek to pursue their aims through the detailed politics of the organization. They form alliances, accommodate power rela- tionships and manage conflict. Such a view is useful in helping organizations to legitimize politics as an inevitable aspect of organizational life.

Forms of organizational structure There are many different ways of defining ‘organizational structure’; here it is seen as the way in which tasks and responsibilities are divided into distinct groupings, and how the responsibility and co-ordination relationships between the groupings are defined. Most organizational designs attempt to divide an organization into discrete parts that are given some degree of authority to make decisions within their part of the organization. All but the very smallest of organizations needs to delegate decision making in this way; it allows specialization so decisions can be taken by the most appropriate people. The main issue is what dimension of specialization should be used when grouping parts of the organization together. There are three basic approaches to this:

● Group resources together according to their functional purpose – so, for example, sales, marketing, operations, research and development, finance, etc.

✽ ✽ ✽ Operations principle Operations principle Operations principle

M09_SLAC8678_08_SE_C09.indd 282 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 283

● Group resources together by the characteristics of the resources themselves – this may be done, for example, by clustering similar technologies together (extrusion technology, roll- ing, casting, etc.). Alternatively, it may be done by clustering similar skills together (audit, mergers and acquisitions, tax, etc.). It may also be done according to the resources required for particular products or services (chilled food, frozen food, canned food, etc.).

● Group resources together by the markets which the resources are intended to serve – again this may be done in various ways. Markets may be defined by location, with distinct geo- graphical boundaries (North America, South America, Europe and the Middle East, South- East Asia, etc.). Alternatively, markets may be defined by the type of customer (small firms, large national firms, large multinational firms, etc.).

There are an almost infinite number of possible organizational structures. However, some pure types of organization have emerged that are useful in illustrating different approaches to organizational design, even if, in their pure form, they are rarely found:

● The U-form organization – The unitary form, or U-form, organization clusters its resources primarily by their functional purpose. Figure 9.3(a) shows a typical U-form organization with a pyramid management structure, each level reporting to the manage- rial level above. Such structures can emphasize process efficiency above customer service and the ability to adapt to changing markets. But the U-form keeps together expertise and can promote the creation and sharing of technical knowledge. The problem then with the U-form organization is not so much the development of capabilities, but the flexibility of their deployment.

● The M-form organization – This form of organizational structure emerged because the functionally based structure of the U-form was cumbersome when companies became

Figure 9.3 (a) U-form organizations give prominence to functional groupings of resources. (b) The M-form separates the organization's resources into separate divisions. (c) Matrix form structures the organization's resources so that they have two (or more) levels of responsibility. (d) N-form organizations form loose networks internally between groups of resources and externally with other organizations

M09_SLAC8678_08_SE_C09.indd 283 6/2/16 1:32 PM

284 PART TWO DESIGNING THE OPERATION

large, often with complex, markets. It groups together either the resources needed for each product or service group, or, alternatively, those needed to serve a particular geographical market, in separate divisions. The separate functions may be distributed throughout the different divisions (see Fig. 9.3(b)), which can reduce economies of scale and operating efficiency. But it does allow each individual division to focus on the specific needs of their markets.

● Matrix forms – Matrix structures are a hybrid, usually combining the M-form with the U-form. In effect, the organization has simultaneously two different structures (see Fig. 9.3(c)). In a matrix structure each resource cluster has at least two lines of authority, for example both to the division and to the functional groups. While a matrix organization ensures the representation of all interests within the company, it can be complex and sometimes confusing.

● The N-form organization – The ‘N’ in N-form stand for ‘network’. In N-form organiza- tions, resources are clustered into groups as in other organizational forms, but with more delegation of responsibility for the strategic management of those resources. N-forms have relatively little hierarchical reporting and control. Each cluster of resources is linked to the others to form a network, with the relative strength of the relationships between clusters changing over time, depending on circumstances (see Fig. 9.3(d)). Senior management set broad goals and attempt to develop a unifying culture but do not ‘command and control’ to the same extent as in other organizational forms.

Operations ‘developers’ – ‘staff ’ and ‘line’ roles Traditionally, it was common to distinguish between two types of roles in organizations. People occupying classic ‘staff’ positions had a monitoring, planning, shaping and ‘developing’ role. They are the ones who are charged with building up the company’s operations strategic capabil- ity. It is a task that needs some organizational ‘space’ to be performed effectively. It is certainly not a task that co-exists easily with the hectic and immediate concerns of running an operation on a day-to-day basis. These people constitute what could be termed the ‘operations developers’ or ‘central operations’. They perform what are called (slightly confusingly) ‘staff’ roles. By con- trast, people occupying ‘line’ roles are those who run the day-to-day operations. Theirs is partly a reactive role, one that involves finding ways round unexpected problems: reallocating resources, adjusting processes, solving quality problems, and so on. They need to look ahead only enough to make sure that resources are available to meet targets. Theirs is the necessary routine. Knowing where the operation is heading, keeping it on budget and pulling it back on course when the unexpected occurs – no less valuable a task than the developer’s but very different.

While these descriptions are clearly stereotypes, they do represent two types of operations task. The issue, for organizational design, is whether it is wise to separate them organizationally. It may cause more problems than it solves. Although it allows each to concentrate on their differ- ent jobs, it also can keep apart the two sets of people who have most to gain by working together. Here is the paradox: the development role does need freedom from the immediate pressures of day-to-day management but it is crucial that it understands the exact nature of these pressures. What makes the operation distinctive? Where do the problems occur? What improvements would make most difference to the performance of the operation? These are questions answered only by living with the operation, not cloistered away from it. Similarly, the day-to-day operations man- ager has to interpret the workings of the operation, collect data, explain constraints and educate developers. Without the trust and co-operation of each, neither set of managers can be effective.

Four types of operations developer role We can use the dimensions which define the perspectives on operations strategy described in Chapter 3 to examine the role that operations developers play within the operations function:

● Top down or bottom up? If operations developers have a predominantly top-down view of the world, they are likely to take a programmatic approach to activities, emphasizing the

M09_SLAC8678_08_SE_C09.indd 284 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 285

implementation of overall company strategy. Conversely, if they take a bottom-up view, they are more likely to favour a more emergent model of operations development where individ- ual business operations together contribute to the overall building of operations expertise.

● Market requirements or operations resource focus? If operations developers take a market requirements view of operations development, they are likely to focus on the explicit performance achieved by each part of the operations function and how far that performance serves to satisfy the operation’s customers. An operations resource focus, on the other hand, emphasizes the way in which each part of the operation function develops its competences and successfully deploys them in its marketplaces.

We can use these two dimensions to define a typology of how the operations developer role could work, as shown in Figure 9.4. It classifies operations developers into four pure types called governors, curators, trainers and facilitators – a typology. Although, in practice, the central operations function of most businesses is a combination of these pure types, usually one type predominates.

● Operations developers as governors – The term ‘governor’ is used to describe an agent of a central authority, interpreting operations strategy and arbitrating over any disputes. The term is also used to denote the mechanism that sets clear goals for each part of the operation, judges their performance and, if performance is not to target, wants to know the reason why.

● Operations developers as curators – Operations developers can be concerned primarily with performance against market requirements without being top down. They may take a more emergent view by acting as the repository of performance data and ideas regarding operations practice for the company as a whole. The term ‘curator’ is used to capture this idea. Operations developers therefore will be concerned with collecting performance information, examples of best practice, and so on. They will also be concerned with disseminating this information so that operations managers in different parts of the business can benchmark themselves against their colleagues and, where appropriate, adopt best practice from elsewhere.

● Operations developers as trainers – Moving from the market requirements to the oper- ations resources emphasis shifts the focus more to the development of internal capabili- ties. If the mindset of operations developers is top down their role becomes ‘trainers’, who

Trainer Governor

Facilitator Curator

Top down

Programmatic

Emergent

C a

p a

b ili

ti e

s fo

c u

s

P e

rf o

rm a

n c

e f

o c

u s

Controlling the performance of the operations by setting clear priorities and measuring performance against targets

Nurturing the performance of the operations by collecting performance data and distributing comparative performance information

Enabling operations in the development and deployment of their capabilities through shared advice, support and learning

Instructing operations in the development and deployment of their capabilities through standardized improvement methods

Bottom up

Market requirements

Operations resources

Figure 9.4 A typology of the ‘operations developer’ role

M09_SLAC8678_08_SE_C09.indd 285 6/2/16 1:32 PM

286 PART TWO DESIGNING THE OPERATION

develop clear objectives, usually derived from overall company strategy, and devise effec- tive methods of developing the various parts of the overall operation. And because the needs of individual parts of the operation may differ, ‘trainer’ operations developers may devise improvement methodologies that can, to some extent, be customized.

● Operations developers as facilitators – In some ways this final type of operations devel- oper role is the most difficult to operate effectively. They are again concerned with the development of operations capabilities but do so by acting as facilitators of change rather than instructors. Their role is to advise, support and generally aid the development and deployment of capabilities through a process of mentoring the various parts of the oper- ation. They share responsibility with day-to-day operations managers in forming a com- munity of operations practice. Implicit in this type of operations developer role is the acceptance of a relatively long-term approach to operations development.

HOW DO WE GO ABOUT DESIGNING JOBS?

Job design is concerned with how we structure each individual’s job, the team to which they belong (if any), their workplace and their interface with the technology they use. In this sec- tion we deal with what is usually considered to be the central people-related responsibility of operations managers – job design. It is a huge topic and we can only deal with some of the influences on, and approaches to, it.

The influences on job design that we deal with here are illustrated in Figure 9.5.

The decisions of job design Job design involves a number of separate yet related elements:

● What tasks are to be allocated to each person in the operation? Producing goods and services involves a whole range of different tasks which need to be divided between the people who staff the operation. Different approaches to the division of labour will lead to different task allocations.

● What is the best method of performing each job? Every job should have an approved (or best) method of completion. And although there are different ideas of what is ‘best’, it is generally the most efficient method that fits the task, and does not unduly interfere with other tasks.

Design individuals’ and groups’ jobs

‘Behavioural’ approaches

Flexible working

Division of labour

Team working

Ergonomics ‘Scientific’

management

Tele- commuting

Design the working environment

Figure 9.5 The main influences on job design, work time allocation and the design of the working environment

M09_SLAC8678_08_SE_C09.indd 286 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 287

● How long will it take and how many people will be needed? Work measurement helps us calculate the time required to do a job, and therefore how many people will be needed.

● How do we maintain commitment? Understanding how we can encourage people and maintain job commitment is, arguably, the most important of the issues in job design. This is why behavioural approaches, including empowerment, team work and flexible working are at the core of job design.

● What technology is available and how will it be used? Many operational tasks require the use of technology. Not only does the technology need to be appropriately designed, but also so does the interface between the people and the hardware.

● What are the environmental conditions of the workplace? The conditions under which jobs are performed will have a significant impact on people’s effectiveness, Although often considered a part of job design, we treat it separately in this chapter.

OPERATIONS IN PRACTICE

Those jobs that are on the front line of dealing directly with customers (particularly a lot of customers, all the time, of all different types) can be particularly stressful. Not all customers will be reasonable, patient, courteous or even sane. The people who have these high customer contact roles need support, training and perhaps a spe- cial aptitude. And there is plenty of advice for staff that have to deal with customers who are angry because they feel that the level of service they have received is inadequate. Such advice usually includes such things as: acknowledge the (perceived) problem, try to put yourself in the position of the complainer, get the all facts straight, and try to rectify the problem. Not easy, but if complaints can be resolved to the satisfaction of the customer, there can be significant benefits. Some surveys indicate that 90 per cent of customers whose complaints are resolved are happy to use the service again, and may even go on to become advocates for the service. Nevertheless, maintaining tolerance and polite- ness in the face of some particularly difficult custom- ers can be more than even experienced staff can bear. That certainly was the case with Steven Slater, formerly an air steward on the US airline JetBlue. He was work- ing on a flight to New York and had to arbitrate when a female passenger began arguing with a male passenger about space in the overhead luggage compartment dur- ing boarding. The female passenger swore at Mr Slater and pulled down the compartment door on his head. Later, when the aircraft landed, she seemingly refused to follow his request to remain in her seat and got up to take her bag from the overhead locker while the aircraft was still taxiing. Again, the woman allegedly swore at Mr Slater. It was then that his patience ran out in a

particularly dramatic fashion. He went to the intercom and broadcast to everybody on board: ‘ To the passenger who just called me a motherf *****: F*** you. I've been in this business for 28 years and I've had it. ’ He then col- lected his hand- luggage (and two beers from the trolley) opened the cabin door, activated the inflatable chute, announced ‘ to those of you who have shown dignity and respect for 20 years, have a great ride ’ and slid out of the (fortunately stationary) aircraft onto the runway. As a way to give up your job, it is not recommended. He was later arrested and charged with criminal mischief and reckless endangerment.

The stress of high customer contact jobs 8

So u

rc e:

S h

u tt

er st

o ck

.c o

m : A

lp h

as p

ir it

M09_SLAC8678_08_SE_C09.indd 287 6/2/16 1:32 PM

288 PART TWO DESIGNING THE OPERATION

Task allocation – the division of labour Any operation must decide on the balance between using specialists or generalists. This idea is related to the division of labour – dividing up the total task into smaller parts, each of which is accomplished by a single person or team. It was first formalized as a concept by the econ- omist Adam Smith in his Wealth of Nations in 1746. Perhaps the epitome of the division of labour is the assembly line, where products move along a single path and are built up by oper- ators continually repeating a single task. This is the predominant model of job design in most mass-produced products and in some mass-produced services (fast food, for example). There are some real advantages in division-of-labour principles:

● It promotes faster learning . It is obviously easier to learn how to do a relatively short and simple task than a long and complex one. This means that new members of staff can be quickly trained and assigned to their tasks when they are short and simple.

● Automation becomes easier . Dividing a total task into small parts raises the possibility of automating some of those small tasks. Substituting technology for labour is considerably easier for short and simple tasks than for long and complex ones.

● Reduced non-productive work . This is probably the most important benefit of division of labour. In large, complex tasks the proportion of time spent picking up tools and materials, putting them down again and generally finding, positioning and searching can be very high indeed. For example, one person assembling a whole motor car engine would take two or three hours and involve much searching for parts, positioning, and so on. Around half of the person’s time would be spent on these reaching, positioning, finding tasks (called non-productive elements of work). Now consider how a motor car engine is actually made in practice. The total job is probably divided into 20 or 30 separate stages, each staffed by a person who carries out only a proportion of the total. Specialist equipment and materials handling devices can be devised to help them carry out their job more efficiently. Furthermore, there is relatively little finding, positioning and reaching involved in this simplified task. Non-productive work can be considerably reduced, perhaps to under 10 per cent, which would be very significant to the costs of the operation.

However, there are also serious drawbacks to highly divided jobs:

● Monotony . The shorter the task, the more often operators will need to repeat it. Repeating the same task, for example every 30 seconds, eight hours a day and five days a week, can hardly be called a fulfilling job. As well as any ethical objections, there are other, more obviously practical objections to jobs which induce such boredom. These include the increased likelihood of absenteeism and staff turnover, the increased likelihood of error and even deliberate sabotage of the job.

● Physical injury . The continued repetition of a very narrow range of movements can, in extreme cases, lead to physical injury. The overuse of some parts of the body (especially the arms, hands and wrists) can result in pain and a reduction in physical capability. This is sometimes called repetitive strain injury (RSI).

● Low f lexibility . Dividing up a task into many small parts often gives the job design a rigidity which is difficult to change under changing circumstances. For example, if an assembly line has been designed to make one particular product but then has to change to manufacture a quite different product, the whole line will need redesigning. This will probably involve changing every operator’s set of tasks, which can be a long and difficult procedure.

● Poor robustness . Highly divided jobs imply materials (or informa- tion) passing between several stages. If one of these stages is not working correctly, for example because some equipment is faulty, the whole operation is affected. On the other hand, if each person is performing the whole of the job, any problems will only affect that one person’s output.

✽ ✽ ✽ Operations principle Operations principle Operations principle

M09_SLAC8678_08_SE_C09.indd 288 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 289

Designing job methods – scientific management The term ‘scientific management’ became established in 1911 with the publication of the book of the same name by Fredrick Taylor (this whole approach to job design is sometimes referred to, pejoratively, as Taylorism). In this work he identified what he saw as the basic tenets of scientific management: 9

● All aspects of work should be investigated on a scientific basis to establish the laws, rules and formulae governing the best methods of working.

● Such an investigative approach to the study of work is necessary to establish what consti- tutes a ‘fair day’s work’.

● Workers should be selected, trained and developed methodically to perform their tasks. ● Managers should act as the planners of the work (analysing jobs and standardizing the

best method of doing the job) while workers should be responsible for carrying out the jobs to the standards laid down.

● Co-operation should be achieved between management and workers based on the ‘maxi- mum prosperity’ of both.

The important thing to remember about scientific management is that it is not particularly ‘scientific’ as such, although it certainly does take an ‘investigative’ approach to improving operations. Perhaps a better term for it would be ‘systematic management’. It gave birth to two separate, but related, fields of study: method study, which determines the methods and activities to be included in jobs; and work measurement, which is concerned with measur- ing the time that should be taken for performing jobs. Together, these two fields are often referred to as work study and are explained in detail in the supplement to this chapter.

Critical commentary

Even in 1915, criticisms of the scientifi c management approach were being voiced. In a sub- mission to the United States Commission on Industrial Relations, scientifi c management is described as:

● being in ‘spirit and essence a cunningly devised speeding up and sweating system’;

● intensifying the ‘modern tendency towards specialization of the work and the task’;

● condemning ‘the worker to a monotonous routine’;

● putting ‘into the hands of employers an immense mass of information and methods that may be used unscrupulously to the detriment of workers’;

● tending to ‘transfer to the management all the traditional knowledge, the judgement and skills of workers’;

● greatly intensifying ‘unnecessary managerial dictation and discipline’;

● tending to ‘emphasize quantity of product at the expense of quality ’.

Two themes evident in this early criticism do warrant closer attention. The fi rst is that scientifi c management inevitably results in standardization of highly divided jobs and thus reinforces the negative eff ects of excessive division of labour previously mentioned. Second, scientifi c management formalizes the separation of the judgemental, planning and skilled tasks, which are done by ‘management’, from the routine, standardized and low-skill tasks, which are left for ‘operators’. Such a separation, at the very least, deprives the majority of staff of an opportunity to contribute in a meaningful way to their jobs (and, incidentally, deprives the organization of their contribution). Both of these themes in the criticisms of scientifi c management lead to the same point: that the jobs designed under strict scientifi c management principles lead to low motivation among staff , frustration at the lack of control over their work, and alienation from the job.

M09_SLAC8678_08_SE_C09.indd 289 6/2/16 1:32 PM

290 PART TWO DESIGNING THE OPERATION

Designing the human interface - ergonomic workplace design Ergonomics is concerned primarily with the physiological aspects of job design. Physiology is about the way the body functions. It involves two aspects: first, how a person interfaces with their immediate working area; second, how people react to environmental conditions. We will examine the second aspect of ergonomics later in this chapter. Ergonomics is sometimes referred to as human factors engineering or just ‘human factors’. Both aspects are linked by two common ideas:

● There must be a fit between people and the jobs they do. To achieve this fit there are only two alternatives. Either the job can be made to fit the people who are doing it, or, alternatively, the people can be made (or perhaps less radically, recruited) to fit the job. Ergonomics addresses the former alternative.

● It is important to take a ‘scientific’ approach to job design, for example collecting data to indicate how people react under different job design conditions and trying to find the best set of conditions for comfort and performance.

Anthropometric aspects Many ergonomic improvements are primarily concerned with what are called the anthro- pometric aspects of jobs – that is, the aspects related to people’s size, shape and other phys-

ical abilities. The design of an assembly task, for example, should be governed partly by the size and strength of the operators who do the job. The data which ergonomists use when doing this is called anthropometric data. Because we all vary in our size and capabilities, ergonomists are particularly interested in our range of capabilities, which is why anthropometric data is usually expressed in percentile terms. Figure 9.6 illustrates the idea. This shows the idea of size (in

this case height) variation. Only 5 per cent of the population are smaller than the person on the extreme left (5th percentile), whereas 95 per cent of the population are smaller than the

✽ ✽ ✽ Operations principle Operations principle Operations principle

Figure 9.6 The use of anthropometric data in job design

M09_SLAC8678_08_SE_C09.indd 290 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 291

person on the extreme right (95th percentile). When this principle is applied to other dimen- sions of the body, for example arm length, it can be used to design work areas. Figure 9.6 also shows the normal and maximum work areas derived from anthropometric data. It would be inadvisable, for example, to place frequently used components or tools outside the maximum work area derived from the 5th percentile dimensions of human reach.

Designing for job commitment – behavioural approaches to job design Jobs that are designed purely on division of labour, scientific management or even purely ergonomic principles can alienate the people performing them. Job design should also take into account the desire of individuals to fulfil their needs for self-esteem and personal devel- opment. This is where motivation theory and its contribution to the behavioural approach to job design is important. This achieves two important objectives of job design. First, it provides jobs which have an intrinsically higher quality of working life – an ethically desirable end in itself. Second, because of the higher levels of motivation it engenders, it is instrumental in achieving better performance for the operation, in terms of both the quality and the quantity of output. This approach to job design involves two conceptual steps: first, exploring how the various characteristics of the job affect people’s motivation; second, exploring how individu- als’ motivation towards the job affects their performance at that job.

Typical of the models which underlie this approach to job design is that by Hackman and Oldham shown in Figure 9.7.10 Here a number of ‘techniques’ of job design are recom- mended in order to affect particular core ‘characteristics’ of the job. These core character- istics are held to influence various positive ‘mental states’ towards the job. In turn, these are assumed to give certain performance outcomes. In Figure 9.7 some of the ‘techniques’ (which Hackman and Oldham originally called ‘implementing concepts’) need a little fur- ther explanation:

● Combining tasks means increasing the number of activities allocated to individuals. ● Forming natural work units means putting together activities which make a coherent

whole. ● Establishing client relationships means that staff make contact with their internal custom-

ers directly. ● Vertical loading means including ‘indirect’ activities (such as maintenance). ● Opening feedback channels means that internal customers feed back perceptions directly.

Figure 9.7 A typical ‘behavioural’ job design model

M09_SLAC8678_08_SE_C09.indd 291 6/2/16 1:32 PM

292 PART TWO DESIGNING THE OPERATION

Hackman and Oldham also indicate how these techniques of job design shape the core characteristics of the resulting job and, further, how the core characteristics inf luence people’s ‘mental states’. Mental states are the attitudes of individuals towards their jobs – specifically, how meaningful they find the job, how much responsibility and control they feel they have over the way the job is done, and how much they understand about the results of their efforts. All of these mental states inf luence people’s performance at their job in terms of their motivation, quality of work, satisfaction with their work, turnover and absenteeism.

Job rotation If increasing the number of related tasks in the job is constrained in some way, for exam- ple by the technology of the process, one approach may be to encourage job rotation. This means moving individuals periodically between different sets of tasks to provide some variety in their activities. When successful, job rotation can increase skill flexibility and make a small contribution to reducing monotony. However, it is not viewed as universally beneficial either by management (because it can disrupt the smooth flow of work) or by the people performing the jobs (because it can interfere with their rhythm of work).

Job enlargement The most obvious method of achieving at least some of the objectives of behavioural job design is by allocating a larger number of tasks to individuals. If these extra tasks are broadly of the same type as those in the original job, the change is called job enlargement. This may not involve more demanding or fulfilling tasks, but it may provide a more complete and therefore slightly more meaningful job. If nothing else, people performing an enlarged job will not repeat themselves as often, which could make the job marginally less monotonous. So, for example, suppose that the manufacture of a product has traditionally been split up on an assembly line basis into 10 equal and sequential jobs. If that job is then redesigned so as to form two parallel assembly lines of five people, the output from the system as a whole would be maintained but each operator would have twice the number of tasks to per- form. This is job enlargement. Operators repeat themselves less frequently and presumably the variety of tasks is greater, although no further responsibility or autonomy is necessarily given to each operator.

Job enrichment Job enrichment means not only increasing the number of tasks, but also allocating extra tasks which involve more decision making, greater autonomy and greater control over the job. For example, the extra tasks could include maintenance, planning and control, or mon- itoring quality levels. The effect is both to reduce repetition in the job and to increase auton- omy and personal development. So, in the assembly line example, each operator, as well as being allocated a job which is twice as long as that previously performed, could also be allo- cated responsibility for carrying out routine maintenance and such tasks as record keeping and managing the supply of materials. Figure 9.8 illustrates the difference between what are called horizontal and vertical changes. Broadly, horizontal changes are those which extend the variety of similar tasks assigned to a particular job. Vertical job changes are those which add responsibilities, decision making or autonomy to the job. Job enlargement implies move- ment only on the horizontal scale, whereas job enrichment certainly implies movement on the vertical scale and perhaps on both scales.

Empowerment Empowerment is an extension of the autonomy job characteristic prominent in the behav- ioural approach to job design. However, it is usually taken to mean more than autonomy. Whereas autonomy means giving staff the ability to change how they do their jobs, empow- erment means giving staff the authority to make changes to the job itself, as well as how

M09_SLAC8678_08_SE_C09.indd 292 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 293

Figure 9.8 Job enlargement and job enrichment

OPERATIONS IN PRACTICE

In what was thought to be the first contract of its type in the UK, McDonald's, the quick-service restaurant chain, announced that it was to allow family members to cover each other's jobs. Under the deal members of the same family working in the same outlet would be able to work each other's shifts without giving any prior notice or getting a manager's permission. The company said that it hoped the contracts would ‘ encourage people to become fully trained and fully rotatable ’ . But that the main aim was to ‘ cut absenteeism and improve staff retention ’. ‘ It's great ’, said one McDonald's employee. ‘ Depending on how we feel in a morning, we decide which one of us wants to go in and work .’ Although the scheme is currently limited to family mem- bers only, McDonald's said that it might consider extending it to cover friends who work at the same restaurant.

McDonald's lets families share job

it is performed. This can be designed into jobs to different degrees. 11 At a minimum, staff could be asked to contribute their suggestions for how the operation might be improved. Going further, staff could be empowered to redesign their jobs. Further still, staff could be included in the strategic direction and performance of the whole organization. The benefits of empowerment are generally seen as providing fast responses to customer needs (includ- ing dissatisfied customers), employees who feel better about their jobs and who will interact with customers with more enthusiasm, promoting ‘word-of-mouth’ advertising and cus- tomer retention. However, there are costs associated with empowerment, including higher selection and training costs, perceived inequity of service and the possibility of poor deci- sions being made by employees.

Team working A development in job design which is closely linked to the empowerment concept is that of team-based work organization (sometimes called self-managed work teams). This is where staff, often with overlapping skills, collectively perform a defined task and have a high degree

So u

rc e:

S h

u tt

er st

o ck

.c o

m : Z

u ri

je ta

M09_SLAC8678_08_SE_C09.indd 293 6/2/16 1:32 PM

294 PART TWO DESIGNING THE OPERATION

of discretion over how they actually perform the task. The team would typically control such things as task allocation between members, scheduling work, quality measurement and improvement, and sometimes the hiring of staff. To some extent most work has always been a group-based activity. The concept of team work, however, is more prescriptive and assumes a shared set of objectives and responsibilities. Groups are described as teams when the virtues of working together are being emphasized, such as the ability to make use of the various skills within the team. Teams may also be used to compensate for other organizational changes such as the move towards flatter organizational structures. When organizations have fewer managerial levels, each manager will have a wider span of activities to control. Teams which are capable of autonomous decision making have a clear advantage in these circumstances. The benefits of team work can be summarized as:

● improving productivity through enhanced motivation and flexibility; ● improving quality and encouraging innovation; ● increasing satisfaction by allowing individuals to contribute more effectively; ● making it easier to implement technological changes in the workplace because teams are

willing to share the challenges this brings.

Critical commentary

Team work not only is diffi cult to implement successfully, but also can place undue stress on the individuals who form the teams. Some teams are formed because more radical solutions, such as total reorganization, are being avoided. Teams cannot compensate for badly designed organizational processes, nor can they substitute for management's responsibility to defi ne how decisions should be made. Often teams are asked to make decisions but are given insuffi cient responsibility to carry them out. In other cases, teams may provide results but at a price. The Swedish car maker Volvo introduced self- governing teams in the 1970s and 1980s which improved motivation and morale but eventually proved prohibitively expensive. Perhaps most seriously, team work is criticized for substituting one sort of pressure for another. Although teams may be autonomous, this does not mean they are stress-free. Top-down managerial control is often replaced by excessive peer pressure, which is in some ways more insidious.

Flexible working The nature of most jobs has changed significantly over the last 25 years. New technol- ogies, more dynamic marketplaces, more demanding customers and a changed under- standing of how individuals can contribute to competitive success have all had their impact. Also changing is our understanding of how home life, work and social life need to be balanced. Alternative forms of organization and alternative attitudes to work are being sought which allow, and encourage, a degree of f lexibility in working practice which matches the need for f lexibility in the marketplace. From an operations management per- spective, three aspects of f lexible working are significant: skills f lexibility, time f lexibility and location f lexibility.

● Skills flexibility – A flexible workforce that can move across several different jobs could be deployed (or deploy themselves) in whatever activity is in demand at the time. In the short term, staff at a supermarket may be moved from warehouse activities to shelf replenishment in the store or to the checkout, depending on what is needed at the time. In the longer term sense, multi-skilling means being able to migrate individuals from one skill set to another as longer term demand trends become obvious. So, for example,

M09_SLAC8678_08_SE_C09.indd 294 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 295

an engineer who at one time maintained complex equipment by visiting the sites where such equipment was installed may now perform most of his or her activities by using remote computer diagnostics and ‘help line’ assistance. The implication of job flexibil- ity is that a greater emphasis must be placed on training, learning and knowledge man- agement. Defining what knowledge and experience are required to perform particular tasks and translating these into training activities are clearly prerequisites for effective multi-skilling.

● Time f lexibility – Not every individual wants to work full-time. Many people, often because of family responsibilities, only want to work for part of their time, sometimes only during specific parts of the day or week (because of childcare responsibilities etc.). Likewise, employers may not require the same number of staff at all times. They may, for example, need extra staff only at periods of heavy demand. Bringing both the supply of staff and the demand for their work together is the objective of ‘f lexible time’ or f lexi-time working systems. These may define a core working time for each individual member of staff and allow other times to be accumulated f lexibly. Other schemes include annual hours schemes, one solution to the capacity management issue described in Chapter 11 .

● Location f lexibility – The sectoral balance of employment has changed. The service sector in most developed economies now accounts for between 70 and 80 per cent of all employment. Even within the manufacturing sector, the proportion of people with indirect jobs (those not directly engaged in making products) has also increased signif- icantly. One result of all this is that the number of jobs which are not ‘location-specific’ has increased. Location-specific means that a job must take place in one fixed location. So a shop worker must work in a shop and an assembly line worker must work on the assembly line. But many jobs could be performed at any location where there are com- munication links to the rest of the organization. The realization of this has given rise to what is known as telecommuting, teleworking, ‘f lexible working’, ‘home working’, mobile working, and creating the ‘virtual office’. See the ‘Operations in practice’ case of telecommuting (or not) at Yahoo.

Critical commentary

There is always a big diff erence between what is technically possible and what is organizationally feasible. Telecommuting does have its problems. In particular, those types that deny individuals the chance to meet with colleagues often face diffi culties. Problems can include the following:

● Lack of socialization – offi ces are social places where people can adopt the culture of an organization as well as learn from each other. It is naive to think that all knowledge can be codifi ed and learnt formally at a distance.

● Eff ectiveness of communication – a large part of the essential communication we have with our colleagues is unplanned and face to face. It happens on ‘chance meet’ occasions, yet it is important in spreading contextual information as well as establishing specifi c pieces of information necessary to the job.

● Problem solving – it is still often more effi cient and eff ective informally to ask a colleague for help in resolving problems than formally to frame a request using communications technology.

● It is lonely – isolation among mobile or home workers is a real problem. For many of us, the workplace provides the main focus for social interaction. A computer screen is no substitute.

M09_SLAC8678_08_SE_C09.indd 295 6/2/16 1:32 PM

296 PART TWO DESIGNING THE OPERATION

How should the working environment be designed? The aspect of ergonomics that we examined earlier was concerned with how a person inter- faces with the physical aspects of their immediate working area, such as its dimensions. But the subject also examines how people interface with their working environment. By this we mean the temperature, lighting, noise environment, and so on. It will obviously influence the way they are performed. Working conditions which are too hot or too cold, insufficiently illu- minated or glaringly bright, excessively noisy or irritatingly silent, will all influence the way

jobs are carried out. Many of these issues are often covered by occu- pational health and safety legislation which controls environmental conditions in workplaces throughout the world. A thorough under- standing of this aspect of ergonomics is necessary to work within the guidelines of such legislation.

OPERATIONS IN PRACTICE

When Marissa Mayer, the new boss of Yahoo, ruled that employees of the company could no longer work from home, but must come into the office to work, it was met with horror throughout Silicon Valley, and beyond. The news also prompted a debate about how much freedom employees should be given to decide how, when and where they should do their jobs. Perhaps most surpris- ing was that Ms Mayer’s decision seemed to go against the trend, especially in hi-tech companies, to allow and even encourage a degree of what had become known as ‘telecommuting ’ (defined as ‘the practice of working from home for a business and communicating through the use of a personal computer and communication systems’). Surveys had recently shown that home-based working in some industries, especially information sys- tems, engineering and science, was rising particularly quickly. Also, given that many of these technology firms produced the hardware and software that make work- ing from home possible, it seemed only sensible to let their employees use them. As one headline read, ‘ The “9 to 5” mentality is dead ’. And it is not surprising that tel- ecommuting is popular; it has a number of advantages for firms. First, it is popular with (most) staff, so it helps retain (and gain access to a larger pool of ) talent. It also is said to improve productivity by avoiding the some- times distracting work environment. And, of course, because staff spend less time in the office, there can be substantial overhead savings.

Which is possibly why Yahoo’s decision was greeted with such criticism (‘An epic fail’, ‘Hypocrite’, ‘Idiotic’ were just some of the reactions). But it was not a fear that her employees were sitting around in their pyjamas all day that had prompted her decision to send the memo to Yahoo employees banning telecommuting. The leaked memo said that ‘ the habit has slowed the firm down and

made it harder to have serendipitous meetings that can give birth to new ideas ’ and it was the innovation that came from these meetings that the firm required. ‘ We can all feel the energy and buzz in our offices ’, the memo explained. Yahoo’s defenders say that their staff are highly skilled people, such as designers and programmers, who needed more face time with colleagues. Quite simply, for Yahoo, the costs of telecommuting were greater than its benefits. And there are some widely accepted disad- vantages of telecommuting. Working from home can be isolating, for staff and for managers who will need to put effort into keeping in touch. In fact, telecommuting can be difficult when employees require constant super- vision. There is also the question of accountability. It is difficult to judge whether staff really are working rather than watching daytime TV. Nevertheless a blanket ban on working from home is still unusual in hi-tech indus- tries. And within a year of Yahoo’s original decision, there were some indications that, under certain circumstances, telecommuting was once more being permitted.

Yahoo clamps down on telecommuting 12

✽ ✽ ✽ Operations principle Operations principle Operations principle

So u

rc e:

S h

u tt

er st

o ck

.c o

m : J

H D

T P

ro d

u ct

io n

s

M09_SLAC8678_08_SE_C09.indd 296 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 297

Working temperature Predicting the reactions of individuals to working temperature is not straightforward. Individuals vary in the way their performance and comfort vary with temperature. Furthermore, most of us judging ‘temperature’ will also be influenced by other factors such as humidity and air movement. Nevertheless, some general points regarding working tempera- tures provide guidance to job designers:

● Comfortable temperature range will depend on the type of work being carried out, lighter work requiring higher temperatures than heavier work.

● The effectiveness of people at performing vigilance tasks reduces at temperatures above about 29°C; the equivalent temperature for people performing light manual tasks is a little lower.

● The chances of accidents occurring increase at temperatures which are above or below the comfortable range for the work involved.

Illumination levels The intensity of lighting required to perform any job satisfactorily will depend on the nature of the job. Some jobs which involve extremely delicate and precise movement, surgery for example, require very high levels of illumination. Other, less delicate jobs do not require such high levels. Table 9.3 shows the recommended illumination levels (measured in lux) for a range of activities.

Noise levels The damaging effects of excessive noise levels are perhaps easier to understand than some other environmental factors. Noise-induced hearing loss is a well-documented consequence of working environments where noise is not kept below safe limits. The noise levels of various activities are shown in Table 9.4. When reading this list, bear in mind that the recommended (and often legal) maximum noise level to which people can be subjected over the working day is 90 decibels (dB) in the UK (although in some parts of the world the legal level is lower than this). Also bear in mind that the decibel unit of noise is based on a logarithmic scale, which means that noise intensity doubles about every 3 dB. In addition to the damaging effects of high levels of noise, intermittent and high-frequency noise can also affect work performance at far lower levels, especially on tasks requiring attention and judgement.13

Table 9.3 Examples of recommended lighting levels for various activities14

Activity illuminance (lx)

Normal activities in the home, general lighting 50

Furnace rooms in glass factory 150

General office work 500

Motor vehicle assembly 500

Proofreading 750

Colour matching in paint factory 1,000

Electronic assembly 1,000

Close inspection of knitwear 1,500

Engineering testing inspection using small instruments 3,000

Watchmaking and fine jewellery manufacture 3,000

Surgery, local lighting 10,000–50,000

M09_SLAC8678_08_SE_C09.indd 297 6/2/16 1:32 PM

298 PART TWO DESIGNING THE OPERATION

Table 9.4 Noise levels for various activities

Noise Decibels (dB)

Quiet speech 40

Light traffi c at 25 metres 50

Large busy offi ce 60

Busy street, heavy traffi c 70

Pneumatic drill at 20 metres 80

Textile factory 90

Circular saw – close work 100

Riveting machine – close work 110

Jet aircraft taking off at 100 metres 120

OPERATIONS IN PRACTICE

Background music at work is not new. It has been used in the workplace for centuries. As far back as the Industrial Revolution orchestras and singers would be hired occa- sionally to perform for workers in the quieter factories. Later, in the 1940s, the BBC launched a radio programme called Music While You Work . Broadcasting twice a day, it was made especially for factory workers. Artists who were booked for the show were told to ‘ play material with an upbeat rhythm that would keep the workers’ attention ', in the belief that it would improve productivity. But playing music at work is not always free. In the UK, for example, the law requires businesses that play any recorded music in public to get licences from the Performing Right Society (PRS), which collects fees and pays royalties to composers and their publishers. Listening to a device through head- phones, however, is free. But does music help or hinder?

Some bodies definitely think that it helps. Musicworks (which is an organization supported by the PRS, so it is not exactly independent) cites studies that show that music in the workplace promotes positive mood, can build team spirit, improves alertness and can reduce the number of workplace accidents. It can also, they say, cut the num- ber of sick days and increase workplace productivity. One study by Teresa Lesiuk at the University of Miami found that IT specialists who listened to music completed tasks more quickly and came up with better ideas than those who did not. But not everyone is convinced. ‘ If people need a high level of concentration, it could be a distraction ’, says Dr Carolyn Axtell, at the Institute of Work Psychology. ‘ When people choose to listen there can be positive effects – it can be relaxing and help manage other distractions such

as noise. But when it's imposed, they can find it annoying and stressful .’ However, individuals can differ in their reac- tion to music and problems occur when colleagues clash. ‘ You can look away if you don't want to see something, but you can't close your ears ’, she says.

In another study researchers at London University studied the apparently common practice of surgeons playing music in the operating theatre (playlists ranged from gentle classical music, through heavy metal, to elec- tronic dance music). Patients did not complain, being anaesthetized, but other members of the surgical team were not always happy. Music could damage commu- nication in a surgical team, preventing team members from hearing instructions. Even worse, when sound lev- els are uneven and a new track blasts out unexpectedly, or when a surgeon turns up the volume when his or her favourite song comes on, other team members can be

Music while you work? 15

So u

rc e:

S h

u tt

er st

o ck

.c o

m : O

lly y

M09_SLAC8678_08_SE_C09.indd 298 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 299

Ergonomics in the office As the number of people working in offices (or office-like workplaces) has increased, ergonomic principles have been applied increasingly to this type of work. At the same time, legislation has been moving to cover office technology such as computer screens and keyboards. For example, European Union directives on working with display screen equipment require organizations to assess all workstations to reduce the risks inherent in their use, plan work times for breaks and changes in activity, and provide information and training for users. Figure 9.9 illustrates some of the ergonomic factors which should be taken into account when designing office jobs.

disturbed. But notwithstanding the sometimes conflict- ing findings from researchers, some themes do emerge:

● How ‘immersive’ a task is makes a difference when evaluating music's effectiveness in increasing pro- ductive output. ‘Immersive’ refers to the variability and creative demand of the task. Creating an entirely original piece of work from scratch that demands a lot of creativity is ‘immersive’. Performing more rou- tine tasks such as answering emails is not. When the task is routine, clearly defined and repetitive, music is probably useful for most people.

● Music affects your mood. Apparently, it is not the background noise of the music itself, but rather the improved mood that your favourite music creates that is the reason for the increase in productivity. In one study, IT specialists who listened to music

completed their tasks more quickly and came up with better ideas than those who did not, because the music improved their mood.

● In open-plan offices where background chatter can be too much for some people to handle, headphones can help some people.

● Music does not help learning. It has a negative effect on absorbing and retaining new information, because it demands too much of your attention.

● Listening to music with lyrics, especially interesting and/or new lyrics, detracts from performing immer- sive tasks. Listening to lyrics activates the language centre of your brain, so trying to perform other language-related tasks is particularly difficult.

(Full disclosure: most of this book was written while listening to music.)

Figure 9.9 Ergonomics in the office environment

M09_SLAC8678_08_SE_C09.indd 299 6/2/16 1:32 PM

300 PART TWO DESIGNING THE OPERATION

HOW ARE WORK TIMES ALLOCATED?

Without some estimate of how long it takes to complete an activity, it will not be possible to know how much work to allocate to teams or individuals, to know when a task will be com- pleted, to know how much it costs, to know if work is progressing according to schedule, and many other vital pieces of information that are needed to manage any operation. Without some estimate of work times, operations managers are ‘flying blind’. At the same time it does not need much thought before it becomes clear that measuring work times must be difficult to do with any degree of accuracy, or confidence. The time you take to do any task will depend on how skilled you are at the task, how much experience you have, how energetic or moti- vated you are, whether you have the appropriate tools, what the environmental conditions are, how tired you are, and so on. So, at best, any ‘measurement’ of how long a task will, or should, take will be an estimate. It will be our ‘best guess’ of how much time to allow for the task. That is why we call this process of estimating work times ‘work time allocation’. We are allocating a time for completing a task because we need to do so for many important opera- tions management decisions. For example, work times are needed for:

● planning how much work a process can perform (its capacity); ● deciding how many staff are needed to complete tasks; ● scheduling individual tasks to specific people; ● balancing work allocation in processes (see Chapter 7); ● costing the labour content of a product or service; ● estimating the efficiency or productivity of staff and/or processes; and ● calculating bonus payments (less important than it was at one time).

Notwithstanding the weak theoretical basis of work measurement, understanding the rela- tionship between work and time is clearly an important part of job design. The advantage of structured and systematic work measurement is that it gives a common currency for the evaluation and comparison of all types of work. So, if work time allocation is important, how should it be done? In fact, there is a long-standing body of knowledge and experience in this area. This is generally referred to as ‘work measurement’, although, as we have said, ‘meas- urement’ could be regarded as indicating a somewhat spurious degree of accuracy. Formally, work measurement is defined as ‘the process of establishing the time for a qualified worker, at a defined level of performance, to carry out a specified job’. Although not a precise definition, generally it is agreed that a specified job is one for which specifications have been established to define most aspects of the job. A qualified worker is ‘one who is accepted as having the necessary physical attributes, intelligence, skill, education and knowledge to perform the task to satisfactory standards of safety, quality and quantity’. Standard performance is ‘The rate of output which qualified workers will achieve without over-exertion as an average over the working day provided they are motivated to apply themselves to their work.’

The techniques of work measurement At one time, work measurement was firmly associated with an image of the ‘efficiency expert’, ‘time and motion’ man, or ‘rate fixer’, who wandered around factories with a stopwatch, look- ing to save a few cents or pennies. And although that idea of work measurement has (almost) died out, the use of a stopwatch to establish a basic time for a job is still relevant, and used in a technique called ‘time study’. Time study and the general topic of work measurement are treated in the supplement to this chapter.

As well as time study, there are other work measurement techniques in use. They include the following:

● Synthesis from elemental data – is a work measurement technique for building up the time for a job at a defined level of performance by totalling element times obtained previously from the studies in other jobs containing the elements concerned or from synthetic data.

M09_SLAC8678_08_SE_C09.indd 300 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 301

● Predetermined motion–time systems (PMTS) – is a work measurement technique whereby times established for basic human motions (classified according to the nature of the motion and the conditions under which it is made) are used to build up the time for a job at a defined level of performance.

● Analytical estimating – is a work measurement technique which is a development of esti- mating whereby the time required to carry out the elements of a job at a defined level of performance is estimated from knowledge and experience of the elements concerned.

● Activity sampling – is a technique in which a large number of instantaneous observations are made over a period of time of a group of machines, processes or workers. Each observa- tion records what is happening at that instant and the percentage of observations recorded for a particular activity or delay is a measure of the percentage of time during which that activity or delay occurs.

Critical commentary

The criticisms aimed at work measurement are many and various. Among the most common are the following:

● All the ideas on which the concept of a standard time is based are impossible to defi ne precisely. How can one possibly give clarity to the defi nition of qualifi ed workers, or spec- ifi ed jobs, or especially a defi ned level of performance?

● Even if one attempts to follow these defi nitions, all that results is an excessively rigid job defi nition. Most modern jobs require some element of fl exibility, which is diffi cult to achieve alongside rigidly defi ned jobs.

● Using stopwatches to time human beings is both degrading and usually counter-productive. At best it is intrusive, at worst it makes people ‘objects for study ’.

● The rating procedure implicit in time study is subjective and usually arbitrary. It has no basis other than the opinion of the person carrying out the study.

● Time study, especially, is very easy to manipulate. It is possible for employers to ‘work back’ from a time which is ‘required’ to achieve a particular cost. Also, experienced staff can ‘put on an act’ to fool the person recording the times.

● Human resources are any organization’s, and therefore any operation’s, greatest asset. Often, most ‘human resources’ are to be found in the operations function.

❯ Why are people so important in operations management?

SUMMARY ANSWERS TO KEY QUESTIONS

● Human resource strategy is the overall long-term approach to ensuring that an organiza- tion’s human resources provide a strategic advantage. It involves identifying the number and type of people that are needed to manage, run and develop the organization so that it meets its strategic business objectives, and putting in place the programmes and initiatives that attract, develop and retain appropriate staff .

❯ How do operations managers contribute to human resource strategy?

M09_SLAC8678_08_SE_C09.indd 301 6/2/16 1:32 PM

302 PART TWO DESIGNING THE OPERATION

Grace Whelan, Managing Partner of McPherson Charles, was puzzled. Three of her most successful teams seemed to be facing similar problems with their staff, even though each team had very different tasks, processes and types of staff. Every year the firm surveyed its entire staff in order to gauge their views, levels of satisfaction with their jobs and development needs. It was the results from the latest survey that surprised Grace. ‘ The results of the survey are really unanticipated. Only last year everything seemed fine. Now staff morale has evidently slumped in all three teams. Yet the partners who lead all of these teams are first class. Outstanding lawyers and good leaders. ’

McPherson Charles, based in Bristol in the West of England, had grown rapidly to be one of the biggest law firms in the region, with 21 partners and around 400 staff. Three years previously the firm had reorganized into 15 teams each headed by a ‘lead partner ’ and specializing in practising one type of law. It had proved to be a good organ- izational structure, which encouraged teams to organize themselves appropriately for the type of clients that they dealt with. In particular three teams had flourished under

this structure: ‘family law ’, ‘property ’ and ‘litigation’. Now it was these very teams whose staff were showing signs of dissatisfaction.

Before the results of the survey were published to all staff, Grace knew that she would need to have worked out some kind of response to the issues raised. She decided to go and see each of the lead partners in the three teams. The first

● One can take various perspectives on organizations. How we illustrate organizations says much about our underlying assumptions of what an ‘organization’ is. For example, organi- zations can be described as machines, organisms, brains, cultures or political systems.

● The relationship between the ‘staff ’ and ‘line’ roles in operation can be modelled using the four perspective on operations strategy that were discussed in Chapter 3 .

● There are an almost infi nite number of possible organizational structures. Most are blends of two or more ‘pure types’, such as the U-form, the M-form, matrix forms, the N-form.

❯ How can the operations function be organized?

● There are many infl uences on how jobs are designed. These include the division of labour, scientifi c management, method study, work measurement, ergonomics, behavioural ap- proaches, including job rotation, job enlargement and job enrichment, empowerment, team working, and fl exible working (including ‘telecommuting ’).

❯ How do we go about designing jobs?

● The best known method is time study, but there are other work measurement techniques including synthesis from elemental data, predetermined motion-time systems (PMTS), an- alytical estimating and activity sampling.

❯ How are work times allocated?

CASE STUDY Grace faces (three) problems

So u

rc e:

S h

u tt

er st

o ck

.c o

m : W

av eb

re ak

m ed

ia

M09_SLAC8678_08_SE_C09.indd 302 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 303

person she decided to talk to was Simon Reece, who led the family law team. Before doing so she explained what his team did.

Family law ‘They are called the “family law” team but basically what they do is to help people through the trauma of divorce, separation and break up. Their biggest “high value” clients come to them because of word of mouth recommendation. Last year they had almost a hundred of these “high value” clients and they all valued the personal touch that they were able to give them, getting to know them well and spending time with them to understand the, often “hidden” aspects of their case. Of course, not all their clients are the super-rich. About a third of the annual family law income comes from about 750 relatively routine divorce and coun- seling cases.’

Simon was blunt about the declining levels of staff sat- isfaction in his team. ‘The problem is that working with the “high value” clients is just more fun and more rewarding than the routine “bread and butter” work. So my people who do that kind of work, usually the more experienced ones, don’t want to take on the routine stuff. With “high value” cases you have to be able to untangle the personal issues from the busi- ness ones. Interviewing these clients cannot be rushed. They tend to be wealthy people with complex assets. We will often have to drop everything and go off half way round the world to meet and discuss their situation. There are no standard procedures, every client is different, and everyone has to be treated as an individual. So we have a team of individuals who rise to the challenge each time and give great service. By contrast, the routine work is a lot less interesting, yet some- times very harrowing. The more junior staff who tend to take on the routine cases can sometimes feel themselves to be “second-class citizens”. Many of them would like to get more experience with the complex high value work, but I can’t take the risk of giving them that degree of responsibility, the work is too valuable. Also, frankly, the senior people who deal with the high value work don’t want to give up their more glam- orous work. I have been trying to make sure that everyone in my team who wants to has a mix of interesting and rou- tine work over the year. It’s the only way to develop them in the long term. You have to encourage them to exercise and develop their professional judgement. They are empowered to deal with any issues themselves or call on one of the more senior members of the team for advice if appropriate. It is important to give this kind of responsibility to them so that they see themselves as part of a team. But there are still ten- sions between senior and more junior staff. We are thinking about adopting an open-plan office arrangement centred around our specialist library of family case law, to try and encourage more cooperation.’

Litigation Grace was less concerned about the litigation team, led by Hazel Lewis. ‘The litigation team has been our best success

story. The have grown far faster than any other part of the firm, and a lot of that is down to Hazel. She provides a key service for our commercial client base. Their primary work consists of handling bulk collections of debt. The group has 17 clients of which 5 provide 85% of total volume. They work closely with the accounts departments of the client companies and have developed a semi-automatic approach to debt col- lection. It’s a great service that Hazel has largely automated.’

Hazel had led the litigation team since it had been set up four years ago. As well as being the partner in charge of litigation, unusually she and her assistant were the only qualified lawyers in the team. ‘Our problems in the litigation team are not really because of any internal ten- sions or disputes. Broadly, our people are happy with what they do and how they are supervised. The issue is just that we are so different from the rest of the firm. Apart from myself and Raymond [her assistant] everyone else in the team are either technicians who look after and develop the systems that we use, or people who have worked in process- ing or call centres, before they came to us. And between us we have developed a smart operation here. Our staff input data received from their clients into the system, from that point everything progresses through a pre-defined process, letters are produced, queries responded to and eventu- ally debts collected, ultimately through court proceedings if necessary. Work tends to come in batches from clients and varies according to the time of year and client sales activities. At the moment things are fairly steady; we had almost 900 new cases to deal with last week. The details of each case are sent over by the client; our people input the data onto our screens and set up a standard diary system for sending letters out. Some people respond quickly to the first letter and often the case is closed within a week or so, other people ignore letters and eventually we initiate court proceedings. We know exactly what is required for court dealings and have a pretty good process to make sure all the right documentation is available on the day. Our problem is that the rest of the firm does not see us as being “proper lawyers”, and they are right, we’re not. But it does get difficult for our people, being looked down upon all the time. Our salary structure is different, our bonus scheme is different, and how we measure performance is differ- ent. But there is a solution. Because we have expanded so much, we need more space than is available in this build- ing. I think that we should think about moving the litiga- tion team. There is a great location out by the airport that could be expanded in the future if needed. There is really no reason for us to be located with the other teams.’

Property The ‘property ’ team was one of the largest parts of the firm and was well established in the local market with an excellent reputation for being fast, friendly and giving value for money. Most of its work was ‘domestic’, acting for individuals buying or selling their home, or their sec- ond home. Each client was allocated to a solicitor who

M09_SLAC8678_08_SE_C09.indd 303 6/2/16 1:32 PM

304 PART TWO DESIGNING THE OPERATION

becomes his or her main point of contact. But, given that they can have up to a hundred domestic clients a week, most of the work was actually carried out by the rest of the team of ‘paralegal’ staff (staff with qualifications less than a fully qualified lawyer) behind the scenes.

Kate Hutchinson, who led the property team, was proud of the process she and her team had set up. ‘ There is a rel- atively standard process to domestic property sales and pur- chases and we think that we are pretty efficient at managing these standard jobs. Our process has four stages, one dealing with land registry searches, one liaising with banks who are providing the mortgage finance, one to make sure surveys are completed and one section that finalises the whole process to completion. We believe that this degree of specialisation can help us achieve the efficiencies that are becoming important, as the market gets more competitive. Our particular prob- lem is that increasingly we are also getting more complex “special” jobs. These are things like “volume re-mortgage” arrangements and rather complex “one-off ” jobs, where a mortgage lender transfers a complex set of loan assets to another lender. These “special” jobs are always more complex than the domestic work and they are not popular with our staff. They don't always fit easily into our standard process, and they disrupt the routine of working. For example, some- times there are occasions when fast completion is particularly important and that can throw us a bit. ’

Grace was more worried about the property team than Kate appeared to be. The firm had recently formed partnerships with two large speculative builders, which dealt in special ‘plot sales’ that would also be classed as non- standard ‘specials’ by Kate. Grace knew that all these ‘specials’ did involve a lot of work and could occupy several members of the team for a time. But they were an important source of revenue. Currently the team was dealing with up to 25 ‘specials’ each week, and this would certainly increase. Grace suspected that Kate was mistaken to try and follow the same process with them as the normal domestic jobs. Maybe trying to do differ- ent things on the same process was the cause of the dis- satisfaction in the team?

QUESTIONS 1 What are the problems amongt the staff of each of the

three teams?

2 What are the individual ‘services’ offered by each of the three teams?

3 How would you describe each team’s process in terms of jobs of its staff?

4 What do you think each team leader should be doing to try and overcome their teams' problems?

PROBLEMS AND APPLICATIONS

1 A hotel has two wings, an east wing and a west wing. Each wing has four ‘room service maids’ working seven-hour shifts to service the rooms each day. The east wing has 40 standard rooms, 12 deluxe rooms and 5 suites. The west wing has 50 standard rooms and 10 deluxe rooms. The standard times for servicing rooms are as follows: standard rooms 20 standard minutes, deluxe rooms 25 standard minutes, and suites 40 standard minutes. In addition, an allowance of 5 standard minutes per room is given for any miscellaneous jobs such as collecting extra items for the room or dealing with customer requests. What is the productivity of the maids in each wing of the hotel? What other factors might also influence the productivity of the maids?

2 In the problem above, one of the maids in the west wing wants to job-share with her partner, each working three hours per day. Her colleagues have agreed to support her and will guar- antee to service all the rooms in the west wing to the same standard each day. If they succeed in doing this, how has it affected their productivity?

3 Step 1 – Make a sandwich. Any type of sandwich, preferably one that you enjoy, and docu- ment the tasks you have to perform in order to complete the job. Make sure you include all the activities including the movement of materials (bread etc.) to and from the work surface. Step 2 – So impressed were your friends with the general appearance of your sandwich that they have persuaded you to make one each for them every day. You have 10 friends so every morning you must make 10 identical sandwiches (to stop squabbling). How would you change the method by which you make the sandwiches to accommodate this higher volume?

M09_SLAC8678_08_SE_C09.indd 304 6/2/16 1:32 PM

CHAPTER 9 PEOPLE IN OPERATIONS 305

Step 3 – The fame of your sandwiches had spread. You now decide to start a business making several different types of sandwich in high volume. Design the jobs of the two or three people who will help you in this venture. Assume that volumes run into at least 100 of three types of sandwich every day.

4 A little-known department of your local government authority has the responsibility for keeping the area’s public lavatories clean. It employs 10 people who each have a number of public lavatories that they visit, clean and report any necessary repairs every day. Draw up a list of ideas for how you would keep this fine body of people motivated and committed to performing this unpleasant task.

5 Visit a supermarket and observe the people who staff the checkouts. (a) What kind of skills do people who do this job need to have? (b) How many customers per hour are they capable of ‘processing ’? (c) What opportunities exist for job enrichment in this activity? (d) How would you ensure motivation and commitment among the staff who do this job?

SELECTED FURTHER READING

Argyris, C. (1998) Empowerment: the emperor ’s new clothes, Harvard Business Review, May–June.

A critical but fascinating view of empowerment.

Bock, L. (2015) Work Rules! Insights from Inside Google That Will Transform How You Live and Lead, John Murray, London.

With an agenda far wider than this chapter, it is nevertheless an absorbing book that gives an insight into an absorbing firm.

Bond, F.W. and Bunce, D. (2001) Job control mediates change in a work reorganization interven- tion for stress reduction, Journal of Occupational Health Psychology, vol. 6, 290–302.

An academic paper that is interesting on stress issues.

Bridger, R. (2003) Introduction to Ergonomics, Taylor & Francis, London.

Exactly what it says in the title, an introduction (but a good one) to ergonomics.

Dul, J. and Weerdmeester, B. (2008) Ergonomics for Beginners: A Quick Reference Guide, 3rd edn, CRC Press, Boca Raton, FL.

Good, practical guidance on the removal from the workplace of physical and mental stresses caused by poor job or environmental design.

Hackman, R. J. and Oldham, G. (1980) Work Redesign, Addison-Wesley, Reading, MA.

Somewhat dated but, in its time, ground breaking and certainly hugely influential.

Herzberg, F. (1987) One more time: how do you motivate employees? (with retrospective com- mentary), Harvard Business Review, January.

An interesting look back by one of the most influential figures in the behavioural approach to job design school.

Lantz, A. and Brav, A. (2007) Job design for learning in work groups, Journal of Workplace Learning, vol. 19, issue 5, 269–285.

Another academic paper, but one that addresses the important issue of learning as a job design objective.

M09_SLAC8678_08_SE_C09.indd 305 6/2/16 1:32 PM

306 PART TWO DESIGNING THE OPERATION

INTRODUCTION A tale is told of Frank Gilbreth (the founder of method study) addressing a scientific conference with a paper entitled ‘The best way to get dressed in a morning’. In his presentation, he rather bemused the scientific audience by analysing the ‘best’ way of buttoning up one’s waistcoat in the morning. Among his conclusions was that waistcoats should always be buttoned from the bottom upwards. (To make it easier to straighten a tie in the same motion; buttoning from the top downwards requires the hands to be raised again). Think of this example if you want to understand scientific management and method study in particular. First of all, he is quite right. Method study and the other techniques of scientific management may often be without any intellectual or scientific validation, but by and large they work in their own terms. Second, Gilbreth reached his conclusion by a systematic and critical analysis of what motions were necessary to do the job. Again, these are characteristics of scientific management – detailed analysis and painstakingly systematic examination. Third (and possibly most important), the results are relatively trivial. A great deal of effort was put into reaching a conclusion that was unlikely to have any earth-shattering consequences. Indeed, one of the criticisms of scientific management, as developed in the early part of the twentieth century, is that it concentrated on relatively limited, and sometimes trivial, objectives.

The responsibility for its application, however, has moved away from specialist ‘time and motion’ staff to the employees who can use such principles to improve what they do and how they do it. Further, some of the methods and techniques of scientific man- agement, as opposed to its philosophy (especially those which come under the general heading of ‘method study’), can in practice prove useful in critically re-examining job designs. It is the practicality of these techniques which possibly explains why they are still influential in job design almost a century after their inception.

METHOD STUDY IN JOB DESIGN

Method study is a systematic approach to finding the best method. There are six steps:

1 Select the work to be studied. 2 Record all the relevant facts of the present method. 3 Examine those facts critically and in sequence. 4 Develop the most practical, economic and effective method. 5 Install the new method. 6 Maintain the method by periodically checking it in use.

Step 1 – Selecting the work to be studied Most operations have many hundreds and possibly thousands of discrete jobs and activities which could be subjected to study. The first stage in method study is to select those jobs to be studied which will give the most return on the investment of the time spent studying them. This means it is unlikely that it will be worth studying activities which, for example, may soon be discontinued or are only performed occasionally. On the other hand, the types of jobs which

Supplement to Chapter 9 Work study

M09_SLAC8678_08_SE_C09.indd 306 6/2/16 1:32 PM

SUPPLEMENT TO CHAPTER 9 WORK STUDY 307

should be studied as a matter of priority are those which, for example, seem to offer the greatest scope for improvement, or which are causing bottlenecks, delays or problems in the operation.

Step 2 – Recording the present method There are many different recording techniques used in method study. Most of them:

● record the sequence of activities in the job; ● record the time interrelationship of the activities in the job; or ● record the path of movement of some part of the job.

Perhaps the most commonly used recording technique in method study is process mapping, which was discussed in Chapter 4. Note that we are here recording the present method of doing the job. It may seem strange to devote so much time and effort to recording what is currently happening when, after all, the objective of method study is to devise a better method. The rationale for this is, first of all, that recording the present method can give a far greater insight into the job itself, and this can lead to new ways of doing it. Second, recording the present meth- od is a good starting point from which to evaluate it critically and therefore improve it. In this last point the assumption is that it is easier to improve the method by starting from the current method and then criticizing it in detail than by starting with a ‘blank sheet of paper’.

Step 3 – Examining the facts This is probably the most important stage in method study and the idea here is to examine the current method thoroughly and critically. This is often done by using the so-called ‘questioning technique’. This technique attempts to detect weaknesses in the rationale for existing methods so that alternative methods can be developed (see Table S9.1).

Table S9.1 The method study questioning technique

Broad question Detailed question

The purpose of each activity (questions the fundamental need for the element)

What is done?

Why is it done?

What else could be done?

What should be done?

The place in which each element is done (may suggest a combination of certain activities or operations)

Where is it done?

Why is it done there?

Where else could it be done?

Where should it be done?

The sequence in which the elements are done (may suggest a change in the sequence of the activity)

When is it done?

Why is it done then?

When should it be done?

The person who does the activity (may suggest a combination and/or change in responsibility or sequence)

Who does it?

Why does that person do it?

Who else could do it?

Who should do it?

The means by which each activity is done (may suggest new methods)

How is it done?

Why is it done in that way?

How else could it be done?

How should it be done?

M09_SLAC8678_08_SE_C09.indd 307 6/2/16 1:32 PM

308 PART TWO DESIGNING THE OPERATION

The approach may appear somewhat detailed and tedious, yet it is fundamental to the method study philosophy – everything must be critically examined. Understanding the natural tendency to be less than rigorous at this stage, some organizations use pro forma questionnaires, asking each of these questions and leaving space for formal replies and/ or justifications, which the job designer is required to complete.

Step 4 – Developing a new method The previous critical examination of current methods has by this stage probably indicat- ed some changes and improvements. This step involves taking these ideas further in an attempt to:

● eliminate parts of the activity altogether; ● combine elements together; ● change the sequence of events so as to improve the efficiency of the job; or ● simplify the activity to reduce the work content.

A useful aid during this process is a checklist such as the revised principles of motion econ- omy. Table S9.2 illustrates these.

Steps 5 and 6 – Installing the new method and regularly maintaining it The method study approach to the installation of new work practices concentrates largely on ‘project managing’ the installation process. It also emphasizes the need to monitor regularly the effectiveness of job designs after they have been installed.

Table S9.2 The principles of motion economy

Broad principle How to do it

Use the human body the way it works best

Work should be arranged so that a natural rhythm can become automatic

Motion of the body should be simultaneous and symmetrical if possible

The full capabilities of the human body should be employed

Arms and hands as weights are subject to the physical laws and energy should be conserved

Tasks should be simplified

Arrange the workplace to assist performance

There should be a defined place for all equipment and materials

Equipment, materials and controls should be located close to the point of use

Equipment, materials and controls should be located to permit the best sequence and path of motions

The workplace should be fitted both to the tasks and to human capabilities

Use technology to reduce human effort

Work should be presented precisely where needed

Guides should assist in positioning the work without close operator attention

Controls and foot-operated devices can relieve the hands of work

Mechanical devices can multiply human abilities

Mechanical systems should be fitted to human use

Source: Adapted from Barnes, Frank C. (1983) ‘Principles of Motion Economy: Revisited, Reviewed, and Restored’, Proceedings of the Southern Management Association Annual Meeting (Atlanta, GA 1983), p. 298.

M09_SLAC8678_08_SE_C09.indd 308 6/2/16 1:32 PM

WORK MEASUREMENT IN JOB DESIGN

Basic times Terminology is important in work measurement. When a qualified worker is working on a specified job at standard performance, the time he or she takes to perform the job is called the basic time for the job. Basic times are useful because they are the ‘building blocks’ of time estimation. With the basic times for a range of different tasks, an operations manager can construct a time estimate for any longer activity which is made up of the tasks. The best- known technique for establishing basic times is probably time study.

Time study Time study is ‘a work measurement technique for recording the times and rate of working for the elements of a specified job, carried out under specified conditions, and for analysing the data so as to obtain the time necessary for the carrying out of the job at a defined level of performance’. The technique takes three steps to derive the basic times for the elements of the job:

● observing and measuring the time taken to perform each element of the job; ● adjusting, or ‘normalizing’, each observed time; ● averaging the adjusted times to derive the basic time for the element.

Step 1 – Observing, measuring and rating A job is observed through several cycles. Each time an element is performed, it is timed using a stopwatch. Simultaneously with the observation of time, a rating of the perceived performance of the person doing the job is recorded. Rating is ‘the process of assessing the worker’s rate of working relative to the observer’s concept of the rate corresponding to standard performance. The observer may take into account, separately or in combina- tion, one or more factors necessary to carrying out the job, such as speed of movement, effort, dexterity, consistency, etc.’ There are several ways of recording the observer’s rating. The most common is on a scale which uses a rating of 100 to represent standard performance. If an observer rates a particular observation of the time to perform an element at 100, the time observed is the actual time which anyone working at standard performance would take.

Step 2 – Adjusting the observed times The adjustment to normalize the observed time is:

Observed rating Standard rating

where standard rating is 100 on the common rating scale we are using here. For example, if the observed time is 0.71 minutes and the observed rating is 90, then:

Basic time = 0.71 * 9 = 0.64 min 100

Step 3 – Averaging the basic times In spite of the adjustments made to the observed times through the rating mechanism, each separately calculated basic time will not be the same. This is not necessarily a function of inac- curate rating, or even the vagueness of the rating procedure itself; it is a natural phenomenon of the time taken to perform tasks. Any human activity cannot be repeated in exactly the same time on every occasion.

SUPPLEMENT TO CHAPTER 9 WORK STUDY 309

M09_SLAC8678_08_SE_C09.indd 309 6/2/16 1:32 PM

310 PART TWO DESIGNING THE OPERATION

Standard times The standard time for a job is an extension of the basic time and has a different use. Whereas the basic time for a job is a piece of information which can be used as the first step in estimating the time to perform a job under a wide range of conditions, standard time refers to the time allowed for the job under specific circumstances. This is because standard time includes allowances which reflect the rest and relaxation allowed because of the conditions under which the job is performed. So the standard time for each element consists principally of two parts, the basic time (the time taken by a qualified worker, doing a specified job at standard performance) and an allowance (this is added to the basic time to allow for rest, relaxation and personal needs).

Allowances Allowances are additions to the basic time intended to provide the worker with the opportu- nity to recover from the physiological and psychological effects of carrying out specified work under specified conditions and to allow for personal needs. The amount of the allowance will depend on the nature of the job. The way in which relaxation allowance is calculated, and the exact allowances given for each of the factors which determine the extent of the allowance, varies between different organizations. Table S9.3 illustrates the allowance table used by

Table S9.3 An allowances table used by a domestic appliance manufacturer

Allowance factors Example Allowance (%)

Energy needed Negligible None 0 Very light 0–3 kg 3 Light 3–10 kg 5 Medium 10–20 kg 10 Heavy 20–30 kg 15 Very heavy Above 30 kg 15–30

Posture required Normal Sitting 0 Erect Standing 2 Continuously erect Standing for long periods 3 Lying On side, face or back 4 Difficult Crouching, etc. 4–10

Visual fatigue Nearly continuous attention 2 Continuous attention with varying focus 3 Continuous attention with fixed focus 5

Temperature Very low Below 0°C Over 10 Low 0–12°C 0–10 Normal 12–23°C 0 High 23–30°C 0–10 Very high Above 30°C Over 10

Atmospheric conditions Good Well ventilated 0 Fair Stuffy/smelly 2 Poor Dusty/needs filter 2–7 Bad Needs respirator 7–12

M09_SLAC8678_08_SE_C09.indd 310 6/2/16 1:32 PM

one company which manufactures domestic appliances. Every job has an allowance of 10 per cent; the table shows the further percentage allowances to be applied to each element of the job. In addition, other allowances may be applied for such things as unexpected contingen- cies, synchronization with other jobs, unusual working conditions, and so on.

Figure S9.1 shows how average basic times for each element in the job are combined with allowances (low in this example) for each element to build up the standard time for the whole job.

Figure S9.1 Time study of a packing task – standard time for the whole task calculated

Worked example

Two work teams in the Monrovian Embassy have been allocated the task of processing visa applications. Team A processes applications from Europe, Africa and the Middle East. Team B processes applications from North and South America, Asia and Australasia. Team A has cho- sen to organize itself in such a way that each of its three team members processes an appli- cation from start to finish. The four members of Team B have chosen to split themselves into two sub-teams. Two open the letters and carry out the checks for a criminal record (no one who has been convicted of any crime other than a motoring offence can enter Monrovia), while the other two team members check for financial security (only people with more than Monrovian $1,000 may enter the country). The head of consular affairs is keen to find out if one of these methods of organizing the teams is more efficient than the other. The problem

SUPPLEMENT TO CHAPTER 9 WORK STUDY 311

▼

M09_SLAC8678_08_SE_C09.indd 311 6/2/16 1:32 PM

312 PART TWO DESIGNING THE OPERATION

is that the mix of applications differs region by region. Team A typically processes around two business applications to every one tourist application. Team B processes around one business application to every two tourist applications.

A study revealed the following data:

Average standard time to process a business visa = 63 standard minutes

Average time to process a tourist visa = 55 standard minutes

Average weekly output from Team A is: 85.2 Business visas 39.5 Tourist visas

Average weekly output from Team B is: 53.5 Business visas 100.7 Tourist visas

All team members work a 40-hour week.

The efficiency of each team can be calculated by comparing the actual output in standard minutes and the time worked in minutes. So Team A processes:

(85.2 * 63) + (39.5 * 55) = 7,540.1 standard minutes of work

in 3 * 40 * 60 minutes = 7,200 minutes

So its effi ciency = 7, 540.1 * 100 = 104.72% 7, 200

Team B processes:

(53.5 * 63) + (100.7 * 55) = 8,909 standard minutes of work

in 4 * 40 * 60 minutes = 9,600 minutes

So its effi ciency = 8, 909 * 100 = 92.8% 9, 600

The initial evidence therefore seems to suggest that the way Team A has organized itself is more efficient.

M09_SLAC8678_08_SE_C09.indd 312 6/2/16 1:32 PM

M09_SLAC8678_08_SE_C09.indd 313 6/2/16 1:32 PM

This page intentionally left blank

10 Planning and control

14 Planning and control systems

12 Supply chain management

11 Capacity management

15 Lean operations

13 Inventory management

M10_SLAC8678_08_SE_C10.indd 314 6/2/16 2:22 PM

All the activities involved in the design of an operation should have provided the nature and shape of the transforming resources that are capable of satisfying customers’ demands. Products and services then have to be created and delivered to customers. This is done by planning and controlling the activities of the transforming resources on a day-to-day basis to ensure the appropriate supply of products and services to meet the requirements of the market. This part of the book will look at six different aspects of planning and controlling the delivery of products and services as they flow through processes, operations and supply networks. The chapters in this part are:

● Chapter 10 Planning and control – This examines how operations organize the delivery of their products and services on an ongoing basis so that customers’ demands are satisfied.

● Chapter 11 Capacity management – This explains how operations need to decide how to vary their capacity (if at all) as demand for their products and services fluctuates.

Part Three DELIVER

Operations management

Direct

Design Develop

Deliver

Deliver

Planning and control

Supply chain management

Planning and control

systems

Capacity management

Lean operations

Inventory management

M10_SLAC8678_08_SE_C10.indd 315 6/2/16 2:22 PM

● Chapter 12 Supply chain management – This describes how operations relate to each other in the context of a wider network of suppliers and customers, and how these relationships can be managed.

● Chapter 13 Inventory management – This looks at how transformed resources accumulate as inventories as they flow through processes, operations or supply networks.

● Chapter 14 Planning and control systems – This describes how systems are needed to manage the very large amounts of information required to plan and control operations, and how enterprise resource planning (ERP) is used to do this.

● Chapter 15 Lean operations – This explains the concepts that underlie one of the most influential sets of ideas to impact operations management.

M10_SLAC8678_08_SE_C10.indd 316 06/02/16 5:00 PM

10 INtrODuctION The design of an operation determines the resources with which it creates its services and products, but then the operation has to deliver those services and products on an ongoing basis. and central to an operation’s ability to deliver is the way it plans its activities and controls them so that customers’ demands are satisfied. This chapter introduces and provides an overview of some of the principles and methods of planning and control. Later chapters in this part of the book develop some specific issues that are vital to an operation delivering its services and products. These issues start with managing capacity and moves through managing inventory, providing an overview of supply chain management and looking at how planning and control systems, particularly enterprise resources planning (ERP), manages the information that ensures effective delivery. We then examine how ‘lean’ philosophy has influenced operations practice. Figure 10.1 shows where this topic fits into the activities of operations management.

Planning and control

Key questions

❯ What is planning and control?

❯ What is the difference between planning and control?

❯ how do supply and demand affect planning and control?

❯ What are the activities of planning and control?

Topic covered in this chapter

Operations management

Direct

Design Develop

Deliver

Deliver

Planning and control

Supply chain management

Planning and control

systems

Capacity management

Lean operations

Inventory management

Figure 10.1 this chapter examines planning and control

M10_SLAC8678_08_SE_C10.indd 317 06/02/16 5:00 PM

318 PART THREE DELIVER

WHAT IS PLANNING AND CONTROL?

Planning and control is concerned with the activities that attempt to reconcile the demands of the market and the ability of the operation’s resources to deliver. It provides the systems, procedures and decisions which bring different aspects of supply and demand together. Consider, for example, the way in which routine surgery is organized in a hospital. When a patient arrives and is admitted to the hospital, much of the planning for the surgery will already have happened. The operating theatre will have been reserved, and the doctors and nurses who staff the operating theatre will have been provided with all the information regarding the patient’s condition. Appropriate preoperative and postoperative care will have been organized. All this will involve staff and facilities in different parts of the hospital, all of

whom must have been given the same information and their activities co-ordinated. Soon after the patient arrives, he or she will be checked to make sure that the condition is as expected (in much the same way as material is inspected on arrival in a factory). Blood, if required, will be cross-matched and reserved, and any medication will be made ready (in the same way that all the different materials are brought together in a factory). Any last-minute changes may require some degree of re-planning. For example, if the patient shows unexpected symptoms, observation may be necessary before the surgery can take

place. Not only will this affect the patient’s own treatment, but other patients’ treatment may also have to be rescheduled (in the same way as machines will need rescheduling if a job is delayed in a factory). All these activities of scheduling, co-ordination and organization are concerned with the planning and control of the hospital.

OPERATIONS IN PRACTICE

Joanne Cheung is the Senior Service Adviser at a pre- mier BMW dealership. She and her team act as the interface between customers who want their cars ser- viced and repaired and the 16 technicians who carry out the work in their state-of-the-art workshop. ‘ There are three types of work that we have to organize ’, says Joanne. ‘ The first is performing repairs on customers’ vehicles. They usually want this doing as soon as possible. The second type of job is routine servicing. It is usually not urgent so customers are generally willing to negotiate a time for this. The remainder of our work involves working on the pre-owned cars which our buyer has bought-in to sell on to customers. Before any of these cars can be sold they have to undergo extensive checks. To some extent we treat these categories of work slightly differently. We have to give good service to our internal car buyers, but there is some flexibility in planning these jobs. At the other extreme, emergency repair work for customers has to be fitted into our schedule as quickly as possible. If someone is desperate to have their car repaired at very short notice, we sometimes ask them to drop their car in as early as they can and pick it up as late as possible. This gives us the maximum amount of time to fit it into the schedule.

‘ There are a number of service options open to custom- ers. We can book short jobs in for a fixed time and do it while they wait. Most commonly, we ask the customer to leave the car with us and collect it later. To help customers we have ten loan cars which are booked out on a first-come first-served basis. Alternatively, the vehicle can be collected from the customer’s home and delivered back there when it is ready. Our four drivers who do this are able to cope with up to twelve jobs a day.

‘ Most days we deal with fifty to eighty jobs, tak- ing from half-an-hour up to a whole day. To enter a job

Joanne manages the schedule 1

✽ ✽ ✽ Operations principle Operations principle Operations principle Operations principle Operations principle Operations principle

So u

rc e:

S h

u tt

er st

o ck

.c o

m : J

o rd

an T

an

M10_SLAC8678_08_SE_C10.indd 318 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 319

WHAT IS THE DIFFERENCE BETWEEN PLANNING AND CONTROL?

Notice that we have chosen to treat ‘planning and control’ together. This is because the divi- sion between ‘planning’ and ‘control’ is not clear, either in theory or in practice. However, there are some general features that help to distinguish between the two. Planning is a formaliza- tion of what is intended to happen at some time in the future. But a plan does not guarantee that an event will actually happen. Rather it is a statement of intention. Although plans are based on expectations, during their implementation things do not always happen as expected. Customers change their minds about what they want and when they want it. Suppliers may not always deliver on time, process technology may fail, or staff may be absent through illness. Control is the process of coping with these types of change. It may mean that plans need to be redrawn in the short term. It may also mean that an ‘intervention’ will need to be made in the operation to bring it back ‘on track’ – for exam- ple, finding a new supplier who can deliver quickly, getting process technology up and running again, or moving staff from another part of the operation to cover for absentees. Control activi- ties make the adjustments which allow the operation to achieve the objectives that the plan has set, even when the assumptions on which the plan was based do not hold true.

Long-, medium- and short-term planning and control The nature of planning and control activities changes over time. In the very long term, oper- ations managers make plans concerning what they intend to do, what resources they need, and what objectives they hope to achieve. The emphasis is on planning rather than control, because there is little to control as such. They will use forecasts of likely demand described in aggregated terms. For example, a hospital will make plans for ‘2,000 patients’ without neces- sarily going into the details of the individual needs of those 2,000 patients. Similarly, the hos- pital might plan to have 100 nurses and 20 doctors, but again without deciding on the specific attributes of the staff. Operations managers will focus mainly on volume and financial targets.

Medium-term planning and control is more detailed. It looks ahead to assess the overall demand which the operation must meet in a partially disaggregated manner. By this time, for example, the hospital must distinguish between different types of demand. The number

into our process all Service Advisers have access to the computer-based scheduling system. On-screen it shows the total capacity we have day-by-day, all the jobs that are booked in, the amount of free capacity still available, the number of loan cars available, and so on. We use this to see when we have the capacity to book a customer in, and then enter all the customer’s details. BMW have issued “standard times” for all the major jobs. However, you have to modify these standard times a bit to take account of cir- cumstances. That is where the Service Adviser’s experience comes in.

‘ We keep all the most commonly used parts in stock, but if a repair needs a part which is not in stock, we can usually get it from the BMW parts distributors within a day. Every evening our planning system prints out the jobs to be done the next day and the parts which are likely to be needed for each job. This allows the parts staff to pick out the parts for each job so that the

technicians can collect them first thing the next morn- ing without any delay.

‘ Every day we have to cope with the unexpected. A technician may find that extra work is needed, custom- ers may want extra work doing, and technicians are sometimes ill, which reduces our capacity. Occasionally parts may not be available so we have to arrange with the customer for the vehicle to be rebooked for a later time. Every day up to four or five customers just don't turn up. Usually they have just forgotten to bring their car in so we have to rebook them in at a later time. We can cope with most of these uncertainties because our technicians are flexible in terms of the skills they have and also are willing to work overtime when needed. Also, it is important to manage customer’s expectations. If there is a chance that the vehicle may not be ready for them, it shouldn't come as a surprise when they try and collect it. '

✽ ✽ ✽ Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 319 6/2/16 2:22 PM

320 PART THREE DELIVER

of patients coming as accident and emergency cases will need to be distinguished from those requiring routine operations. Similarly, different categories of staff will have been identified and broad staffing levels in each category set. Just as important, contingencies will have been put in place which allow for slight deviations from the plans. These contingencies will act as ‘reserve’ resources and make planning and control easier in the short term.

In short-term planning and control, many of the resources will have been set and it will be difficult to make large changes. However, short-term interventions are possible if things are not going to plan. By this time, demand will be assessed on a totally disaggregated basis, with all types of surgical procedures treated as individual activities. More importantly, individual patients will have been identified by name, and specific time slots booked for their treatment. In making short- term interventions and changes to the plan, operations managers will be attempting to balance the quality, speed, dependability, flexibility and costs of their operation dynamically on an ad hoc basis. It is unlikely that they will have the time to carry out detailed calculations of the effects of their short-term planning and control decisions on all these objectives, but a general understand- ing of priorities will form the background to their decision making. Figure 10.2 shows how the control aspects of planning and control increase in significance closer to the date of the event.

The volume–variety effect on planning and control As we have found previously, the volume and variety characteristics of an operation will have an effect on its planning and control activities. Operations which produce a high variety of services or products in relatively low volume will have customers with different requirements

Figure 10.2 The balance between planning and control activities changes in the long, medium and short terms

M10_SLAC8678_08_SE_C10.indd 320 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 321

and use different processes from operations which create standardized services or products in high volume ( see Table 10.1 ).

Take two contrasting operations – an architects’ practice and an electricity utility. The architects’ high variety of customized services means they cannot produce designs in advance of customers requesting them. Because of this, the time it will take to deliver their services finally to customers will be relatively slow. Customers will understand this, but will expect to be consulted extensively as to their needs. The details and requirements of each job will emerge only as each individual building is designed to the client’s requirements, so planning occurs on a relatively short-term basis. The individual decisions which are taken in the plan- ning process will usually concern the timing of activities and events – for example, when a design is to be delivered, when building should start, when each individual architect will be needed to work on the design. Control decisions also will be at a relatively detailed level. A small delay in fixing one part of the design could have significant implications in many other parts of the job. For an architect, planning and control cannot be a totally routine matter; pro- jects need managing on an individual basis. However, the robustness of the operation (that is, its vulnerability to serious disruption if one part of the operation fails) will be relatively high. There are probably plenty of other things to get on with if an architect is prevented from progressing one part of the job.

The electricity utility, on the other hand, is very different. Volume is high, production is continuous and variety is non-existent. Customers expect instant ‘delivery’ whenever they plug in an appliance. The planning horizon in electricity generation can be very long. Major decisions regarding the capacity of power stations are made years in advance. Even the fluctuations in demand over a typical day can be forecast in advance. Popular television programmes can affect minute-by-minute demand and these are scheduled weeks or months ahead. The weather, which also affects demand, is more uncertain, but can to some extent be predicted. Individual planning decisions made by the electricity utility are not concerned with the timing, but rather the volume of output. Control decisions will concern aggre- gated measures of output such as the total kilowatts of electricity generated, because the product is more or less homogeneous. However, the robustness of the operation is very low because, if a generator fails, the operation’s capability of supplying electricity from that part of the operation also fails.

HOW DO SUPPLY AND DEMAND AFFECT PLANNING AND CONTROL?

If planning and control is the process of reconciling demand with supply, then the nature of the decisions taken to plan and control an operation will depend on both the nature of demand and the nature of supply in that operation. In this section, we examine some differ- ences in demand and supply which can affect the way in which operations managers plan and control their activities.

Table 10.1 The volume–variety effects on planning and control

Volume Variety Customer Planning Major planning Control Robustness responsiveness horizon decision decisions

Low High Slow Short Timing Detailed High

High Low Fast Long Volume Aggregated Low

✽ ✽ ✽ Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 321 6/2/16 2:22 PM

322 PART THREE DELIVER

OPERATIONS IN PRACTICE

‘ In many ways a major airline can be viewed as one large planning problem which is usually approached as many independent, smaller (but still difficult) planning prob- lems. The list of things which need planning seems endless: crews, reservation agents, luggage, flights, through trips, maintenance, gates, inventory, equipment purchases. Each planning problem has its own considerations, its own com- plexities, its own set of time horizons, its own objectives, but all are interrelated. ’

Air France has 80 flight planners working 24-hour shifts in its flight planning office at Roissy, Charles de Gaulle. Their job is to establish the optimum flight routes, anticipate any problems such as weather changes and minimize fuel consumption. Overall the goals of the flight planning activity are first, and most important, safety, followed by economy and passenger comfort. Increasingly powerful computer programs process the mountain of data necessary to plan the flights, but in the end many decisions still rely on human judgement. Even the most sophisticated expert systems only serve as support for the flight planners. Planning Air France’s schedule is a massive job. Just some of the considera- tions which need to be taken into account include the following:

● Frequency – for each airport how many separate ser- vices should the airline provide?

● Fleet assignment – which type of aircraft should be used on each leg of a flight?

● Banks – at any airline hub where passengers arrive and may transfer to other flights to continue their journey, airlines like to organize flights into ‘banks’ of several plans which arrive close together, pause to let passengers change planes, and all depart close together. So, how many banks should there be and when should they occur?

● Block times – a block time is the elapsed time between an aircraft leaving the departure gate at an airport and arriving at its gate in the arrival airport.

The longer the allowed block time, the more likely an aircraft will keep to schedule even if it suffers minor delays. However, longer block times also mean fewer flights can be scheduled.

● Planned maintenance – any schedule must allow time for aircraft to have time at a maintenance base.

● Crew planning – pilot and cabin crew must be sched- uled to allocate pilots to fly aircraft on which they are licensed and to keep within maximum ‘on duty ’ times for all staff.

● Gate plotting – if many aircraft are on the ground at the same time there may be problems in loading and unloading them simultaneously.

● Recovery – many things can cause deviations from any plan in the airline industry. Allowances must be built in to allow for recovery.

For flights within and between Air France’s 12 geo- graphic zones, the planners construct a flight plan that will form the basis of the actual flight only a few hours later. All planning documents need to be ready for the flight crew who arrive two hours before the scheduled departure time. Being responsible for passenger safety and comfort, the captain always has the final say and, when satisfied, co-signs the flight plan together with the planning officer.

Operations control at Air France 2

Uncertainty in supply and demand Uncertainty is important in planning and control because it makes it more difficult. Sometimes the supply of inputs to an operation may be uncertain. Local village carnivals, for example, rarely work to plan. Events take longer than expected, some of the acts scheduled in the programme may be delayed en route and some traders may not even arrive. In other operations supply is relatively predictable, and the need for control is minimal. For exam- ple, cable TV services provide programmes to a schedule into subscribers’ homes. It is rare to change the programme plan. Similarly demand may be unpredictable. A fast food outlet

So u

rc e:

S h

u tt

er st

o ck

.c o

m : A

n ge

lo

M10_SLAC8678_08_SE_C10.indd 322 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 323

inside a shopping centre does not know how many people will arrive, when they will arrive and what they will order. It may be possible to predict certain patterns, such as an increase in demand over the lunch and teatime periods, but a sudden rainstorm that drives shoppers indoors into the centre could significantly and unpredictably increase demand in the very short term. Conversely, demand may be more pre- dictable. In a school, for example, once classes are fixed and the term or semester has started, a teacher knows how many pupils are in the class. Both supply and demand uncertainty make planning and control more difficult, but a combination of supply and demand uncertainty is particularly difficult.

Dependent and independent demand Some operations can predict demand with relative certainty because demand for their ser- vices or products is dependent upon some other factor which is known. This is known as dependent demand. For example, the demand for tyres in an automobile factory is not a totally random variable. The process of demand forecasting is relatively straightforward. It will consist of examining the manufacturing schedules in the car plant and deriving the demand for tyres from these. If 600 cars are to be manufactured on a particular day, then it is simple to calculate that 3,000 tyres will be demanded by the car plant (each car has five tyres) – demand is dependent on a known factor, the number of cars to be manufactured. Because of this, the tyres can be ordered from the tyre manufacturer to a delivery schedule which is closely related to the demand for tyres from the plant (as in Fig. 10.3 ). In fact, the demand for every part of the car plant will be derived from the assembly schedule for the fin- ished cars. Other operations will act in a dependent demand manner because of the nature of the service or product which they provide. For example, a custom-made dressmaker will not buy fabric and make up dresses in many different sizes just in case someone comes along and

✽ ✽ ✽ Operations principle Operations principle Operations principle

Tyre manufacturer

Tyre fitter

Auto plant

Demand for tyres is largely governed by random factors

Demand for tyres is governed by the planned number of

cars to be made by the auto plant

Planned dem and for tyres

Random demand for tyres

Figure 10.3 Dependent demand is derived from the demand for something else; independent demand is more random

M10_SLAC8678_08_SE_C10.indd 323 6/2/16 2:22 PM

324 PART THREE DELIVER

wants to buy one. Nor will a high-class restaurant begin to cook food just in case a customer arrives and requests it. In both these cases, a combination of risk and the perishability of the product or service prevents the operation from starting to create the goods or services until it has a firm order. Planning and control in dependent demand situations is largely concerned with how the operation should respond when demand has occurred.

By contrast, some operations are subject to independent demand. They need to supply future demand without knowing exactly what that demand will be; or in the terminology of planning and control, they do not have firm ‘forward visibility’ of customer orders. For

example, the tyre fitter, which operates a drive-in tyre replacement service, will need to manage a stock of tyres. In that sense it is exactly the same task that faced the manager of tyre stocks in the car plant. However, demand is very different for the tyre fitter. It cannot predict either the volume or the specific needs of customers. It must make decisions on how many and what type of tyres to stock, based on demand forecasts and in the light of the risks it is prepared to run of

being out of stock. This is the nature of independent demand planning and control . It makes ‘best guesses’ concerning future demand, attempts to put the resources in place which can sat- isfy this demand, and attempts to respond quickly if actual demand does not match the fore- cast. Inventory planning and control, treated in Chapter 12 , is typical of independent demand planning and control.

Responding to demand It is clear then that the nature of planning and control in any operation will depend on how it responds to demand, which is in turn related to the type of services or products it produces. For example, an advertising agency will only start the process of planning and controlling the creation of an advertising campaign when the customer (or client as the agency will refer to them) confirms the contract with the agency. The creative ‘design’ of the advertisements will be based on a ‘brief’ from the client. Only after the design is approved will the appropriate resources (director, scriptwriters, actors, production company, etc.) be contracted. The actual shooting of the advertisement and post-production (editing, putting in the special effects, etc.) then goes ahead after which the finished advertisements are ‘delivered’ through tele- vision slots. This is shown in Figure 10.4 as a ‘Design, resource, create and deliver to order’ operation.

Other operations might be sufficiently confident of the nature of demand, if not its exact details, to keep ‘in stock’ most of the resources the operation requires to satisfy its customers. Certainly it will keep its transforming resources, if not its transformed resources. However, it would still make the actual service or product only when it receives a firm customer order. For example, a website designer will have most of its resources (graphic designers, software developers, specialist development software, etc.) in place, but must still design, create and deliver the website after it understands its customer’s requirements. (See the ‘Operations in practice’ case on Torchbox, in Chapter 1 .) This is shown in Figure 10.4 as a ‘Design, create and deliver to order’ operation.

Some operations offer relatively standard services or products, but do not create them until the customer has chosen which particular service or product to have. So a house builder who has standard designs might choose to build each house only when a customer places a firm order. Because the design of the house is relatively standard, suppliers of materials will have been identified, even if the building operation does not keep the items in stock itself. This is shown in Figure 10.4 as a ‘Create and deliver to order’ operation. In manufacturing it would be called a ‘Make to order’ operation.

Some operations have services or products that are so predictable that they can start to ‘create’ them before specific customer orders arrive. Possibly the best known example of this is Dell Computers, where customers can ‘specify’ their computer by selecting between vari- ous components through the company’s website. These components will have already been

✽ ✽ ✽ Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 324 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 325

created (usually by suppliers) but assembled to a specific customer order. This is shown in Figure 10.4 as a ‘ Partially create and deliver to order’ operation. In manufacturing it would be called an ‘Assemble to order’ operation.

When an operation’s services or products are standardized, there is the potential to create them entirely before demand is known. Almost all domestic products for example are ‘Created to stock’, or ‘Make to stock’ (shown in Fig. 10.4 ) from which they are delivered to customers. Taking this evolving logic to its conclusion, some operations require their customers to collect their own services or products. This is the ‘Choose/collect from stock’ illustration in Figure 10.4 . IKEA, and most high street retail operations, are like this.

✽ ✽ ✽ Operations principle Operations principle Operations principle

Obtain resources

Obtain resources

Create

Create

Create

CreateCreate

Choose

Choose

Design, resource, create and deliver to order

Design, create and deliver to order

Create and deliver to order

Partially create and deliver to order

Examples: advertising agency, construction project

Examples: website development, custom furniture production

Examples: hair blow-dry bar, house builder with standard designs

Examples: internet retail fulfilment, assemble to order computers (e.g. Dell)

Deliver

Deliver

Deliver

Deliver

Customer

Customer

Customer

Customer

Design

Design

P

D

P D

D P

D P

Create

Choose

Create to stock

Collect/choose from stock

Examples: preserved food production, domestic appliance production

Examples: collect retailer (e.g. IKEA), vending machines

Customer

D P

D P

Choose

Create Deliver Customer

High volume Low variety

Low volume High variety

Obtain resources

Obtain resources

DeliverObtain resources

Obtain resources

Figure 10.4 The P : D ratio of an operation indicates how long the customer has to wait for the service or product as compared with the total time to carry out all the activities to make the service or product available to the customer

M10_SLAC8678_08_SE_C10.indd 325 6/2/16 2:22 PM

326 PART THREE DELIVER

One point to note in the operations illustrated in Figure 10.4 is that there is a relation- ship between how operations respond to demand and their volume–variety characteristics. It is easy to see that ‘Design, resource, create and deliver to order’ operation is intended for low-volume and high-variety businesses. By definition, designing different services or prod- ucts will result in high variety, and performing each activity for each customer would be too cumbersome for a high-volume business. Conversely, ‘Create for stock’ or ‘Choose/collect from stock’ clearly rely on standardized services or products.

OPERATIONS IN PRACTICE

Uber is a company that has gathered many opponents, namely conventional taxi drivers in several countries, some regulatory bodies around the world, its compet- itors (obviously) – and sometimes even its customers. Why? Well partly because the company, and especially its boss, Travis Kalanick, is so outspoken, but mainly because it has challenged the way that the business of ordering a taxi is managed. Now, the smartphone-based car service has succeeded in disrupting the taxi industry in many of its markets. It has done this because the key input to any planning and control system is information: that is, infor- mation on what customers have ordered, or are likely to order, information on what resources are available to meet customer orders, information on priorities, and so on. And this holds true when you order a taxi. The job of managing the constant flow of customer requests and matching them to taxi availability has traditionally been the responsibility of central ‘despatching centres’. They are an information clearing-house, offering customers a central point of contact and offering busy drivers direc- tions to the nearest prospective passenger. Taxi drivers pay these despatchers a fee to keep the jobs coming. But then the central despatching operations started to be threatened by location-enabled smartphones.

These offered the potential directly to connect custom- ers with drivers, effectively cutting out the intermediary. It was an opportunity spotted by Travis Kalanick and his friend Garrett Camp one snowy night in Paris when they could not get a cab. The frustration led them to the idea of a simple app that could solve the problem. Why should it

not be possible to ‘push a button and get a car’? (It is ironic that the idea came to them in Paris, which became a cen- tre of resistance to the company; Parisian taxi drivers have been known to slash the tyres of Uber cars and smash their windows.) The advantages to the customer were obvious. Instead of either calling the despatching centre or stand- ing on the street until an available taxi happens to come along, then explaining where you want to go and having to find the cash (or fiddling with a credit card machine), using the app-based process is much simpler. Passengers sum- mon a vehicle from a smartphone, whose location facility allows them to be told when the cab is outside. It enters the destination into the driver’s navigation software and, because the cost is automatically charged to the custom- er’s account, lets them walk out immediately upon arrival.

Uber 3

P:D ratios 4 Another way of characterizing the graduation between ‘Design, resource, create and deliver to order’ and ‘Choose/collect from stock’ planning and control is by using a P : D ratio. This contrasts the total length of time customers have to wait between asking for the service or product and receiving it, called the demand time, D , and the total throughput time from start to finish, P . Throughput time is how long the operation takes to design the service or product (if it is customized), obtain the resources, create and deliver it.

So u

rc e:

G et

ty Im

ag es

: B lo

o m

b er

g /

Ju n

ko K

im u

ra -M

at su

m o

to

M10_SLAC8678_08_SE_C10.indd 326 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 327

P and D times depend on the operation P and D are illustrated for each type of operation in Figure 10.4 . Generally the ratio of P to D gets larger as operations move from ‘Design, resource, create and deliver to order’ to ‘Choose/ collect from stock’. In other words, as one moves down this spectrum towards the ‘Create to stock’ and ‘Choose/collect from stock’ end, the operation has anticipated customer demand and already created the services and products even though it has no guarantee that the antic- ipated demand will really happen. This is a particularly important point for the planning and control activity. The larger the P : D ratio, the more speculative the operation’s planning and control activities will be. In its extreme form, the ‘Choose/collect from stock’ operation, such as a high street retailer, has taken a gamble by designing, resourcing, creating and delivering (or, more likely, paying someone else to do so) products to its shops before it has any certainty that any customers will want them. Contrast this with a ‘Design, resource, create and deliver to order’ operation as in the advertising agency mentioned earlier. Here, D is the same as P and speculation regarding the volume of demand in the short term is eliminated because everything happens in response to a firm order. So by reducing their P : D ratio operations reduce their degree of speculative activity and also reduce their dependence on forecasting (although bad forecasting will lead to other problems).

But do not assume that when the P : D ratio approaches 1, all uncertainty is eliminated. The volume of demand (in terms of the number of customer ‘orders’) may be known, but not the time taken to perform each ‘order’. Take the advertising agency again: during each stage of the process, from design to delivery, it is common to have to seek the customer’s approval and/or feedback many times during each stage. Moreover, there will almost certainly be some recycling back through stages as modifications are made. And, in a similar way to how simultaneous development works in new service and product design (see Chapter 5 ) , a stage can be started before the previous one has been completed. So, for example, the video shoot director will have started prior to the artwork design being completed. This is illustrated in Figure 10.5 . So here it is the timings that are uncertain.

WHAT ARE THE ACTIVITIES OF PLANNING AND CONTROL?

Planning and control requires the reconciliation of supply and demand in terms of volumes, timing and quality. In this chapter we will focus on an overview of the activities that plan and control volume and timing (most of this part of the book is concerned with these issues). There

Create

Customer approval

P = D = ?

Deliver

Customer

Obtain resources

Design service/product

Figure 10.5 The relationship between stages in some ‘Design, resource, create and deliver to order’ operations such as an advertising agency, can be complex with frequent consultation and unpredictable recycling

✽ ✽ ✽ Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 327 6/2/16 2:22 PM

328 PART THREE DELIVER

are four overlapping activities: loading, sequencing, scheduling, and monitoring and control ( see Fig. 10.6 ). Some caution is needed when using these terms. Different organizations may use them in different ways, and even textbooks in the area adopt different definitions. For example, some authorities term what we have called planning and control as ‘operations scheduling’. However, the terminology of plan- ning and control is less important than understanding the basic ideas

described in the remainder of this chapter.

Loading Loading is the amount of work that is allocated to a work centre. For example, a machine on the shop floor of a manufacturing business is available, in theory, 168 hours a week. However, this does not necessarily mean that 168 hours of work can be loaded onto that machine. Figure  10.7 shows what erodes this available time. For some periods the machine cannot be worked; for exam- ple, it may not be available on statutory holidays and weekends. Therefore, the load put onto the machine must take this into account. Of the time that the machine is available for work, other

losses further reduce the available time. For example, time may be lost while changing over from making one component to another. If the machine breaks down, it will not be available. If there is machine relia- bility data available, this must also be taken into account. Sometimes the machine may be waiting for parts to arrive or be ‘idling’ for some other reason. Other losses could include an allowance for the machine being run below its optimum speed (for example, because it has not been main- tained properly) and an allowance for the ‘quality losses’ or defects which

the machine may produce. Of course, many of these losses (shown in Fig. 10.6 ) should be small or non-existent in a well-managed operation. However, the valuable operating time available for productive working, even in the best operations, can be significantly below the maximum time available. This idea is taken further in Chapter 11 when we discuss the measurement of capacity.

Finite and infinite loading Finite loading is an approach which only allocates work to a work centre (a person, a machine, or perhaps a group of people or machines) up to a set limit. This limit is the estimate of capac- ity for the work centre (based on the times available for loading). Work over and above this

✽ ✽ ✽ Operations principle Operations principle Operations principle

Figure 10.6 Planning and control activities

✽ ✽ ✽ Operations principle Operations principle Operations principle Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 328 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 329

capacity is not accepted. Figure 10.8 first shows how the load on the work centres is not allowed to exceed the capacity limit. Finite loading is particularly relevant for operations where:

● it is possible to limit the load – for example, it is possible to run an appointment system for a general medical practice or a hairdresser;

● it is necessary to limit the load – for example, for safety reasons only a finite number of peo- ple and weight of luggage are allowed on aircraft;

● the cost of limiting the load is not prohibitive – for example, the cost of maintaining a finite order book at a specialist sports car manufacturer does not adversely affect demand, and may even enhance it.

Infinite loading is an approach to loading work which does not limit accepting work, but instead tries to cope with it. The second diagram in Figure 10.8 illustrates this loading pattern where capacity constraints have not been used to limit loading so the work is completed ear- lier. Infinite loading is relevant for operations where:

● it is not possible to limit the load – for example, an accident and emergency department in a hospital should not turn away arrivals needing attention;

Figure 10.7 The reduction in the time available for valuable operating time

Figure 10.8 Finite and infinite loading of jobs on three work centres A, B and C. Finite loading limits the loading on each centre to their capacities, even if it means that jobs will be late. Infinite loading allows the loading on each centre to exceed their capacities to ensure that jobs will not be late

M10_SLAC8678_08_SE_C10.indd 329 6/2/16 2:22 PM

330 PART THREE DELIVER

● it is not necessary to limit the load – for example, fast food outlets are designed to flex capac- ity up and down to cope with varying arrival rates of customers. During busy periods, customers accept that they must queue for some time before being served. Unless this is extreme, the customers might not go elsewhere;

● the cost of limiting the load is prohibitive – for example, if a retail bank turned away custom- ers at the door because a set number were inside, customers would feel less than happy with the service.

In complex planning and control activities where there are multiple stages, each with dif- ferent capacities and with a varying mix arriving at the facilities, such as a machine shop in an engineering company, the constraints imposed by finite loading make loading calculations complex and not worth the considerable computational power which would be needed.

Sequencing Whether the approach to loading is finite or infinite, when work arrives, decisions must be taken on the order in which the work will be tackled. This activity is termed ‘sequencing’. The priorities given to work in an operation are often determined by some predefined set of rules, some of which are relatively complex. Some of these are summarized below.

Physical constraints The physical nature of the inputs being processed may determine the priority of work. For example, in an operation using paints or dyes, lighter shades will be sequenced before darker shades. On completion of each batch, the colour is slightly darkened for the next batch. This is because darkness of colour can only be added to and not removed from the colour mix. Sometimes the mix of work arriving at a part of an operation may determine the priority given to jobs. For example, when fabric is cut to a required size and shape in garment manu- facture, the surplus fabric would be wasted if not used for another product. Therefore, jobs that physically fit together may be scheduled together to reduce waste. The sequencing issue described in the case on airline passengers is of this type.

OPERATIONS IN PRACTICE

Like many before him, Dr Jason Steffen, a professional astrophysicist from the world-famous Fermilab, was frustrated by the time it took to load him and his fel- low passengers onto an aircraft. He decided to devise a way to make the experience a little less tedious. So, for a while, he neglected his usual work of examining extra-solar planets, dark matter and cosmology, and experimentally tested a faster method of boarding air- craft. He found that, by changing the sequence in which passengers are loaded onto the aircraft, airlines could potentially save both time and money. Using a computer simulation and the arithmetic techniques routinely used in his day-to-day job, he was able to find what seemed to be a superior sequencing method. In fact the most common way of boarding airliners proved to be the least efficient. This is called the ‘block method’ where blocks of seats are called for boarding, starting from the back.

Previously other experts in the airline industry had sug- gested boarding those in window seats first followed by middle and aisle seats. This is called the Wilma method.

Can airline passengers be sequenced? 5

So u

rc e:

S h

u tt

er st

o ck

.c o

m : T

ra vn

ik o

vS tu

d io

M10_SLAC8678_08_SE_C10.indd 330 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 331

But according to Dr Steffen’s simulations, two things slow down the boarding process. The first is that passengers may be required to wait in the aisle while those ahead of them store their luggage before they can take their seat. The second is that passengers already seated in aisle or middle seats frequently have to rise and move into the aisle to let others take seats nearer the window. So Dr Steffen suggested a variant of the Wilma method that minimized the first type of disturbance and eliminated the second. He suggested boarding in alternate rows, progressing from the rear forward, window seats first. Using this approach (now called the Steffen method), first the window seats for every other row on one side of the aircraft are boarded. Next, alternate rows of window seats on the opposite side are boarded. Then, the win- dow seats in the skipped rows are filled in on each side. The procedure then repeats with the middle seats and the aisles. See Figure 10.9 .

Later, the effectiveness of the various approaches were tested using a mock-up of a Boeing 757 aircraft and 72 luggage-carrying volunteers. Five different scenarios were tested: ‘block’ boarding in groups of rows from

back to front, one by one from back to front, the Wilma method, the Steffen method and completely random boarding. In all cases, parent–child pairs were allowed to board first. It was assumed that families were likely to want to stay together. As Dr Steffen had predicted, the conventional block approach came out as the slowest, with the strict back-to-front approach not much bet- ter. Completely random boarding (unallocated seating), which is used by several low-cost airlines, fared much better, most probably because it randomly avoids space conflicts. The times in minutes and seconds for fully boarding the 72 passengers using each method were as follows: block boarding, 6:54; back-to-front, 6:11; random boarding, 4:44; Wilma method, 4:13; Steffen method 3:36.

The big question is: ‘would passengers really be pre- pared to be sequenced in this way as they queue to board the aircraft?’ Some airlines argue that directing passengers onto an aircraft is a little like herding cats. But if they could adopt Dr Steffen’s system it would save time for customers and very significant amounts of money for airlines.

3 3 3

3 3 3

3 3 3

3 3 3

2 2 2

2 2 2

2 2 2

2 2 2

1 1 1

1 1 1

1 1 1

1 1 1

3 3 3

3 3 3

3 3 3

3 3 3

2 2 2

2 2 2

2 2 2

2 2 2

1 1 1

1 1 1

1 1 1

1 1 1

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

1 2 3

3 2

3 2

3 2

3 2 1

1

1

1

3 2 1

3 2 1

3 2 1

3 2 1

3 2 1

3 2 1

3 2 1

3 2 1

11

22

10

21

9

20

8

19

7

18

6

17

5

16

28 4

15

27 3

14

26 2

13

25 1

Block (conventional) method Wilma method Ste�en method

23

12

24

Figure 10.9 The best way to sequence passengers onto an aircraft

Customer priority Operations will sometimes use customer priority sequencing, which allows an important or aggrieved customer, or item, to be ‘processed’ prior to others, irrespective of the order of arrival of the customer or item. This approach is typically used by operations whose customer

M10_SLAC8678_08_SE_C10.indd 331 6/2/16 2:22 PM

332 PART THREE DELIVER

base is skewed, containing a mass of small customers and a few large, very important custom- ers. Some banks, for example, give priority to important customers. Similarly, in hotels, com- plaining customers will be treated as a priority because their complaint may have an adverse effect on the perceptions of other customers. More seriously, the emergency services often have to use their judgement in prioritizing the urgency of requests for service. For example, Figure 10.10 shows a typical triage system used in hospitals to prioritize patients (see the ‘Operations in practice’ case). However, customer priority sequencing, although giving a high level of service to some customers, may erode the service given to many other. This may lower the overall performance of the operation if work flows are disrupted to accommodate impor- tant customers.

Due date (DD) Prioritizing by due date means that work is sequenced according to when it is ‘due’ for deliv- ery, irrespective of the size of each job or the importance of each customer. For example, a support service, such as a printing unit, will often ask when copies are required, and then sequence the work according to that due date. Due date sequencing usually improves the delivery dependability and average delivery speed. However, it may not provide optimal pro- ductivity, as a more efficient sequencing of work may reduce total costs. However, it can be flexible when new, urgent work arrives at the work centre.

Last in, first out (LIFO) Last in, first out (LIFO) is a method of sequencing usually selected for practical reasons. For example, unloading an elevator is more convenient on a LIFO basis, as there is only one entrance and exit. However, it is not an equitable approach. Patients at hospital clinics may be infuriated if they see newly arrived patients examined first.

First in, first out (FIFO) Some operations serve customers in exactly the sequence they arrive in. This is called first in, first out (FIFO) sequencing, or sometimes ‘first come, first served’ (FCFS). For example, UK passport offices receive mail and sort it according to the day when it arrived. They work through the mail, opening it in sequence, and process the passport applications in order of arrival. Queues in theme parks may be designed so that one long queue snakes around the lobby area until the row of counters is reached. When customers reach the front of the queue, they are served at the next free counter.

Patients whose conditions are not true accidents or emergencies

Non-urgent5

Standard cases without immediate danger or distress

Standard4

Patients with serious problems, but apparently stable conditions

Urgent3

Seriously ill or injured patients whose lives are not in immediate danger

Very urgent2

Patient in need of immediate treatment for preservation of life

Immediate resuscitation

1

Figure 10.10 A triage prioritization scale

M10_SLAC8678_08_SE_C10.indd 332 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 333

Longest operation time (LOT) Operations may feel obliged to sequence their longest jobs first, called longest operation time sequencing. This has the advantage of occupying work centres for long periods. By contrast, relatively small jobs progressing through an operation will take up time at each work cen- tre because of the need to change over from one job to the next. However, although longest operation time sequencing keeps utilization high, this rule does not take into account deliv- ery speed, reliability or flexibility. Indeed, it may work directly against these performance objectives.

Shortest operation time (SOT) first Most operations at some stage become cash constrained. In these situations, the sequencing rules may be adjusted to tackle short jobs first; this is called shortest operation time sequenc- ing. These jobs can then be invoiced and payment received to ease cash-flow problems. Larger jobs that take more time will not enable the business to invoice as quickly. This has an effect of improving delivery performance, if the unit of measurement of delivery is jobs. However, it may adversely affect total productivity and can damage service to larger customers.

Judging sequencing rules All five performance objectives, or some variant of them, could be used to judge the effec- tiveness of sequencing rules. However, the objectives of dependability, speed and cost are particularly important. So, for example, the following performance objectives are often used:

● Meeting ‘due date’ promised to customer (dependability). ● Minimizing the time the job spends in the process, also known as ‘flow time’ (speed). ● Minimizing work-in-progress inventory (an element of cost). ● Minimizing idle time of work centres (another element of cost).

OPERATIONS IN PRACTICE

One of the hospital environments that is most difficult to schedule is the accident and emergency depart- ment , where patients arrive at random, without any prior warning, throughout the day. It is up to the hos- pital’s reception and the medical staff to devise very rapidly a schedule which meets most of the neces- sary criteria. In particular, patients who arrive having had very serious accidents, or presenting symptoms of a serious illness, need to be attended to urgently. Therefore, the hospital will schedule these cases first. Less urgent cases – perhaps patients who are in some discomfort , but whose injuries or illnesses are not life-threatening – will have to wait until the urgent cases are treated. Routine non-urgent cases will have the lowest priority of all. In many circumstances, these patients will have to wait for the longest time, which may be many hours, especially if the hospital is busy. Sometimes these non-urgent cases may even be turned away if the hospital is too busy with more important cases. In situations where hospitals expect

sudden influxes of patients, they have developed what is known as a triage system, whereby medical staff hurriedly sort through the patients who have arrived to determine which category of urgency each patient fits into. In this way a suitable schedule for the various treatments can be devised in a short period of time.

The hospital triage system

So u

rc e:

S h

u tt

er st

o ck

.c o

m : S

V L

u m

a

M10_SLAC8678_08_SE_C10.indd 333 6/2/16 2:22 PM

334 PART THREE DELIVER

Worked example

Steve Smith is a website designer in a business school. On returning from his annual vacation (he finished all outstanding jobs before he left), he is given five design jobs upon arrival at work. He gives them the codes A to E. Steve has to decide in which sequence to undertake the jobs. He wants both to minimize the average time the jobs are tied up in his office and, if possible, to meet the deadlines (delivery times) allocated to each job.

His first thought is to do the jobs in the order they were given to him, that is first in, first out (FIFO):

Sequencing rule – first in, first out (FIFO)

Sequence of jobs

Process time (days)

Start time Finish time Due date Lateness (days)

A 5 0 5 6 0 B 3 5 8 5 3 C 6 8 14 8 6 D 2 14 16 7 9 E 1 16 17 3 14

Total time in process 60 Total lateness 32 Average time in process (total/5) 12 Average lateness

(total/5) 6.4

Alarmed by the average lateness, he tries the due date (DD) rule:

Sequence of jobs

Process time (days)

Start time Finish time Due date Lateness (days)

E 1 0 1 3 0 B 3 1 4 5 0 A 5 4 9 6 3 D 2 9 11 7 4 C 6 11 17 8 9

Total time in process 42 Total lateness 16 Average time in process (total/5) 8.4 Average lateness

(total/5) 3.2

Better! But Steve tries out the shortest operation time (SOT) rule:

Sequencing rule – shortest operation time (SOT)

Sequence of jobs

Process time (days)

Start time Finish time Due date Lateness (days)

E 1 0 1 3 0 D 2 1 3 7 0 B 3 3 6 5 1 A 5 6 11 6 5 C 6 11 17 8 9

Total time in process 38 Total lateness 16 Average time in process (total/5) 7.6 Average lateness

(total/5) 3.2

This gives the same degree of average lateness but with a lower average time in the pro- cess. Steve decides to use the SOT rule.

M10_SLAC8678_08_SE_C10.indd 334 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 335

Comparing the results from the three sequencing rules described in the worked example together with the two other sequencing rules described earlier, and applied to the same prob- lem, gives the results summarized in Table 10.2. The SOT rule resulted in both the best aver- age time in process and the best (or least bad) in terms of average lateness. Although different rules will perform differently depending on the circumstances of the sequencing problem, in practice the SOT rule generally performs well.

Scheduling Having determined the sequence that work is to be tackled in, some operations require a detailed timetable showing at what time or date jobs should start and when they should end – this is scheduling. Schedules are familiar statements of volume and timing in many consumer environments. For example, a bus schedule shows that more buses are put on routes at more frequent intervals during rush-hour periods. The bus schedule shows the time each bus is due to arrive at each stage of the route. Schedules of work are used in operations where some planning is required to ensure that customer demand is met. Other operations, such as rapid-response service operations where customers arrive in an unplanned way, cannot schedule the operation in a short-term sense. They can only respond at the time demand is placed upon them.

The complexity of scheduling The scheduling activity is one of the most complex tasks in operations management. First, schedulers must deal with several different types of resource simultaneously. Machines will have different capabilities and capacities; staff will have different skills. More importantly, the number of possible schedules increases rapidly as the number of activities and processes increases. For example, suppose one machine has five different jobs to process. Any of the five jobs could be processed first and, following that, any one of the remaining four jobs, and so on. This means that there are:

5 * 4 * 3 * 2 =120 different schedules possible

In other words, for n jobs there are n! (factorial n) different ways of scheduling the jobs through a single process. But when there are (say) two machines, there is no reason why the sequence on machine 1 would be the same as the sequence on machine 2. If we consider the two sequencing tasks to be independent of each other, for two machines there would be:

120 * 120 = 14,400 possible schedules of the two machines and five jobs

So a general formula can be devised to calculate the number of possible schedules in any given situation, as follows:

Number of possible schedules =(n!)m

where n is the number of jobs and m is the number of machines.

Table 10.2 Comparison of five sequencing decision rules

Rule Average time in process (days) Average lateness (days)

FIFO 12 6.4

DD 8.4 3.2

SOT 7.6 3.2

LIFO 8.4 3.8

LOT 12.8 7.4

M10_SLAC8678_08_SE_C10.indd 335 6/2/16 2:22 PM

336 PART THREE DELIVER

In practical terms, this means that there are often many millions of feasible schedules, even for relatively small scheduling tasks. This is why scheduling rarely attempts to provide an ‘optimal’ solution but rather satisfies itself with an ‘acceptable’ feasible one.

Forward and backward scheduling Forward scheduling involves starting work as soon as it arrives. Backward scheduling involves starting jobs at the last possible moment to prevent them from being late. For example, assume that it takes six hours for a contract laundry to wash, dry and press a batch of overalls. If the work is collected at 8.00 am and is due to be picked up at 4.00 pm, there are more than six hours available to do it. Table 10.3 shows the different start times of each job, depending on whether they are forward or backward scheduled.

The choice of backward or forward scheduling depends largely upon the circumstances. Table 10.4 lists some advantages and dis- advantages of the two approaches. In theory, both materials require- ments planning (MRP , see Chapter 14 ) and lean, or just-in-time, planning (JIT , see Chapter 15 ) use backward scheduling, only start- ing work when it is required. In practice, however, users of MRP have tended to allow too long for each task to be completed, and therefore

each task is not started at the latest possible time. In comparison, JIT is started, as the name suggests, just in time.

Gantt charts One crude but simple method of scheduling is by use of the Gantt chart. This is a simple device which represents time as a bar, or channel, on a chart. The start and finish times for activities can be indicated on the chart and sometimes the actual progress of the job is also indicated. The advantages of Gantt charts are that they provide a simple visual representation both of what should be happening and of what actually is happening in the operation. Furthermore, they can be used to ‘test out’ alternative schedules. It is a relatively simple task to represent alternative schedules (even if it is a far from simple task to find a schedule which fits all the resources satisfactorily). Figure 10.11 illustrates a Gantt chart for a specialist software developer. It indicates the progress of several jobs as they are expected to progress through five stages of the process. Gantt charts are not an optimizing tool; they merely facilitate the development of alternative schedules by communicating them effectively.

Table 10.3 The effects of forward and backward scheduling

Task Duration Start time (backwards) Start time (forwards)

Press 1 hour 3.00 pm 1.00 pm

Dry 2 hours 1.00 pm 11.00 am

Wash 3 hours 10.00 am 8.00 am

Table 10.4 Advantages of forward and backward scheduling

Advantages of forward scheduling Advantages of backward scheduling

High labour utilization – workers always start work to keep busy

Lower material costs – materials are not used until they have to be, therefore delaying added value until the last moment

Flexible – the time slack in the system allows unexpected work to be loaded

Less exposed to risk in case of schedule change by the customer

Tends to focus the operation on customer due dates

✽ ✽ ✽ Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 336 6/2/16 2:22 PM

CHAPTER 10 PLANNING AND CONTROL 337

Figure 10.11 Gantt chart showing the schedule for jobs at each process stage

OPERATIONS IN PRACTICE

Pre-packed sandwiches are a growth product around the world as consumers put convenience and speed above relaxation and cost. But if you have recently consumed a pre-packed sandwich, think about the schedule of events which has gone into its making. For example, take a chicken salad sandwich. Less than five days ago, the chicken was on the farm, unaware that it would never see another weekend. The Gantt chart schedule shown in Figure 10.12 tells the story of the sandwich, and (posthumously) of the chicken.

From the forecast, orders for non-perishable items are placed for goods to arrive up to a week in advance of their use. Orders for perishable items will be placed daily, a day or two before the items are required. Tomatoes, cucumbers and lettuces have a three-day shelf life so may be received up to three days before production. Stock is held on a strict first in, first out (FIFO) basis. If today is (say) Wednesday, vegetables are processed that have been received dur- ing the last three days. This morning the bread arrived from a local bakery and the chicken arrived fresh, cooked and in strips ready to be placed directly in the sandwich during assembly. Yesterday (Tuesday) it had been killed, cooked, prepared and sent on its journey to the factory. By mid-day orders for tonight’s production will have been received on the Internet. From 2.00 pm until 10.00 pm the production lines are closed down for maintenance and a very thorough cleaning. During this time the pro- duction planning team is busy planning the night’s pro- duction run. Production for delivery to customers furthest away from the factory will have to be scheduled first. By 10 pm production is ready to start. Sandwiches are made on production lines. The bread is loaded onto a conveyor belt

by hand and butter is spread automatically by a machine. Next the various fillings are applied at each stage accord- ing to the specified sandwich ‘design’, see Figure 10.13 . After the filling has been assembled, the top slice of bread is placed on the sandwich and machine-chopped into two triangles, packed and sealed by machine. It is now early Thursday morning and by 2.00 am the first refrigerated lorries are already departing on their journeys to various customers. Production continues through until 2.00 pm on the Thursday, after which once again the maintenance and cleaning teams move in. The last sandwiches are des- patched by 4.00 pm on the Thursday. There is no finished goods stock.

Part two of the life and times of a chicken salad sand- wich is in Chapter 14 .

The life and times of a chicken salad sandwich – part one 6

So u

rc e:

S h

u tt

er st

o ck

.c o

m : Y

ek o

P h

o to

S tu

d io

M10_SLAC8678_08_SE_C10.indd 337 6/2/16 2:23 PM

338 PART THREE DELIVER

Figure 10.12 Simplified schedule for the manufacture and delivery of a chicken salad sandwich

Figure 10.13 Design for a chicken salad sandwich

Scheduling work patterns Where the dominant resource in an operation is its staff, then the schedule of work times effectively determines the capacity of the operation itself. The main task of scheduling, therefore, is to make sure that sufficient numbers of people are working at any point in time to provide a capacity appropriate for the level of demand at that point in time. This is often called staff rostering. Operations such as call centres, postal delivery, policing, holiday cou- riers, retail shops and hospitals will all need to schedule the working hours of their staff

M10_SLAC8678_08_SE_C10.indd 338 6/2/16 2:23 PM

CHAPTER 10 PLANNING AND CONTROL 339

with demand in mind. This is a direct consequence of these operations having relatively high ‘visibility’ (we introduced this idea in Chapter 1). Such operations cannot store their outputs in inventories and so must respond directly to customer demand. For example, Figure 10.14 shows the scheduling of shifts for a small technical ‘hot line’ support service for a small software company. It gives advice to customers on their technical problems. Its service times are 4:00 to 20:00 hours on Monday, 4:00 to 22:00 hours Tuesday to Friday, 6:00 to 22:00 hours on Saturday, and 10:00 to 20:00 hours on Sunday. Demand is heaviest Tuesday to Thursday, starts to decrease on Friday, is low over the weekend and starts to increase again on Monday.

The scheduling task for this kind of problem can be considered over different timescales, two of which are shown in Figure 10.14. During the day, working hours need to be agreed with individual staff members. During the week, days off need to be agreed. During the year, vacations, training periods, and other blocks of time where staff are unavailable need to be agreed. All this has to be scheduled such that:

● capacity matches demand; ● the length of each shift is neither excessively long nor too short to be attractive to staff; ● working at unsocial hours is minimized; ● days off match agreed staff conditions (in this example staff prefer two consecutive days

off every week); ● vacation and other ‘time-off’ blocks are accommodated; ● sufficient flexibility is built into the schedule to cover for unexpected changes in supply

(staff illness) and demand (surge in customer calls).

Scheduling staff times is one of the most complex of scheduling problems. In the relatively simple example shown in Figure 10.14 we have assumed that all staff have the same level and type of skill. In very large operations with many types of skill to schedule and uncertain demand (for example, a large hospital) the scheduling problem becomes extremely complex. Some mathematical techniques are available but most scheduling of this type is, in practice, solved using heuristics (rules of thumb), some of which are incorporated into commercially available software packages.

Monitoring and controlling the operation Having created a plan for the operation through loading, sequencing and scheduling, each part of the operation has to be monitored to ensure that planned activities are indeed happening. Any deviation from the plans can then be rectified through some kind

Figure 10.14 Shift scheduling for a small software company

M10_SLAC8678_08_SE_C10.indd 339 6/2/16 2:23 PM

340 PART THREE DELIVER

of intervention in the operation, which itself will probably involve some re-planning. Figure 10.15 illustrates a simple view of control. The output from a work centre is monitored and compared with the plan which indicates what the work centre is supposed to be doing. Deviations from this plan are taken into account through a re-planning activity and the necessar y inter ventions made to the work centre which will (hopefully) ensure that the new plan is carried out. Eventually, however, some further deviation from

planned activity will be detected and the cycle is repeated.

Push and pull control One element of control, then, is periodic inter vention into the activities of the opera- tion. An important decision is how this intervention takes place. The key distinction is between intervention signals which push work through the processes within the oper- ation and those which pull work only when it is required. In a push system of control, activities are scheduled by means of a central system and completed in line with central instructions, such as an MRP system (see Chapter 14 ) . Each work centre pushes out work without considering whether the succeeding work centre can make use of it. Work cen- tres are co-ordinated by means of the central operations planning and control system. In practice, however, there are many reasons why actual conditions differ from those planned. As a consequence, idle time, inventory and queues often characterize push sys- tems. By contrast, in a pull system of control, the pace and specification of what is done are set by the ‘customer’ workstation, which ‘pulls’ work from the preceding (supplier) workstation. The customer acts as the only ‘trigger’ for movement. If a request is not passed back from the customer to the supplier, the supplier cannot produce anything or move any materials. A request from a customer not only triggers production at the supplying stage, but also prompts the supplying stage to request a further delivery from its own suppliers. In this way, demand is transmitted back through the stages from the original point of demand by the original customer.

The inventory consequences of push and pull Understanding the differing principles of push and pull is important because they have different effects in terms of their pro-

pensities to accumulate inventory in the operation. Pull systems are far less likely to result in inventory build-up and are therefore favoured by lean operations (see Chapter 15 ) . To understand why this is so, consider an analogy: the ‘gravity’ analogy is illustrated in Figure 10.16 . Here a push system is represented by an operation,

✽ ✽ ✽ Operations principle Operations principle Operations principle Operations principle Operations principle Operations principle

Figure 10.15 A simple model of control

✽ ✽ ✽ Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 340 6/2/16 2:23 PM

CHAPTER 10 PLANNING AND CONTROL 341

each stage of which is on a lower level than the previous stage. When items are processed by each stage, gravity pushes them down the slope to the next stage. Any delay or varia- bility in processing time at that stage will result in the items accumulating as inventory. In the pull system, items cannot naturally f low uphill, so they can only progress if the next stage along deliberately pulls them forward. Under these circumstances, inventory cannot accumulate as easily.

Drum, buffer, rope The drum, buffer, rope concept comes from the theory of constraints (TOC) and a con- cept called optimized production technology (OPT), originally described by Eli Goldratt in his novel The Goal . 7 (We will deal more with his ideas in Chapter 15 .) It is an idea that helps to decide exactly where in a process control should occur. Most do not have the same amount of work loaded onto each separate work centre (that is, they are not perfectly balanced). This means there is likely to be a part of the process which is acting as a bot- tleneck on the work f lowing through the process. Goldratt argued that the bottleneck in the process should be the control point of the whole process. It is called the drum because it sets the ‘beat’ for the rest of the process to follow. Because it does not have sufficient capacity, a bottleneck is (or should be) working all the time. Therefore, it is sensible to keep a buffer of inventory in front of it to make sure that it always has something to work on. Because it constrains the output of the whole process, any time lost at the bottleneck will affect the output from the whole process. So it is not worthwhile for the parts of the process before the bottleneck to work to their full capacity. All they would do is produce work which would accumulate further along in the process up to the point where the bottleneck is constrain- ing the f low. Therefore, some form of communication between the bottleneck and the input to the process is needed to make sure that activities before the bottleneck do not overproduce. This is called the rope ( see Fig. 10.17 ).

A push system of control is where items are moved onto the next stage as soon as they have been processed

A pull system of control is when items are moved only when the next stage wants them

Figure 10.16 Push versus pull: the gravity analogy

✽ ✽ ✽ Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 341 6/2/16 2:23 PM

342 PART THREE DELIVER

Controlling operations is not always routine The simple monitoring control model in Figure 10.15 helps us to understand the basic func- tions of the monitoring and control activity. But, as the critical commentary box says, it is a simplification. Some simple routine processes may approximate to it, but many other opera- tions do not. In fact, some of the specific criticisms cited in the critical commentary box pro- vide a useful set of questions which can be used to assess the degree of difficulty associated with control of any operation, In particular:

● Is there consensus over what the operation’s objectives should be? ● Are the effects of interventions into the operation predictable? ● Are the operation’s activities largely repetitive?

Starting with the first question, are strategic objectives clear and unambiguous? It is not always possible (or necessarily desirable) to articulate every aspect of an operation’s objec- tives in detail. Many operations are just too complex for that. Nor does every senior manager always agree on what the operation’s objectives should be. Often the lack of a clear objective is because individual managers have different and conflicting interests. In social care organ- izations, for example, some managers are charged with protecting vulnerable members of

Figure 10.17 The drum, buffer, rope concept

Critical commentary

Most of the perspectives on control taken in this chapter are simplifi cations of a far more messy reality. They are based on models used to understand mechanical systems such as car engines. But anyone who has worked in real organizations knows that organizations are not machines. They are social systems, full of complex and ambiguous interactions. Simple models such as these assume that operations objectives are always clear and agreed, yet organizations are political entities where diff erent and often confl icting objectives compete. Local government operations, for example, are overtly political. Furthermore, the outputs from operations are not always easily measured. A university may be able to measure the number and qualifi cations of its students, for example, but it cannot measure the full impact of its education on their future happiness. Also, even if it is possible to work out an appropriate intervention to bring an operation back into ‘control’, most operations cannot perfectly predict what eff ect the intervention will have. Even the largest of burger bar chains does not know exactly how a new shift allocation system will aff ect performance. Also, some operations never do the same thing more than once anyway. Most of the work done by construction operations are one-off s. If every output is diff erent, how can ‘controllers’ ever know what is supposed to happen? Their plans themselves are mere speculation.

M10_SLAC8678_08_SE_C10.indd 342 6/2/16 2:23 PM

CHAPTER 10 PLANNING AND CONTROL 343

society, others with ensuring that public money is not wasted, and yet others may be required to protect the independence of professional staff. At other times objectives are ambiguous because the strategy has to cope with unpredictable changes in the environment, making the original objectives redundant. A further assumption in the simplified control model is that there is some reasonable knowledge of how to bring about the desired outcome. That is, when a decision is made, one can predict its effects with a reasonable degree of confidence. In other words, operational control assumes that any interventions which are intended to bring a process back under control will indeed have the intended effect. Yet, this implies that the relationships between the intervention and the resulting consequence within the process are predictable, which in turn assumes that the degree of process knowledge is high. For exam- ple, if an organization decides to relocate in order to be more convenient for its customers, it may or may not prove to be correct. Customers may react in a manner that was not predicted. Even if customers seem initially to respond well to the new location, there may be a lag before negative reactions become evident. In fact many operations decisions are taken on activities about which the cause–effect relationship is only partly understood. The final assumption about control is that control interventions are made in a repetitive way and occur frequently (for example, checking on a process, hourly or daily). This means that the operation has the opportunity to learn how its interventions affect the process which considerably facilitates control. However, some control situations are non-repetitive: for example, those involving unique services or products. So because the intervention, or the deviation from plan that caused it, may not be repeated, there is little opportunity for learning.

Figure 10.18 illustrates how these questions can form a ‘decision tree’ type of model that indicates how the nature of operations con- trol may be influenced. 8 Operational control is relatively straightfor- ward: objectives are unambiguous, the effects of interventions are known, and activities are repetitive. This type of control can be done using predetermined conventions and rules. There are, however, still some challenges to successful routine control. It needs opera- tional discipline to make sure that control procedures are system- atically implemented. The main point, though, is that any divergence from the conditions necessary for routine control implies a different type of control.

Yes No Negotiated control

Intuitive control

Trial-and-error control

Expert control

Routine control

Needs systematization skills

Needs networking skills

Needs knowledge building skills

Needs decision skills

Needs ‘political’ skills

No

No

NoYes

Yes

Yes

Are objectives unambiguous?

Is process knowledge complete?

Is activity repetitive?

Is activity repetitive?

Figure 10.18 Control is not always routine; different circumstances require different types of control

✽ ✽ ✽ Operations principle Operations principle Operations principle Operations principle Operations principle Operations principle

M10_SLAC8678_08_SE_C10.indd 343 6/2/16 2:23 PM

344 PART THREE DELIVER

Expert control If objectives are unambiguous, yet the effects of interventions relatively well understood, but the activity is not repetitive (for example, installing or upgrading software or IT sys- tems), control can be delegated to an ‘expert’ – someone for whom such activities are repetitive because they have built their knowledge on previous experience elsewhere. Making a success of expert control requires that such experts exist and can be ‘acquired’ by the firm. It also requires that the expert takes advantage of the control knowledge already present in the firm and integrates his or her ‘expert’ knowledge with the sup- port that potentially exists internally. Both of these place a stress on the need to ‘net- work’, in terms of both acquiring expertise and then integrating that expertise into the organisation.

Trial-and-error control If strategic objectives are relatively unambiguous, but effects of interventions not known, yet the activity is repetitive, the operation can gain knowledge of how to control successfully through its own failures. In other words, although simple prescriptions may not be available in the early stages of making control interventions, the organization can learn how to do it through experience. For example, if a fast food chain is opening new stores in new markets, it may not be sure how best to arrange the openings at first. But if the launch is the first of several, the objective must be not only to make a success of each launch, but equally (or more) important, to learn from each experience. It is these knowledge-building skills that will ulti- mately determine the effectiveness of trial-and-error control.

Intuitive control If objectives are relatively unambiguous (so it is clear what the operation is trying to do), but effects of control interventions not known, and neither are they repetitive, learning by trial and error is not possible. Here control becomes more of an art than a science. And in these circumstances control must be based on the management team using its intuition to make control decisions. Many strategic operations processes fall into this category – for example, setting up a strategic supply partnership (see Chapter 13). Objectives are clear ( jointly survive in the long term, make an acceptable return, and so on) but not only are con- trol interventions not repetitive and their effects not fully understood, but also sometimes the supplier’s interests may be in conf lict with yours. Yet, simply stating that ‘intuition’ is needed in these circumstances is not particularly helpful. Instinct and feelings are, of course, valuable attributes in any management team, but they are the result, at least partly, of understanding how best to organize the team members’ shared understanding, knowl- edge and decision-making skills. It requires thorough decision analysis, not to ‘mechanis- tically’ make the decision, but to frame it so that connections can be made, consequences understood and insights gained.

Negotiated control The most difficult circumstance for strategic control is when objectives are ambiguous. This type of control involves reducing ambiguity in some way by making objectives less uncertain. Sometimes this is done simply by senior managers ‘pronouncing’ or arbitrarily deciding what objectives should be, irrespective of opposing views. For example, controlling the activities of a childcare service can involve very different views among the professional social work- ers making day-to-day decisions. Often a negotiated settlement may be sought which then can become an unambiguous objective. Alternatively, outside experts could be used, either to help with the negotiations or to remove control decisions from those with conflicting views. But, even within the framework of negotiation, there is almost always a political element when ambiguities in objectives exist. Negotiation processes will be, to some extent, depend- ent on power structures.

M10_SLAC8678_08_SE_C10.indd 344 6/2/16 2:23 PM

CHAPTER 10 PLANNING AND CONTROL 345

● Planning and control is the reconciliation of the potential of the operation to supply prod- ucts and services, and the demands of its customers on the operation. It is the set of day-to- day activities that run the operation on an ongoing basis.

❯ What is planning and control?

SUMMARY ANSWERS TO KEY QUESTIONS

● A plan is a formalization of what is intended to happen at some time in the future. Control is the process of coping with changes to the plan and the operation to which it relates. Although planning and control are theoretically separable, they are usually treated together.

● The balance between planning and control changes over time. Planning dominates in the long term and is usually done on an aggregated basis. At the other extreme, in the short term, control usually operates within the resource constraints of the operation but makes interventions into the operation in order to cope with short-term changes in circumstances.

❯ What is the difference between planning and control?

● The degree of uncertainty in demand aff ects the balance between planning and control. The greater the uncertainty, the more diffi cult it is to plan, and greater emphasis must be placed on control.

● This idea of uncertainty is linked with the concepts of dependent and independent demand. Dependent demand is relatively predictable because it is dependent on some known fac- tor. Independent demand is less predictable because it depends on the chances of the mar- ket or customer behaviour.

● The diff erent ways of responding to demand can be characterized by diff erences in the P:D ratio of the operation. The P:D ratio is the ratio of total throughput time of services or prod- ucts to demand time.

❯ How do supply and demand affect planning and control?

● In planning and controlling the volume and timing of activity in operations, four distinct activities are necessary:

● loading, which dictates the amount of work that is allocated to each part of the operation;

● sequencing, which decides the order in which work is tackled within the operation;

● scheduling, which determines the detailed timetable of activities and when activities are started and fi nished;

● monitoring and control, which involve detecting what is happening in the operation, re-planning if necessary, and intervening in order to impose new plans. Two important types are ‘pull’ and ‘push’ control. Pull control is a system whereby demand is triggered by requests from a work centre’s (internal) customer. Push control is a centralized system whereby control (and sometimes planning) decisions are issued to work centres which are then required to perform the task and supply the next workstation. In manufactur- ing, ‘pull’ schedules generally have far lower inventory levels than ‘push’ schedules.

● The ease with which control can be maintained varies between operations.

❯ What are the activities of planning and control?

M10_SLAC8678_08_SE_C10.indd 345 6/2/16 2:23 PM

346 PART THREE DELIVER

C.K. was clearly upset. Since he had founded subText in the fast-growing South-East Asian computer-generated imag- ing (CGI) market, three years ago, this was the first time that he had needed to apologize to his clients. In fact, it had been more than an apology; he had agreed to reduce his fee, though he knew that did not make up for the delay. He admitted that, up to that point, he had not fully realized just how much risk there was, both reputational and finan- cial, in failing to meet schedule dates. It was not that either he or his team was unaware of the importance of reliability. On the contrary. ‘Imagination', ‘expertise’ and ‘reliability ’ all figured prominently in the studios’ promotional literature, mission statements, and so on. It was just that the ‘imagina- tion’ and ‘expertise’ parts had seemed to be the things that had been responsible for their success so far. Of course, it had been bad luck that, after more than a year of perfect reliability (not one late job), the two that had been late in the first quarter of this year had been particularly critical. ‘ They were both for new clients ’, said C.K., ‘ And neither of them indicated just how important the agreed delivery date was to them. We should have known, or found out, I admit. But it’s always more difficult with new clients, because with- out a track record with them, you don't really like even to admit the possibility of being late. ’

The company After studying computer science at the National University of Singapore, C.K. Ong had worked in CGI workshops in and around the Los Angeles area of California, after which he returned to Singapore to start subText Studios. At the heart of the company were the three ‘core’ departments that dealt sequentially with each job taken on. These three departments were ‘Pre-production’, ‘Production’ and ‘Post-production’.

Pre-production was concerned with taking and refining the brief as specified by the client, checking with and liaising with the client to iron out any ambiguities in the brief, sto- ryboarding the sequences, and obtaining outline approval of the brief from the client. In addition, pre-production also acted as account liaison with the client and was also respon- sible for estimating the resources and timing for each job. Pre-production also had nominal responsibility for monitor- ing the job through the remaining two stages, but generally it only did this if the client needed to be consulted during the production and post-production processes.

Production involved the creation of the imagery itself. This could be a complex and time-consuming process involving the use of state-of-the-art workstations and CGI software. Around 80 per cent of all production work was carried out in-house, but for some jobs other specialist

workshops were contracted. This was only done for work that subText either could not do, or would find difficult to do.

Post-production had two functions: the first was to inte- grate the visual image sequences produced by Production with other effects such as sound effects, music, voiceovers, etc.; the second was to cut, edit and generally produce the finished ‘product’ in the format required by the client.

Each of the three department employed teams of two people. Pre-production had two teams, Production three teams and Post-production two teams. For Pre-production and Post-production work, one team is always exclusively devoted to one job. ‘ We never allow either one team to be working on two jobs at the same time, or have both teams working on one job. It just doesn’t work because of the con- fusion it creates. That doesn’t apply to Production. Usually (but not always) the Production work can be parcelled up so that two or even all three of the teams could be working on different parts of it at the same time. Provided there is close coordination between the teams and provided that they are all committed to pulling it together at the end, there should be a more or less inverse relationship between the number of bodies working on the job and the length of time it takes. In fact, with the infamous “fifty-three slash F” job that’s exactly what we had to do. However, not withstanding what I just said about shortening the time, we probably did lose some efficiency there by having all three teams working on it. Our teams generally work until the job is finished. That level of work is factored in to the time estimates we make for each stage of the process. And, although we can be a little inac- curate sometimes, it’s because this type of thing is difficult to estimate. ’ (C.K. Ong)

The fifty-three slash F job The fifty-three slash F job, recently finished (late) and delivered to the client (dissatisfied), had been the source

CASE STUDY subText Studios Singapore 9

So u

rc e:

S h

u tt

er st

o ck

.c o

m : H

ai d

er Y

. A b

d u

lla

M10_SLAC8678_08_SE_C10.indd 346 6/2/16 2:23 PM

CHAPTER 10 PLANNING AND CONTROL 347

of much chaos, confusion and recrimination over the last two or three weeks. Although the job was only three days late, it had caused the client to postpone a presentation to its own client. Worse, subText had given only five days’ notice of late delivery, trying until the last minute to pull back onto schedule.

The full name of the job that had given subText so much trouble was 04/53/F. Table 10.5 shows the data for all the jobs started this year up to the current time (day 58; every working day was numbered throughout the year). Figure 10.19 shows the schedule for this period. The job had been accepted on day 18 and had seemed relatively

A

B

B C

C

D FSara/Liz

J/TC

Guy/J

Ri/Tim

Cy/Wes

Tom/Jo

P re

-p ro

d .

P ro

d u

c ti

o n

P o

st -p

ro d

.

Days

H

H

J

KI

I

G

G

E

E E

A D

F

F

F

D

A F

E

D

J

G

0 5 10 15 20 25 30 35 40 45 50 55 60

B

CMW/Tim

Figure 10.19 subText Studios Singapore – actual schedule for day 02 to 58

Job (04) Day in Estimated total time

Actual total time

Due date

Actual delivery

Pre-prod Prod Post-prod

Est Actual Est Actual Est Actual

06/A −4 29 30 40 34 6 8 11 10 12 12

11/B −4 22 24 42 31 4 5.5 7 7.5 11 11

04/C 2 31 30.5 43 40 9 9.5 12 13 10 9

54/D 5 28 34 55 58 10 12 12 17 6 5

31/E 15 34 25 68 57 10 11 12 14 12 –

53/F 18 32 49 50 53 6 10 18 28 8 11

24/G 25 26 20 70 – 9 11 9 9 8 –

22/H 29 32 26 70 – 10 12 14 14 8 –

22/I 33 30 11 75 – 10 11 12 – 8 –

09/J 41 36 14 81 – 12 14 14 – 10 –

20/K 49 40 – 89 – 12 – 14 – 14 –

Table 10.5 subText Studios Singapore – planning data for day 02 to day 58

M10_SLAC8678_08_SE_C10.indd 347 6/2/16 2:23 PM

348 PART THREE DELIVER

straightforward, although it was always clear that it would be a long production job. It was also clear that time was always going to be tight. There were 32 days in which to finish a job that was estimated to take 30 days.

‘ In hindsight we underestimated how much having three teams working on the production stage of this job at one point or other would increase its complexity. OK, it was not an easy piece of CGI to carry off, but we probably would have been OK if we had organized the CGI stage better. It was also real bad luck that, in our efforts to deliver the fifty-three slash F job on time, we also disrupted the fifty-four slash D job that turned out to be the only other new client we have had this year .’ (C.K. Ong)

The job had proved difficult from the start. The pre- production stage took longer than estimated, mainly because the client’s creative team changed just before the start of subText beginning the work. But it was the actual CGI itself that proved to be the major problem. Not only was the task intrinsically difficult, but it was also difficult to par- cel it up into separate packages that could be co-ordinated for working on by the two teams allocated to the job . More seriously, it became apparent within two or three days of starting the production work that they would need the help of another studio for some of the effects. Although the other studio was a regular supplier at short notice, this time it was too busy with its own work to help out. Help eventually came from a specialist studio in Hong Kong. ‘ The subcon- tracting delay was clearly a problem, but it was only half way through the production phase that we first realized just how

much difficulty the fifty-three slash F job was in. It was at that stage that we devoted all our production resources to finish- ing it. Unfortunately, even then, the job was late. The decision eventually to put all three teams on to the fifty-three slash F job was not easy because we knew that it would both disrupt other jobs and potentially cause more coordination problems .’

‘No way will we be doing that again’ ‘ No way will we be doing that again ’, said C.K. to the core teams when they met to pick over what had gone wrong. ‘ We are desperately in need of a more professional approach to keeping track of our activities. There is no point in me telling everyone how good we are if we then let them down. The problem is that I don't want to encourage a “command and control” culture in the studio. We depend on all staff feeling that they have the freedom to explore seemingly crazy options that may just lead to something real special. We aren't a factory. But we do need to get a grip on our esti- mating so that we have a better idea of how long each job really will take. After that each of the core departments can be responsible for their own planning. ’

QUESTIONS 1 What went wrong with the fifty-three slash F job

and how could the company avoid making the same mistakes again?

2 What would you suggest that subText do to tighten up its planning and control procedures?

1 Reread the ‘Operations in practice’ case on automobile service scheduling at the beginning of the chapter and also the case on Air France. What are the differences and what are the similarities between the planning and control task in these two operations?

2 A specialist sandwich retailer must order sandwiches at least eight hours before they are delivered. When they arrive in the shop, they are immediately displayed in a temperature- controlled cabinet. The average time that the sandwiches spend in the cabinet is six hours. What is the P : D ratio for this retail operation?

3 It is the start of the week and Marie, Willy and Silvie have three jobs to complete. The three of them can work on these jobs in any order. Job A requires 4 hours of Marie’s time, 5 hours of Willy ’s time and 3 hours of Silvie’s time. Job B requires 2 hours of Marie’s time, 8 hours of Willy ’s time and 7 hours of Silvie’s time. Job C requires 10 hours of Marie’s time, 4 hours of Willy ’s time and 5 hours of Silvie’s time. Devise a schedule for Marie, Willy and Sylvie that details when they will be working on each job. (Assume that they work seven hours per day.)

4 For the problem above, what is the loading on Marie, Willy and Silvie? If all the jobs have to be finished within two days, how much extra time must each of them work?

PROBLEMS AND APPLICATIONS

M10_SLAC8678_08_SE_C10.indd 348 6/2/16 2:23 PM

CHAPTER 10 PLANNING AND CONTROL 349

Step 1 – Make a list of all the jobs you have to do in the next week. Include in this list jobs relating to your work and/or study, jobs relating to your domestic life, in fact all the things you have to do. Step 2 – Prioritize all these jobs on a ‘most important’ to ‘least important’ basis. Step 3 – Draw up an outline schedule of exactly when you will do each of these jobs. Step 4 – At the end of the week compare what your schedule said you would do with what you actually have done. If there is a discrepancy, why did it occur? Step 5 – Draw up you own list of planning and control rules from your experience in this exer- cise in personal planning and control.

6 From your own experience of making appointments at your GP’s surgery, or by visiting who- ever provides you with primary medical care, reflect on how patients are scheduled to see a doctor or nurse. (a) What do you think planning and control objectives are for a GP’s surgery? (b) How could your own medical practice be improved?

SELECTED FURTHER READING

Chapman, S.N. (2005) Fundamentals of Production Planning and Control, Pearson, Harlow.

A detailed textbook, intended for those studying the topic in depth.

Goldratt, E.Y. and Cox, J. (1984) The Goal, North River Press, Croton-on-Hudson, NY.

Do not read this if you like good novels, but do read this if you want an enjoyable way of understanding some of the complexities of scheduling. It particularly applies to the drum, buffer, rope concept described in this chapter and it also sets the scene for the discussion of OPT in Chapter 14.

Kehoe, D.F. and Boughton, N. J. (2001) New paradigms in planning and control across manu- facturing supply chains – the utilization of Internet technologies, International Journal of Operations & Production Management, vol. 21, issue 5/6, 582–593.

An academic study of planning and control.

Vollmann, T.E., Berry, W.L., Whybark, D.C. and Jacobs, F.R. (2004) Manufacturing Planning and Control Systems for Supply Chain Management: The Definitive Guide for Professionals, McGraw- Hill Higher Education, New York.

The latest version of the ‘bible’ of manufacturing planning and control.

M10_SLAC8678_08_SE_C10.indd 349 6/2/16 2:23 PM

Figure 11.1 this chapter examines capacity management

Operations management

Direct

Design Develop

Deliver

Deliver

Planning and control

Supply chain management

Planning and control

systems

Capacity management

Lean operations

Inventory management

Topic covered in this chapter

introduction Capacity management is the activity of understanding the nature of an operation’s supply and demand, and of coping with any differences between them. It involves selecting supply-side responses (called capacity plans) and demand- side responses (called demand management and yield management). It aims to meet the needs of customers while maintaining the efficiency of the operation’s resources. and to do this, operations managers must be able to understand and reconcile two competing requirements. On the one hand there is the importance of maintaining customer satisfaction by delivering products and services to customers reasonably quickly. On the other there is the need for operations (and their extended supply networks) to maintain efficiency by minimizing the costs of excess capacity. In this chapter, we

look at these competing tensions at an aggregated level. at this level, managers do not discriminate between the different products and services that might be produced by the operation. Instead, they aim to ensure that the overall ability to supply is in line with the overall demand placed on the operation. Figure 11.1 shows where this chapter fits in the structure of the book. at the end of the chapter there is a supplement on queuing for those wishing to go into more detail on this important sub-topic of capacity management.

capacity management 11 Key questions

❯ what is capacity management?

❯ How are demand and capacity measured?

❯ How should the operation’s base capacity be set?

❯ what are the ways of coping with mismatches between demand and capacity?

❯ How can operations understand the consequences of their capacity decisions?

M11_SLAC8678_08_SE_C11.indd 350 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 351

wHat is caPacity managEmEnt?

The capacity of an operation is the maximum level of value-added activity over a period of time that the process can achieve under normal operating conditions. Critically, this definition reflects the scale of capacity but, more importantly, its processing capabilities. Suppose a phar- maceutical manufacturer invests in a new 1,000-litre capacity reactor or a property company purchases a 500-vehicle capacity car park. This information gives you a good sense of the scale of capacity but it is far from a useful measure of capacity for an operations manager. Instead, the pharmaceutical company will be concerned with the level of output (that is, the process- ing capability) that can be achieved from the 1,000-litre reactor vessel. If a batch of standard products can be produced every hour, the planned processing capacity could be as high as 24,000 litres per day. If the reaction takes four hours, and two hours are used for cleaning between batches, the vessel may only produce 4,000 litres per day. Similarly, the car park may be fully occupied by office workers during the working day, ‘processing’ only 500 cars per day. Alternatively, it may be used for shoppers staying on average only one hour, and theatre-goers occupying spaces for three hours, in the evening. The processing capability would then be up to 5,000 cars per day.

Capacity management is the activity of understanding the nature of demand for products and services, and effectively planning and controlling capacity in the short term, medium term and long term. All this must be done while reconciling the competing demands of customer satisfaction and resource efficiency. In Chapter 5 , we examined how long-term changes in demand can be considered in the structure and scope of operations. These decisions, termed ‘long- term capacity strategy’, are concerned with introducing (or removing) major increments of capacity. In this chapter we focus more on the medium- and short-term aspects of capacity management, where decisions are being made largely within the constraints of the physical capacity set by the operation’s long-term capacity strategy.

medium- and short-term capacity management Having established long-term capacity, operations managers must decide how to adjust the capacity of the operation in the medium term. This usually involves an assessment of the demand forecasts over a period of 2–18 months ahead, during which time planned output can be varied, for example by changing the number of hours when resources are used. In practice, however, few forecasts are accurate, and most operations also need to respond to changes in demand that occur over an even shorter timescale – termed ‘short-term capacity management’. Hotels and restaurants have unexpected and apparently random changes in demand from night to night, but also know from experience that certain days are on average busier than others. So operations managers also have to make short-term capacity adjust- ments, which enable them to flex output for a short period, either on a predicted basis (for example, bank checkouts are always busy at lunchtimes) or at short notice (for example, a warm sunny day at a theme park).

aggregate demand and capacity The important characteristic of capacity management, as we are treating it here, is that it is concerned with setting capacity levels over the medium and short terms in aggregated terms (in fact what we call capacity management here is sometimes called ‘aggregate planning’). That is, it is making overall, broad capacity decisions, but is not concerned with all of the detail of the individual products and services offered. This is what ‘aggregated’ means – different products and services are bundled together in order to get a broad view of demand and capacity. This may mean some degree of approximation, especially if the mix of

✽ operations principle Capacity is the maximum level of value- added activity over a period of time that the process or operation can achieve under normal operating conditions.

M11_SLAC8678_08_SE_C11.indd 351 06/02/16 7:20 PM

352 Part tHrEE DELIVER

products or services being created varies significantly (as we will see later in this chapter). Nevertheless, as a first step in capacity management, aggregation is necessary. For exam- ple, a hotel might think of demand and capacity in terms of ‘room nights per month’; this ignores the number of guests in each room and their individual requirements, but it is a good first approximation. A woollen knitwear factory might measure demand and capacity in the number of units (garments) it is capable of making per month, ignoring size, col- our or style variations. Aluminium producers could use tonnes per month, ignoring types of alloy, gauge and batch size variation. The ultimate aggregation measure is money. For example, retail stores, which sell an exceptionally wide variety of products, use revenue per month, ignoring variation in spend, number of items bought, the gross margin of each item and the number of items per customer transaction. If all this seems very approximate, remember that most operations have sufficient experience of dealing with aggregated data to find it useful.

oPErations in PracticE

With over 469,000 flights and 72.3 million passengers arriving and departing each year, London Heathrow is one of the busiest inter- national hubs in the world. Its size and location give it powerful ‘network effects’. this means that it can match incoming passengers with outgoing flights to hundreds of different cit- ies. and yet its attractiveness to the airlines is one of its main problems. On an average day, 60 per cent of arrivals, totalling over 55,000 customers, spend time in one of Heathrow ’s four ‘holding stacks’. these delays range from 4 to 10 minutes, rising to 20 minutes in the late morning peak, when between 32 and 40 jets typically circle over London. the costs of these delays include £119,000 of wasted fuel per day, 600 tonnes of additional CO 2 emissions, and the frustration of many customers losing valuable work and leisure time. the key problem is operating capacity, which currently stands at 98 per cent compared with around 70 per cent at most other major airports. ‘[When] you have [one of ] the most utilised pieces of infrastructure in the world, then one of the results is that you have air- borne holding ’, says Jon proudlove, managing Director of the national air traffic service (naS) at Heathrow. With no slack in capacity, the effect (as we have seen with the operations triangle in Chapter 6 ) is that any varia- tions (such as poor weather or poor conditions on the ground) have an immediate impact on aircraft processing speeds. the effects of Heathrow’s capacity management problem are starting to be felt, with several airlines seek- ing alternative capacity in the UK and Europe to expand their operations. yet the solutions to the problem are far

less clear. Considering medium- and short-term capacity management, significant investments have been made to air traffic systems to try to increase existing capacity (that is, to expand the ‘ability to serve’), and improve- ments to boarding processes have been trialled to ensure rapid plane turnaround. However, in the longer term the capacity question (within which medium- and short- term capacity management decisions are made) inev- itably turns to runway expansion. Heathrow currently operates two runways, compared with the four or five of its major European competitors Schiphol, madrid, paris and Frankfurt. yet building a new runway, while perhaps the obvious capacity solution, has recently been vetoed by UK politicians in the face of strong, largely environ- mental, opposition. So for the time being, Heathrow ’s operations managers must manage existing capacity as best they can.

Heathrow ’s capacity crises 1

So u

rc e:

a la

m y

Im ag

es : R

o ge

r B

am b

er

M11_SLAC8678_08_SE_C11.indd 352 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 353

capacity constraints Many organizations operate at below their maximum processing capacity, either because there is insufficient demand completely to ‘fill’ their capacity, or as a deliberate policy, so that the operation can respond quickly to every new order. Often, though, organizations find themselves with some parts of their operation operating below their capacity while other parts are at their capacity ‘ceiling’. It is the parts of the operation that are operating at their capacity ‘ceiling’ which are the capacity constraint for the whole operation. For example, a retail superstore might offer a gift-wrapping service that at normal times can cope with all requests for its services without delaying customers unduly. At Christmas, how- ever, the demand for gift wrapping might increase proportionally far more than the overall increase in custom for the store as a whole. Unless extra resources are provided to increase the capacity of this micro operation, it could constrain the capacity of the whole store.

the objectives of capacity management The decisions taken by operations managers in devising their capacity plans will affect several different aspects of performance:

● Costs will be affected by the balance between capacity and demand. Capacity levels in excess of demand could mean under-utilization of capacity and therefore high units cost.

● Revenues will also be affected by the balance between capacity and demand, but in the opposite way. Capacity levels equal to or higher than demand at any point in time will ensure that all demand is satisfied and no revenue lost.

● Working capital will be affected if an operation decides to build up finished goods inven- tory prior to demand. This might allow demand to be satisfied, but the organization will have to fund the inventory until it can be sold.

● Quality of goods or services might be affected by a capacity plan that involves large fluc- tuations in capacity levels, by hiring temporary staff, for example. The new staff and the disruption to the routine working of the operation could increase the probability of errors being made.

● Speed of response to customer demand could be enhanced either by the build-up of inventories (allowing customers to be satisfied directly from the inventory rather than having to wait for items to be manufactured) or by the deliberate provision of surplus capacity to avoid queuing.

● Dependability of supply will also be affected by how close demand levels are to capacity. The closer demand gets to the operation’s capacity ceiling, the less able it is to cope with any unexpected disruptions and the less dependable its deliveries of goods and services could be.

● Flexibility , especially volume flexibility, will be enhanced by surplus capacity. If demand and capacity are in balance, the operation will not be able to respond to any unexpected increase in demand.

the process of managing capacity The process of managing capacity is illustrated in Figure 11.2 . Typically, operations managers are faced with a forecast of demand which is unlikely to be either certain or constant. They will also have some idea of their own ability to meet this demand. Nevertheless, before any further decisions are taken, they must have quantitative data on both capacity and demand. So the first step will be to measure the aggregate demand and capacity levels and understand changes in these levels for the planning period. The second step is to determine the operation’s base level of capacity from which adjustments up or down will be made. This will largely be determined by the performance objectives of the operation, as well as the perishability of outputs, and degree of variability in both demand and supply. The third step is to identify and select methods of coping with mismatches between demand and capacity . Often operations managers will use a combination of ‘pure’ approaches – level capacity plan, chase demand

✽ operations principle Capacity is usually expressed in aggregated terms.

M11_SLAC8678_08_SE_C11.indd 353 06/02/16 7:20 PM

354 Part tHrEE DELIVER

plan and demand management. The fourth and final step is to understand the consequences of different capacity decisions. This often involves the use of cumulative representations, queuing principles and longer term outlooks.

How arE dEmand and caPacity mEasurEd?

understanding demand for products and services The first task of capacity management is to understand the nature of demand. Key questions include: What is the overall demand for a product or service over a period of time? How much does demand change? Are the total requirements and/or the changes in demand easy or dif- ficult to predict? In any period of time, how much does demand change, and how accurate is forecast demand likely to be? Although demand forecasting is often the responsibility of the sales and/or marketing functions, it is a very important input into the capacity manage- ment decisions, and so is of interest to operations managers. After all, without an estimate of future demand it is not possible to plan effectively for future events, only to react to them. It is therefore important to understand the basis and rationale for these demand forecasts (see the supplement on forecasting after Chapter 5). As far as capacity management is con- cerned, there are three requirements from a demand forecast:

● It is expressed in terms that are useful for capacity management. If forecasts are expressed only in money terms and give no indication of the demands that will be placed on an operation’s capacity, they will need to be translated into realistic expectations of demand, expressed in the same units as the capacity (for example, machine hours per year, opera- tives required, space, etc.).

● It is as accurate as possible. In capacity management, the accuracy of a forecast is important because, whereas demand can change instantaneously, there is usually a lag between decid- ing to change capacity and the change taking effect. Thus, many operations managers are faced with a dilemma. In order to attempt to meet demand, they must often decide output in advance, based on a forecast, which might change before the demand occurs or, worse, prove not to reflect actual demand at all.

● It gives an indication of relative uncertainty. Decisions to operate extra hours and recruit extra staff are usually based on forecast levels of demand, which could in practice differ considerably from actual demand, leading to unnecessary costs or unsatisfactory customer service. For example, a forecast of demand levels in a supermarket may show initially slow business that builds up to a lunchtime rush. After this, demand slows, only to build up again

Figure 11.2 the process of managing capacity

Measure aggregate demand and capacity Understand changes to demand and capacity

Determine the operation’s base level of capacity

Identify and select methods of coping with mismatches between demand and capacity

Understand the consequences of di�erent capacity decisions

M11_SLAC8678_08_SE_C11.indd 354 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 355

for the early evening rush, and finally falls again at the end of trading. The supermarket manager can use this forecast to adjust (say) checkout capacity throughout the day. But although this may be an accurate average demand forecast, no single day will exactly con- form to this pattern. Of equal importance is an estimate of how much actual demand could differ from the average. This can be found by examining demand statistics to build up a distribution of demand at each point in the day. The importance of this is that the manager now has an understanding of when it will be important to have reserve staff, perhaps filling shelves, but on call to staff the checkouts should demand warrant it. Generally, the advan- tage of probabilistic forecasts such as this is that it allows operations managers to make a judgement between possible plans that would virtually guarantee the operation’s ability to meet actual demand and plans that minimize costs. Ideally, this judgement should be influenced by the nature of the way the business wins orders: price-sensitive markets may require a risk-avoiding cost minimization plan that does not always satisfy peak demand, whereas markets that value responsiveness and service quality may justify a more generous provision of operational capacity.

understanding changes in demand Most markets are influenced by some kind of seasonality. Sometimes the causes of season- ality are climatic (holidays), sometimes festive (gift purchases), sometimes financial (tax processing), or social, or political; in fact there are many factors that affect the volume of activity in everything from construction materials to clothing, from healthcare to hotels. Typically, the term ‘seasonality’ is used to describe changes to demand over a period of a year. Yet similar variations in demand can also occur for some products and services over a shorter cycle. The daily and weekly demand patterns of a supermarket will fluctuate, with some degree of predictability. Demand might be low in the morning, higher in the afternoon, with peaks at lunchtime and after work in the evening. Demand might be low on Monday and Tuesday, build up during the latter part of the week and reach a peak on Friday and Saturday. Banks, public offices, telephone sales organizations and electricity utilities all have weekly and daily, or even hourly, demand patterns which require capacity adjustment. The extent to which an operation will have to cope with very short-term demand fluctuations is partly determined by how long its customers are prepared to wait for their products or services. An operation whose customers are incapable of, or unwilling to, wait will have to plan for very short-term demand fluctuations. Emergency services, for example, will need to understand the hourly variation in the demand for their services and plan capacity accordingly.

better forecasting or better operations responsiveness? The degree of effort (and cost) to devote to forecasting is often a source of heated debate within organizations. This often comes down to two opposing arguments. One goes some- thing like this. ‘ Of course it is important for forecasts to be as accurate as possible; we cannot plan operations capacity otherwise. This invariably means we finish up with too much capacity (thereby increasing costs), or too little capacity (thereby losing revenue and dissatisfying cus- tomers) .’ The counter-argument is very different. ‘ Demand will always be uncertain, that is the nature of demand. Get used to it. The only way to satisfy customers is to make the operation suf- ficiently responsive to cope with demand, almost irrespective of what it is .’ Both these arguments have some merit, but both are extreme positions. In practice, operations must find some bal- ance between having better forecasts and being able to cope without perfect forecasts.

Trying to get forecasts right has particular value where the operation finds it difficult or impossible to react to unexpected demand fluctuations in the short term. Internet-based retailers at some holiday times, for example, find it difficult to flex the quan- tity of goods they have in stock in the short term. Customers may

✽ operations principle Capacity management requires combining attempts to increase market knowledge with attempts to increase operations flexibility.

M11_SLAC8678_08_SE_C11.indd 355 06/02/16 7:20 PM

356 Part tHrEE DELIVER

not be willing to wait. On the other hand, other types of operations working in intrinsi- cally uncertain markets may develop fast and f lexible processes to compensate for the difficulty in obtaining accurate forecasts. For example, fashion garment manufactur- ers try to overcome the uncertainty in their market by shortening their response time to new fashion ideas (catwalk to rack time) and the time taken to replenish stocks in the stores (replenishment time). Similarly, when the cost of not meeting demand is very high, processes also have to rely on their responsiveness rather than accurate forecasts. For example, accident and emergency departments in hospitals must be responsive even if it means under-utilized resources at times.

oPErations in PracticE

panettone has become a national symbol of the Italian Christmas. the light and fluffy, dome-shaped, confection is dotted with sultanas and candied citrus peel, and is the Italian Christmas cake. traditionally made in milan, about 40 million of them are consumed throughout Italy over the holiday period. now, they are becoming popular around the world. Over a million are exported to the USa, while an endorsement from Delia Smith, a celebrity chef, caused a surge in demand in the UK with a well-publicized recipe for trifle made with panet- tone. this boost to production is good news for the big Italian manufacturers, but although volumes are higher, the product is still seasonal, which poses a problem for even the experienced milanese confectioners. Smaller, ‘artisan’ producers simply squeeze a few batches of pan- ettone into their normal baking schedules as Christmas approaches. But for the large industrial producers which need to make millions for the Christmas season it is not possible. and no panettone manufacturer is larger than the Bauli group. It is one of the foremost manufactur- ers of confectionery in Europe. Founded over 70 years ago, and in spite of its mass production approach, it has a reputation for quality and technological improvement. the company ’s output of panettone accounts for 38 per cent of Italian sales. the key to its success, according to the company, is in having ‘ combined the skill of home- made recipes with high technology [and] quality guaran- teed by high standards that are unattainable in craftsman production, but that can only be reached by selecting top quality raw materials, by thousands of tests and checks on the entire production line and the production process ’. In fact, the company says that its size is an advantage. ‘ High investment in research and technology allow us to man- age natural fermentation and guarantee a uniform quality that artisanal bakeries find hard to achieve .’

In fact, although Bauli has diversified into year-round products like croissants and biscuits, it has acquired a leadership role in the production of products for festive occasions. Seasonal cakes account for over 50 per cent of its turnover of around €420 million.

and so successful has it been in its chosen markets that in 2009 it bought motta and alemagna, the two big milanese brands that pioneered the manufacture of panettone. So how does Bauli cope with such sea- sonality? partly it is by hiring large numbers of tempo- rary seasonal workers to staff its dedicated production lines. at peak times there can be 1,200 seasonal work- ers in the factory, more than its permanent staff of around 800. It also starts to build up inventories before demand begins to increase for the Christmas peak. production of panettone lasts about four months, start- ing in September. ‘ Attention to ingredients and the use of new technologies in production give a shelf-life of five months without preservatives ’, says michele Bauli, Deputy Chairman, who comes from the firm’s founding family. temporary workers are also hired to bake other seasonal cakes such as the colomba , a dove-shaped Easter treat, which keeps them occupied for a month and a half in the spring.

Panettone: how italy ’s bakers cope with seasonal demand 2

M11_SLAC8678_08_SE_C11.indd 356 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 357

oPErations in PracticE

Founded as an online bookseller by Jeff Bezos, amazon, now the world’s biggest online retailer, started business in Europe in 1998. Since then the Seattle-based firm has experienced remarkable growth, employing around 40,000 people around the world and dominating a fiercely com- petitive consumer market, where its suc- cess is unquestionably based partly on its keen pricing. But low prices are not the only thing supporting amazon’s success. Without fast, accurate and efficient deliv- ery it could not have secured its ‘top e-re- tailer’ position.

this is why amazon devotes so much investment and effort to its fulfilment cen- tres, customer service centres and software development centres across Europe, north america, Latin america and asia that organ- ize the shipment of millions of individual items from bird tables to baby clothes. (amazon says that the exact num- ber of different items it sells is difficult to define, especially if you take the articles into account that are offered via the ‘marketplace’, and changes every day.) typical of amazon’s shipment operations is its 46,000 m 2 warehouse in milton Keynes, one of eight in the UK. In the warehouse prod- ucts are stocked within its extensive shelving arrangement with the position of every item referenced using a porta- ble satellite navigation system. amazon says that it saves time when their staff retrieve items. ‘ The computer shows the shortest pick-path ’, said arthur Valdez, vice-president of amazon’s UK operations. the item is then scanned and picked, after which it moves along a conveyor belt to be packed or gift wrapped and then labelled. at this point an email is automatically sent to the customer informing them that the product is on its way.

mr Valdez manages a network of fast-moving oper- ations that must always maintain a tight control of its activities, but no time is more testing than the run-up to Christmas. the gift-buying habits of Western con- sumers mean that up to 40 per cent of annual sales value can come in the final three months of the year, with half of the multibillion online Christmas sales tak- ing place over the end of november and the first two weeks of December. the average number of articles being sold each day soars from 300,000 to, at its peak, 3.6 million being sold in one day. In the UK, this day – which it calls ‘Cyber monday ’ – is at the beginning of December, or, to be more precise, at 9 pm on that

day, when shoppers, having normally been paid for the month and having spent the weekend browsing the high street, return from work to begin their Christmas shopping in earnest. It makes for a hectic time: ‘ A full truck is dispatched every three minutes and 24 seconds on our busiest trading day ’, says mr Valdez. But careful fore- casting can at least stop the Christmas peak being a sur- prise. and careful monitoring of customer behaviour has revealed a further trend: after ‘Cyber monday ’ comes ‘Boomerang thursday ’, when customers start to return their unwanted items. ‘ As the online retail sector contin- ues to grow, so too has consumer demand and confidence to return items, often before Christmas ’, says mark Lewis, Chief Executive of Collectplus, which allows customers to return items to a local convenience store. ‘ This suits retailers. They want to get [items] back as soon as possible, so they can sell them on. ’ mark Lewis says that half of his customers return items at off-peak times. ‘ It peaks at 7 pm. It reflects how we live our lives these days .’

However, some retail analysts believe that the advance of technology in the form of mobile phone transactions and broadband has also meant that the significance of ‘Cyber monday ’ and ‘Boomerang thursday ’ will diminish because such technology makes it easier to stagger transactions. But for mr Valdez, it is continual vigilance that allows amazon to keep up with demand trends. ‘ Every year it feels like [Christmas starts on] January 1. We are all-year-long focused on under- standing the lessons learnt from the previous Christmas ,’ he says.

christmas at amazon 3

So u

rc e:

S h

u tt

er st

o ck

.c o

m : J

ez p

er

M11_SLAC8678_08_SE_C11.indd 357 06/02/16 7:20 PM

358 Part tHrEE DELIVER

understanding capacity The second task of capacity management is to understand the nature of capacity or supply. Measuring capacity may sound simple, but can in fact be relatively hard to define unambigu- ously unless the operation is standardized and repetitive. So if a television factory produces only one basic model, the weekly capacity could be described as 2,000 Model A televisions. A govern- ment office may have the capacity to print and post 500,000 tax forms per week. A fast ride at a theme park might be designed to process batches of 60 people every three minutes – a capac-

ity to convey 1,200 people per hour. In each case, an output capacity measure is the most appropriate measure because the output from the operation does not vary in its nature. For many operations, however, the definition of capacity is not so obvious. When a much wider range of outputs places varying demands on the process, for instance, out- put measures of capacity are less useful. Here input capacity measures are frequently used to define capacity. Almost every type of operation

could use a mixture of both input and output measures, but in practice most choose to use one or the other ( see Table 11.1 ).

capacity depends on activity mix How much an operation can do depends on what it is being required to do. For example, a hospital may have a problem in measuring its capacity because the nature of the products and service may vary significantly. If all its patients required relatively minor treatment with only short stays in hospital, it could treat many people per week. Alternatively, if most of its patients required long periods of observation or recuperation, it could treat far fewer. Output depends

on the mix of activities in which the hospital is engaged and, because most hospitals perform many different types of activities, output is difficult (though not impossible!) to predict. Some of the problems caused by variation mix can be partially overcome by using aggregated capacity measures. (Remember that ‘aggregated’ means that differ- ent products and services are bundled together in order to get a broad view of demand and capacity.)

✽ operations principle any measure of capacity should reflect the ability of an operation or process to supply demand.

✽ operations principle Capacity is a function of service/product mix, duration, and product service specification.

table 11.1 input and output capacity measures for different operations

operation input measure of capacity output measure of capacity

air-conditioner plant machine hours available number of units per week

Hospital beds available number of patients treated per week

theatre number of seats number of customers entertained per week

University number of students Students graduated per year

Retail store sales fl oor area number of items sold per day

airline number of seats available on the sector

number of passengers per week

Electricity company generator size megawatts of electricity generated

Brewery Volume of fermentation tanks Litres per week

Note: The most commonly used measure is shown in bold.

M11_SLAC8678_08_SE_C11.indd 358 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 359

capacity depends on the duration over which output is required Capacity is the output that an operation can deliver in a defined unit of time. The level of activity and output that may be achievable over short periods of time is not the same as the capacity that is sustainable on a regular basis. For example, a tax return processing office, during its peak periods at the end (or beginning) of the financial year, may be capable of processing 120,000 applications a week. It does this by extending the working hours of its staff, discouraging its staff from taking vacations during this period, avoiding any potential disruption to its IT systems (not allowing upgrades during this period etc.), and maybe just by working hard and intensively. Nevertheless, staff do need vacations, nor can they work long hours continually, and eventually the information system will have to be upgraded. As such, when measuring capacity, operations managers should consider three different measures of capacity as shown in Figure 11.3 :

● Design capacity – the theoretical capacity of an operation that one of its technical designers had in mind when they commissioned it. For example, a company coating photographic paper will have several coating lines which deposit thin layers of chemicals onto rolls of paper at high speed. Each line will be capable of running at a particular speed. Multiplying the maximum coating speed by the operating time of the plant gives the theoretical design capacity of the line.

● Effective capacity – the capacity of an operation after planned losses are accounted for. For example, in the case above, the line cannot realistically be run continuously at its maximum

worked example

Suppose an air-conditioner factory produces three different models of air-conditioner unit: the deluxe, the standard and the economy. the deluxe model can be assembled in 1.5 hours, the standard in 1 hour and the economy in 0.75 hours. the assembly area in the factory has 800 staff hours of assembly time available each week.

If demand for deluxe, standard and economy units is in the ratio 2:3:2, the time needed to assemble 2 + 3 + 2 = 7 units is:

12 * 1.52 + 13 * 12 + 12 * 0.752 = 7.5 hours the number of units produced per week is:

800 7.5

* 7 = 746.7 units

If demand changes to a ratio of deluxe, economy, standard units of 1:2:4, the time needed to assemble 1 + 2 + 4 = 7 units is:

11 * 1.52 + 12 * 12 + 14 * 0.752 = 6.5 hours now the number of units produced per week is:

800 6.5

* 7 = 861.5 units

✽ operations principle Useable capacity is rarely equal to theoretical or ‘design’ capacity.

Unplanned loss

Actual output

Planned loss

E�ective capacity

Design capacity

Figure 11.3 design capacity, effective capacity and actual output

M11_SLAC8678_08_SE_C11.indd 359 06/02/16 7:20 PM

360 Part tHrEE DELIVER

rate. Different products will have different coating requirements, so the line will need to be stopped while it is changed over. Maintenance will need to be performed on the line, which will take out further productive time. Technical scheduling difficulties might mean further lost time. Not all of these losses are the operations manager’s fault; they have occurred because of the market and technical demands on the operation.

● Actual output – the capacity of an operation after both planned and unplanned losses are accounted for. For example, quality problems, machine breakdowns, absenteeism and other avoidable problems all take their toll. This means that the actual output of the line will be even lower than the effective capacity. The ratio of the output actually achieved by an operation to its design capacity, and the ratio of output to effective capacity, are called, respectively, the utilization and the efficiency of an operation:

Utilization = Actual output

Design capacity

Efficiency = Actual output

Effective capacity

worked example

Suppose the photographic paper manufacturer has a coating line with a design capacity of 200 m 2 per minute, and the line is operated on a basis of 24 hours a day, 7 days per week (168 hours per week).

Design capacity is 200 * 60 * 24 * 7 = 2.016 million m 2 per week. the records for a week’s production show the following lost production time:

1 2 3 4 5 6 7 8 9 10

product changeovers (set-ups) Regular preventative maintenance no work scheduled Quality sampling checks Shift change times maintenance breakdown Quality failure investigation Coating material stockouts Labour shortages Waiting for paper rolls

20 hours 16 hours 8 hours 8 hours 7 hours 18 hours 20 hours 8 hours 6 hours 6 hours

During this week the actual output was only 582,000 m 2 . the first five categories of lost production occur as a consequence of reasonably una-

voidable, planned occurrences and amount to a total of 59 hours. the last five categories are unplanned, and avoidable, losses and amount to 58 hours.

measured in hours of production:

Design capacity = 168 hours per week

Effective capacity = 168 - 59 = 109 hrs actual output = 168 - 59 - 58 = 51 hours

therefore:

Utilization = actual output

Design capacity =

51 hours 168 hours

= 0.304 = 30%

Efficiency = actual output

Effective capacity =

51 hours 109 hours

= 0.468 = 47%

M11_SLAC8678_08_SE_C11.indd 360 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 361

capacity depends on the specification of output Some operations can increase their output by changing the specification of the product or service (although this is more likely to apply to a service). For example, a postal service may effectively reduce its delivery dependability at peak times. So, during the busy Christmas period, the number of letters delivered the day after being posted may drop from 95 per cent to 85 per cent. Similarly, accounting firms may avoid long ‘relationship-building’ meetings with clients during busy periods. Important though these are, they can usually be deferred to less busy times. The important task is to distinguish between the ‘must do’ elements of the ser- vice that should not be sacrificed and the ‘nice to do’ parts of the service that can be omitted or delayed in order to increase capacity in the short term.

capacity ‘leakage’ Even after allowing for all the difficulties inherent in measuring capacity, the theoretical capacity of a process (the capacity that it was designed to have) is rarely achieved in prac- tice. Some reasons for this are, to some extent, predictable. Different products or services may have different requirements, so people and machinery will have delays when switch- ing between tasks. Maintenance will need to be performed on machines while training will be required for employees. Scheduling difficulties could mean further lost time. Not all of these losses are necessarily avoidable; they may occur because of the market and technical demands on the process. However, some of the reduction in capacity can be the result of less predictable events. For example, labour shortages, quality problems, delays in the delivery of bought-in products and services, and machine, or system, breakdown can all reduce capacity. This reduction in capacity is sometimes called ‘capacity leakage’ and one popular method of assessing this leakage is the overall equipment effectiveness (OEE) measure that is calculated as follows ( see Fig. 11.4 ):

OEE = a * p * q

where a is the availability of a process, p is the performance or speed of a process, and q is the quality of product or services that the process creates. OEE works on the assumption that some capacity leakage occurs causing reduced availability. For example, availability can be lost through time losses such as set-up and changeover losses (when equipment, or people in a service context, are being prepared for the next activity) and breakdown failures (when the machine is being repaired or in a service context where employees are being trained/absent).

critical commentary

For such an important topic, there is surprisingly little standardization in how capacity is measured. not only is a reasonably accurate measure of capacity needed for operations planning and control, but it is also needed to decide whether it is worth investing in extra capacity. yet not all practitioners would agree with the way in which design and eff ective capacity have been defi ned or measured in the previous worked example. For example, some would argue that the fi rst fi ve categories do not occur as a consequence of reasonably unavoidable, planned occurrences. product changeover set-ups can be reduced, allocating work in a diff erent manner between processes could reduce the amount of time when no work is scheduled, and re-examining preventative maintenance schedules could lead to a reduction in lost time. One school of thought is that whatever capacity effi ciency measures are used, they should be useful as diagnostic measures which can highlight the root causes of ineffi cient use of capacity. the idea of overall equipment eff ectiveness (OEE), described in the ‘capacity leakage’ section, is often put forward as a useful way of measuring capacity effi ciencies.

M11_SLAC8678_08_SE_C11.indd 361 06/02/16 7:20 PM

362 Part tHrEE DELIVER

Some capacity is lost through speed losses, such as when equipment is idling (for example, when it is temporarily waiting for work from another process) and when equipment is being run below its optimum work rate. In a service context, the same principle can be seen when individuals are not working at an optimum rate, for example mail order call centre employees in the quiet period after the winter holiday season. Finally, not everything processed by an operation will be error-free. So some capacity is lost through quality losses.

For processes to operate effectively, they need to achieve high levels of performance against all three dimensions – availability, performance (speed) and quality. Viewed in isola- tion, these individual metrics are important indicators of performance, but they do not give a complete picture of the process’s overall effectiveness. And critically, all these losses in the cal- culation mean that OEE represents the valuable operating time as a percentage of the capac- ity something was designed to have.

understanding changes in capacity While many operations are most concerned with dealing with changes in demand, some oper- ations also have to cope with variation in capacity (if it is defined as ‘the ability to supply’). For example, Figure 11.5 shows the demand and capacity variation of two businesses. The first is a domestic appliance repair service. Both demand and capacity vary month on month. Capacity varies because the field service operatives in the business prefer to take their vaca- tions at particular times of the year. Nevertheless, capacity is relatively stable throughout the year. Demand, by contrast, fluctuates more significantly. It would appear that there are two peaks of demand through the year, with peak demand being approximately twice the level of the low point in demand. The second business is a food manufacturer producing frozen spin- ach. The demand for this product is relatively constant throughout the year but the capacity of the business varies significantly. During the growing and harvesting season capacity to

Figure 11.4 operating equipment effectiveness (oEE)

Performance rate = p = net operating time/ total operating time

Quality rate = q = valuable operating time/

net operating time

Not worked (unplanned, e.g. forecast errors)

Breakdown failure

Rework losses

Complaint handling

Inspection activities

Learning losses (e.g. new sta�)

Under-performing sta�

Technology related delays

Available operating time

Total operating time Availa- bility

losses

Speed losses

Net operating time

Quality losses

Valuable operating time

Planning/ communication

delays

Availability rate = a = total operating time/

available operating time

M11_SLAC8678_08_SE_C11.indd 362 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 363

supply is high, but it falls off almost to zero for part of the year. Yet although the mismatch between demand and capacity is driven pri- marily by fluctuations in demand in the first case, and capacity in the second case, the essence of the capacity management activity is essen- tially similar for both.

✽ operations principle Capacity management decisions should reflect both predictable and unpredictable variations in capacity and demand.

worked example

In a typical seven-day period, the planning department programmes a particular machine to work for 150 hours – its loading time. Changeovers and set-ups take an average of 10 hours and breakdown failures average 5 hours every seven days. the time when the machine cannot work because it is waiting for material to be delivered from other parts of the process is five hours on average and during the period when the machine is running, it averages 90 per cent of its rated speed. Of the parts processed by the machine, 3 per cent are subse- quently found to be defective in some way.

maximum time available = 7 * 24 hours = 168 hours

Loading time = 150 hours availability losses = 10 hours (set-ups) + 5 hours (breakdowns)

= 15 hours So, total operating time = Loading time - availability

= 150 - 15 = 135 hours

Speed losses = 5 hours 1idling2 + 31135 - 52 * 0.1] (10% of remaining time)

= 18 hours So, net operating time = total operating time - Speed losses

= 135 - 18 = 117 hours

Quality losses = 117 1net operating time2 * 0.03 1error rate2 = 3.51 hours

So, valuable operating time = net operating time - quality losses = 117 - 3.51 = 113.49 hours

therefore, availability rate, a = total operating time

Loading time

= 117 135

= 86.67%

performance rate, p = net operating time

total operating time

= 117 135

= 86.67%

and quality rate, q = Valuable operating time

net operating time

= 113 .49

117 = 97%

OEE 1a * p * q2 = 75.6%

M11_SLAC8678_08_SE_C11.indd 363 06/02/16 7:20 PM

364 Part tHrEE DELIVER

How sHouLd tHE oPEration’s basE caPacity bE sEt?

The most common way of managing capacity is to decide the ‘base level’ of capacity and then adjust it periodically up or down to reflect fluctuations in demand. In fact, the concept of ‘base’ capacity is unusual because, although nominally it is the capacity level from which increases and decreases in capacity level are planned, in very unstable markets, where fluc- tuations are significant, it may never occur. Also, the two decisions of ‘what should the base level of capacity be?’ and ‘how do we adjust capacity around that base to reflect demand?’ are

interrelated. An operation could set its base level of capacity at such a high level compared with demand that there is never a need to adjust capacity levels. However, this is clearly wasteful, which is why most operations will adjust their capacity level over time. Nevertheless, although the two decisions are interrelated it is usually worthwhile setting a nominal base level of capacity before going on to consider how it can be adjusted.

setting base capacity The base level of capacity in any operation is influenced by many factors, but should be related to three in particular:

● The relative importance of the operation’s performance objectives. ● The perishability of the operation’s outputs. ● The degree of variability in demand or supply.

operation’s performance objectives Base levels of capacity should be set primarily to reflect an operation’s performance objec- tives, see Figure 11.6 . For example, setting the base level of capacity high compared with aver- age demand will result in relatively high levels of under-utilization of capacity and therefore high costs. This is especially true when an operation’s fixed costs are high and therefore the consequences of under-utilization are also high. Conversely, high base levels of capacity result in a capacity ‘cushion’ for much of the time, so the ability to flex output to give responsive customer service will be enhanced. When the output from the operation is capable of being stored, there may also be a trade-off between fixed capital and working capital where base capacity level is set. A high level of base capacity can require considerable investment while a lower base level would reduce the need for capital investment but may require inventory to

✽ operations principle the higher the base level of capacity, the less capacity fluctuation is needed to satisfy demand.

Figure 11.5 Volatility in demand versus volatility in capacity

Demand Demand

Ability to supply = Capacity

V o

lu m

e

Month on month volumes

Frozen spinach

Ability to supply = Capacity

V o

lu m

e

Month on month volumes

Domestic appliance repair service

M11_SLAC8678_08_SE_C11.indd 364 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 365

be built up to satisfy future demand, thus increasing working capital. For some operations, building up inventory is risky either because products have a short shelf life (for example, perishable food, high-performance computers, or fashion items) or because the output can- not be stored at all (most services).

the perishability of the operation’s outputs When either supply or demand is perishable, base capacity will need to be set at a relatively high level because inputs to the operation or outputs from the operation cannot be stored for long periods. For example, a factory that produces frozen fruit will need sufficient freezing, packing and storage capacity to cope with the rate at which the fruit crop is being harvested during its harvesting season. Similarly, a hotel cannot store its accommodation services. If an individual hotel room remains unoccupied, the ability to sell for that night has ‘perished’. In fact, unless a hotel is fully occupied every single night, its capacity is always going to be higher than the average demand for its services.

the degree of variability in demand or supply Variability, either in demand or capacity, will reduce the ability of an operation to process its inputs. That is, it will reduce its effective capacity. This effect was explained in Chapter 6 when the consequences of variability in individual processes were discussed. As a reminder, the greater the variability in arrival time or activity time at a process, the more the process will suffer both high throughput times and reduced utilization. This principle holds true for whole operations, and because long throughput times mean that queues will build up in the operation, high variability also affects inventory levels. This is illustrated in Figure 11.7. The implication of this is that the greater the variability, the more extra capacity will need to be provided to compensate for the reduced utilization of available capacity. Therefore, opera- tions with high levels of variability will tend to set their base level of capacity relatively high in order to provide this extra capacity. Of course, as we have seen earlier, not all operations have the option of simply increasing capacity! (See the ‘Operations in practice’ case on capacity constraints at Heathrow.)

Figure 11.6 the base level of capacity should reflect the relative importance of the operation’s performance objectives

Factors tending to increase base level of capacity • Low fixed costs • Need for high levels of customer service • High perishability (food, fashion, most services, etc.) • Inexpensive fixed capacity

Time

Demand

Base level of capacity

Factors tending to decrease base level of capacity • High fixed costs • Need for high-capacity utilisation • Ability to store output • Expensive fixed capacity

M11_SLAC8678_08_SE_C11.indd 365 06/02/16 7:20 PM

366 Part tHrEE DELIVER

wHat arE tHE ways oF coPing witH mismatcHEs bEtwEEn dEmand and caPacity?

With an understanding of both demand and capacity, the next step is to consider the alterna- tive methods of responding to any mismatches between them. For most organizations, this equates to understanding how demand might vary (although the same logic would apply to variation in capacity). In particular, the balance between predictable and unpredictable vari- ation in demand affects the nature of capacity management. When demand is stable and pre- dictable, the life of an operations manager is relatively easy! If demand is changeable, but this change is predictable, capacity adjustments may be needed, but at least they can be planned in advance. With unpredictable variation in demand, if an operation is to react to it at all, it must do so quickly; otherwise, the change in capacity will have little effect on the operation’s ability to deliver products and services as needed by their customers. Figure 11.8 illustrates how the objective and tasks of capacity management vary depending on the balance between predictable and unpredictable variation.

There are three ‘pure’ options available for coping with such variation (illustrated in Fig. 11.9), though organizations do tend to use a combination of all of these in practice:

● Level capacity plan – Ignore demand f luctuations and keep nominal capacity levels constant.

● Chase demand plan – Adjust capacity to reflect the fluctuations in demand. ● Demand management – Attempt to change demand to align with capacity.

Level capacity plan In a level capacity plan, the capacity is fixed throughout the planning period, regardless of the fluctuations in forecast demand. This means that the same number of staff operate the same processes and should therefore be capable of producing the same aggregate output in each period. Where non-perishable materials are processed, but not immediately sold, they can be transferred to finished goods inventory in anticipation of sales at a later time.

Level capacity plans of this type can achieve the objectives of stable employment patterns, high process utilization, and usually also high productivity with low unit costs.

Figure 11.7 the effect of variability on the utilization of capacity

1000

In ve

n to

ry /t

h ro

u g

h p

u t

tim e

0

Inventory/throughput time with higher base capacity

Inventory/throughput time with low base capacity

Capacity utilisation with low base capacity

Capacity utilisation with higher base capacity

X

X

M11_SLAC8678_08_SE_C11.indd 366 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 367

Unfortunately, they can also create considerable inventory which has to be financed and stored. Perhaps the biggest problem, however, is that decisions have to be taken as to what to produce for inventory rather than for immediate sale. Will green woollen sweaters knitted in July still be fashionable in October? Could a particular aluminium alloy in a specific sectional shape still be sold months after it has been produced? Most firms operating this plan, therefore, give priority to only creating inventory where future sales are relatively certain and unlikely to be affected by changes in fashion or design. Clearly, such plans are not suitable for ‘perishable’ products, such as foods and some pharmaceuticals, for products where fashion changes rap- idly and unpredictably (for example, fashion garments), or for customized products.

A level capacity plan could also be used by a hotel and supermarket, although this would not be the usual approach of such organizations, because it usually results in a waste of staff resources, reflected in low productivity. Because service cannot be stored as inventory, a level capacity plan would involve running the operation at a uniformly high level of capacity

Figure 11.8 the nature of capacity management depends on the mixture of predictable and unpredictable demand and capacity variation

Objective – Adjust capacity as fast as possible

Capacity management tasks • Identify sources of extra capacity and/or uses for surplus capacity • Work on how to adjust capacity and/or uses of capacity quickly

Objective – Make sure the base capacity is appropriate

Capacity management tasks • Seek ways of providing steady capacity e�ectively

Low

Objective – Adjust planned capacity as e�ciently as possible and enhance capability for further fast adjustments

Capacity management tasks • Combination of those for predictable and unpredictable variation

Objective – Adjust planned capacity as e�ciently as possible

Capacity management tasks • Evaluate optimum mix of methods for capacity fluctuation • Work on how to reduce cost of putting plan into e�ect

High

HighLow

Predictable variation

Unpredictable variation

Figure 11.9 managing mismatches between demand and capacity: ‘level capacity’, ‘chase demand’ and ‘demand management’ plans

V o

lu m

e

Time

V o

lu m

e

V o

lu m

e

Time Time

(a) Level capacity plan – absorb fluctuations

(b) Chase demand plan – change capacity to reflect demand fluctuations

(c) Demand management plan – attempt to change demand to reduce fluctuations

Demand

Capacity

Demand

Capacity Capacity

Original demand

Changed demand

M11_SLAC8678_08_SE_C11.indd 367 06/02/16 7:20 PM

368 Part tHrEE DELIVER

availability. The hotel would employ sufficient staff to service all the rooms, to run a full restaurant and to staff the reception even in months when demand was expected to be well below capacity. Similarly, the supermarket would plan to staff all the checkouts, warehousing operations, and so on, even in quiet periods.

Low utilization can make level capacity plans prohibitively expensive in many service operations, but may be considered appropriate where the opportunity costs of individual lost sales are very high, for example in the high-margin retailing of jewellery and in (real) estate agents. It is also possible to set the capacity somewhat below the forecast peak demand level

in order to reduce the degree of under-utilization. However, in the periods where demand is expected to exceed planned capacity, cus- tomer service may deteriorate. Customers may have to queue for long periods or may be ‘processed’ faster and less sensitively. While this is obviously far from ideal, the benefits to the organization of stability and productivity may outweigh the disadvantages of upsetting some customers.

chase demand plan The opposite of a level capacity plan is one which attempts to match capacity closely to the varying levels of forecast demand. This is much more difficult to achieve than a level capacity plan, as different numbers of staff, different working hours and even different amounts of equipment may be necessary in each period. For this reason, pure chase demand plans are unlikely to appeal to operations that manufacture standard, non-perishable products. Also, where manufacturing operations are particularly capital intensive, the chase demand policy would require a level of physical capacity, all of which would only be used occasionally. It is for this reason that such a plan is less likely to be appropriate for an aluminium producer than for a woollen garment manufacturer, for example. A pure chase demand plan is more usually adopted by operations that are not able to store their output, such as customer-processing operations or manufacturers of perishable products. It avoids the wasteful provision of excess staff that occurs with a level capacity plan, and yet should satisfy customer demand through- out the planned period. Where output can be stored, the chase demand policy might still be adopted in order to minimize or eliminate finished goods inventory, especially if the nature of future demand (in terms of volume or mix) is relatively unpredictable. There are a number of different methods for adjusting capacity, although they may not all be feasible for all types of operation. Some of these methods are shown in Table 11.2 .

demand management plan The third pure capacity management approach is demand management. Here, the objective is to change the pattern of demand to bring it closer to available capacity, by either stimulat- ing off-peak demand or by constraining peak demand. There are a number of methods for achieving this:

● Constraining customer access – customers may only be allowed access to the operation’s products or services at particular times. For example, reservation and appointment sys- tems in hospitals.

● Price differentials – adjusting price to reflect demand. That is, increasing prices during periods of high demand and reducing prices during periods of low demand. For example, skiing and camping holidays are cheapest at the beginning and end of the season and are particularly expensive during school vacations, while ice cream is on offer in many super- markets during the winter.

● Scheduling promotion – varying the degree of market stimulation through promotion and advertising in order to encourage demand during normally low periods. For example, tur- key growers in the UK and the USA make vigorous attempts to promote their products at times other than Christmas and Thanksgiving.

✽ operations principle the higher the base level of capacity, the less capacity fluctuation is needed to satisfy demand.

M11_SLAC8678_08_SE_C11.indd 368 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 369

table 11.2 methods of executing a chase demand plan

method of adjusting capacity advantages disadvantages

Overtime – staff working longer than their normal working times

Quickest and most convenient Extra payment normally necessary and agreement of staff to work can reduce productivity over long periods

annualized hours – staff contracting to work a set number of hours per year rather than a set number of hours per week

Without many of the costs associated with overtime the number of staff time available to an organization can be varied throughout the year to reflect demand

When very large and unexpected fluctuations in demand are possible, all the negotiated annual working time flexibility can be used before the end of the year

Staff scheduling – arranging working times (start and finish times) to vary the aggregate number of staff available for working at any time

Staffing levels can be adjusted to meet demand without changing job responsibilities or hiring new staff

providing start and finish (shift) times that both satisfy staffs’ need for reasonable working times and shift patterns as well as providing appropriate capacity can be difficult

Varying the size of the workforce – hiring extra staff during periods of high demand and laying them off as demand falls, or hire and fire

Reduces basic labour costs quickly Hiring costs and possible low productivity while new staff go through the learning curve. Lay-offs may result in severance payments and possible loss of morale in the operation and loss of goodwill in the local labour market

Using part-time staff – recruit staff who work for less than the normal working day (at the busiest periods)

good method of adjusting capacity to meet predictable short-term demand fluctuations

Expensive if the fixed costs of employment for each employee (irrespective of how long he or she works) are high

Skills flexibility – designing flexibility in job design and job demarcation so that staff can transfer across from less busy parts of the operation

Fast method of reacting to short-term demand fluctuations

Investment in skills training needed and may cause some internal disruption

Subcontracting/outsourcing – buying, renting or sharing capacity or output from other operations

no disruption to the operation Can be very expensive because of subcontractor’s margin and subcontractor may not be as motivated to give same service, or quality. also a risk of leakage of knowledge

Change output rate – expecting staff (and equipment) to work faster than normal

no need to provide extra resources Can only be used as a temporary measure, and even then can cause staff dissatisfaction, a reduction in the quality of work, or both

● Service differentials – allowing service levels to reflect demand (implicitly or explicitly), allowing service to deteriorate in periods of high demand and increase in periods of low demand. If this strategy is used explicitly, customers are being educated to expect varying levels of service and hopefully move to periods of lower demand.

A more radical approach attempts to create alternative products or services to fill capacity in quiet periods. It can be an effective demand management method but, ideally, new prod- ucts or services should meet three criteria: (a) they can be produced on the same processes, (b) they have different demand patterns to existing offerings, and (c) they are sold through similar marketing channels. For example, most universities fill their accommodation and lecture theatres with conferences and company meetings during vacations. Ski resorts may provide organized mountain activity holidays in the summer, and garden tractor companies may make snow blowers in the autumn and winter. However, the apparent benefits of filling

M11_SLAC8678_08_SE_C11.indd 369 06/02/16 7:20 PM

370 Part tHrEE DELIVER

critical commentary

to many, the idea of fl uctuating the workforce to match demand, either by using part- time staff or by hiring and fi ring, is more than just controversial. It is regarded as unethical. It is any business’s responsibility, they argue, to engage in a set of activities which are capable of sustaining employment at a steady level. Hiring and fi ring merely for seasonal fl uctuations, which can be predicted in advance, is treating human beings in a totally unacceptable manner. Even hiring people on a short-term contract, in practice, leads to their being off ered poorer conditions of service and leads to a state of permanent anxiety as to whether they will keep their jobs. On a more practical note, it is pointed out that, in an increasingly global business world where companies may have sites in diff erent countries, those countries that allow hiring and fi ring are more likely to have their plants ‘downsized’ than those where legislation makes this diffi cult.

oPErations in PracticE

Lowaters nursery is a garden plant and horticulture spe- cialist in the south of England employing around 25 peo- ple. Like any business that depends on seasonal weather conditions, it faces fluctuating demand for its services and products. It also prides itself on offering ‘ the best service in partnership with our customers, by communicating in a friendly professional manner and listening to our custom- ers to provide the result required ’. (Lowaters mission state- ment). But to maintain its quality of service throughout the seasonal ups and down in workload means keeping your core team happy and employed throughout the year. this is why Lowaters introduced its annualized hours scheme, a method of fluctuating capacity as demand var- ies throughout the year without many of the costs asso- ciated with overtime or hiring temporary staff. It involves staff contracting to work a set number of hours per year rather than a set number of hours per week. the main advantage of this is that the amount of staff time available to an organization can be varied throughout the year to reflect the real state of demand. annual hours plans can also be useful when supply varies throughout the year. maria Fox, one of the management team at Lowaters, says that annualized hours give the company several advantages. ‘ It simplifies administration and gives us the flexibility we need to run the business while delivering some real advantages to the employees. They are all effectively on salary with fixed monthly payments. We can flex the hours worked over the year – when we are busy we work longer and when things are quiet, in the winter, they can take time off. Everyone other than directors is contracted to work 39 hours on average over 52 weeks of the year. ’

the company created a simple spreadsheet that sets out the actual hours worked and compares them with a target distribution of the annualized hours that

are expected to be worked over the year. this allows employees to see at a glance whether someone is over or under target. ‘ We email them a copy of their sheet at the beginning of the year so they can keep track of their own progress as they go ’, says maria Fox. ‘ It also allows us to keep track of how many hours they do. If at the end of the year they come in plus or minus 50 hours we simply adjust it up or down for the next year. If there is a bigger discrepancy than that we’ll look at the job structure. ’

However, not all experiments with annualized hours have been as successful as that at Lowaters. In cases where demand is very unpredictable, staff can be asked to come in to work at very short notice. this can cause con- siderable disruption to social and family life. For example, at one news broadcasting company, the scheme caused problems. Journalists and camera crew who went to cover a foreign crisis found that they had worked so many hours that they were asked to take the whole of one month off to compensate. Since they had no holiday plans, many would have preferred to work for additional income.

annualized hours at Lowaters 4

So u

rc e:

a lis

ta ir

B ra

n d

o n

-J o

n es

M11_SLAC8678_08_SE_C11.indd 370 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 371

capacity in this way must be weighed against the risks of damaging the core product or ser- vice, and the operation must be fully capable of serving both markets.

yield management In operations which have relatively fixed capacities, such as airlines and hotels, it is impor- tant to use the capacity of the operation for generating revenue to its full potential. One approach used by such operations is called yield management. 6 This is really a collection of methods, some of which we have already discussed, which can be used to ensure that an operation maximizes its potential to generate profit. Yield management is especially useful where capacity is relatively fixed; the market can be fairly clearly segmented; the service cannot be stored in any way; the service is sold in advance; and the marginal cost of making a sale is relatively low.

Airlines, for example, fit all these criteria. They adopt a collection of methods to try to max- imize the yield (that is, profit) from their capacity. Over-booking capacity may be used to com- pensate for passengers who do not show up for the flight. However, if more passengers show up than they expect, the airlines will have a number of upset passengers. By studying past data on flight demand, airlines try to balance the risks of over-booking and under-booking. Operations may also use price discounting at quiet times, when demand is unlikely to fill

oPErations in PracticE

In service settings, queues will build up when demand exceeds capacity. nowhere is this more evident than during the daily rush hour. But in California, USa, a novel gaming approach may just offer an innovative solution to this age-old problem. Balaji prabhakat , a computer science professor, wanted to support Stanford’s efforts to alleviate rush-hour traffic in the local Santa Clara County. First, he managed to persuade nearly half of the university ’s 8,000 parking permit holders to install tracking devices in their cars. Second, he created a simple system that awards points to a person each time they arrive or leave an hour before or after rush hour. third, these points can then be used in an online game of chance with ran- dom cash rewards from $2 to $50. While the prizes are small, the idea has proved popular, with around 15 per cent of trips taken shifting out of the rush-hour peri- ods. Students are tending to arrive at and leave univer- sity later, while faculty are arriving and leaving earlier! the scheme is unlike classic peak-load pricing schemes, where customers pay more to use capacity when there is naturally high demand for it. For example, Seattle’s transit system charges a 75-cent excess to travel between 6.00 and 9.00 am. Rather, the gaming idea is

that some participants will change their habits for little or no reward, while others gain a (relatively) much big- ger reward. So, it is the randomness of the reward that makes this example a particularly interesting one. If the scheme can be rolled out to a wider population in the area, the effect of reducing load on stretched capacity by 15 per cent will be significant. the challenge, as with many improvement projects, is to retain the incentives for individuals to stick with the game even when they are not ‘winning ’ and when their attention is on other things.

demand management through gaming 5

So u

rc e:

g et

ty Im

ag es

: L o

o k

/ H

en d

ri k

H o

lle r

M11_SLAC8678_08_SE_C11.indd 371 06/02/16 7:20 PM

372 Part tHrEE DELIVER

capacity. For example, hotels will typically offer cheaper room rates outside of holiday peri- ods to try and increase naturally lower demand. In addition, many larger chains will sell heav- ily discounted rooms to third parties who in turn take on the risk (and reward) of finding customers for these rooms.

oPErations in PracticE

Baseball remains to this day one of the most popular sports in the USa and going to a match is a national pastime. Still, some games are a great deal more popular than others. For the new york mets, a Saturday night game against their cross-town rivals, the new york yankees, will have a great deal more natural demand than a mid-week game against a low-performing team from the other side of the country. and yet, for years, the price of a ticket at most baseball matches was based solely on the location of the ticket in the stadium rather than what the actual demand for the game was. However, this is changing across major League Baseball and nowhere more so than at Wrigley Field, home of the Chicago Cubs. Here, yield management, often referred to as ‘dynamic pricing ’, is being used to set ticket prices for games. So now, rather than having a ‘face value’ for a specific seat location within Wrigley Field, prices fluctuate daily based on ‘changing market factors’, as the team website states. So what are these factors? there are two that appear to dominate this approach to pricing – the day of the week and the opponent. First, prices of tickets rise steadily through the week, with a monday game averaging just $27 in the bleachers (the section of the stadium where analysis has been carried out), while a Saturday game ticket averages around $76.

On Sunday, the price falls back to $46. the opponent is also key to dynamic pricing too. games against the Boston Red Sox and the Chicago White Sox fetch the highest average ticket prices, given the prestige of the former and the local rivalry of the latter. Conversely, tickets for (apologies to their fans!) games against the atlanta Braves and the milwaukee Brewers average just $19 and $16. So for those with limited resources, go to a mid-week game against lowly opposition; for those who cannot skip work, go to a Sunday game; and for those where money is no object, enjoy Saturday night against the mighty Boston Red Sox!

Knowing when to watch a baseball game 7

mixed plans Each of the three ‘pure’ plans is applied only where its advantages strongly outweigh its dis- advantages. For many organizations, however, these ‘pure’ approaches do not match their required combination of competitive and operational objectives. Most operations managers are required simultaneously to reduce costs and inventory, to minimize capital investment, and yet to provide a responsive and customer-oriented approach at all times. For this reason, most organizations choose to follow a mixture of the three approaches. This can be illustrated by looking at a woollen knitwear company ( see Fig. 11.10 ). Here some of the peak demand has been brought forward by the company offering discounts to selected retail customers (manage demand plan). Capacity has also been adjusted at two points in the year to reflect the broad changes in demand (chase demand plan). Yet the adjustment in capacity is not suffi- cient to avoid totally the build-up of inventories (level capacity plan).

So u

rc e:

g et

ty Im

ag es

: R o

n V

es el

y

M11_SLAC8678_08_SE_C11.indd 372 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 373

How can oPErations undErstand tHE consEquEncEs oF tHEir caPacity dEcisions?

Before an operation adopts one or more of the three ‘pure’ capacity plans (level capacity, chase demand or demand management), it should examine the likely consequences. Three methods are particularly useful in helping to assess the consequences of adopting particular capacity plans:

● Consider capacity decisions using cumulative representations. ● Consider capacity decisions using queuing principles. ● Consider capacity decisions over time.

considering capacity decisions using cumulative representations Figure 11.11 shows the forecast aggregated demand for a chocolate factory that makes con- fectionery products. Demand for its products in the shops is greatest at Christmas. To meet this demand and allow time for the products to work their way through the distribution sys- tem, the factory must supply a demand that peaks in September, as shown. One method of assessing whether a particular level of capacity can satisfy the demand would be to calculate the degree of over-capacity below the graph which represents the capacity levels (areas A and C) and the degree of under-capacity above the graph (area B). If the total over-capacity is greater than the total under-capacity for a particular level of capacity, then that capacity could be regarded as adequate to satisfy demand fully, the assumption being that inventory has been accumulated in the periods of over-capacity. However, there are two problems with this approach. The first is that each month shown in Figure 11.11 may not have the same amount of productive time. Some months (August, for example) may contain vacation periods that reduce the availability of capacity. The second problem is that a capacity level that seems adequate may only be able to supply products after the demand for them has occurred. For example, if the period of under-capacity occurred at the beginning of the year, no inventory

Figure 11.10 a mixed capacity plan for a woollen knitwear factory

M11_SLAC8678_08_SE_C11.indd 373 06/02/16 7:20 PM

374 Part tHrEE DELIVER

could have accumulated to meet demand. A far superior way of assessing capacity plans is first to plot demand on a cumulative basis. This is shown as the heavier line in Figure 11.11 .

The cumulative representation of demand immediately reveals more information. First, it shows that although total demand peaks in September, because of the restricted number of available productive days, the peak demand per productive day occurs a month earlier in August. Second, it shows that the fluctuation in demand over the year is even greater than it seemed. The ratio of monthly peak demand to monthly lowest demand is 6.5:1, but the ratio of peak to lowest demand per productive day is 10:1. Demand per productive day is more relevant to operations managers, because productive days represent the time element of capacity.

The most useful consequence of plotting demand on a cumulative basis is that, by plot- ting capacity on the same graph, the feasibility and consequences of a capacity plan can be assessed. Figure 11.12 also shows a level capacity plan which produces at a rate of 14.03 tonnes per productive day. This meets cumulative demand by the end of the year. It would also pass our earlier test of total over-capacity being the same as or greater than under-capacity.

However, if one of the aims of the plan is to supply demand when it occurs, the plan is inadequate. Up to around day 168, the line representing cumulative production is above that representing cumulative demand. This means that at any time during this period, the factory has produced more products than has been demanded from it. In fact the vertical distance between the two lines is the level of inventory at that point in time. So by day 80, 1,122 tonnes have been produced but only 575 tonnes have been demanded. The surplus of production above demand, or inventory, is therefore 547 tonnes. When the cumulative demand line lies above the cumulative production line, the reverse is true. The vertical dis- tance between the two lines now indicates the shortage, or lack of supply. So by day 198, 3,025 tonnes have been demanded but only 2,778 tonnes produced. The shortage is there- fore 247 tonnes.

For any capacity plan to meet demand as it occurs, its cumulative production line must always lie above the cumulative demand line. This makes it a straightforward task to judge the adequacy of a plan, simply by looking at its cumulative representation. An impression of the inventory implications can also be gained from a cumulative rep- resentation by judging the area between the cumulative production and demand curves. This represents the amount of inventory carried

Figure 11.11 if the over-capacity areas (a + c) are greater than the under-capacity area (b), the capacity level seems adequate to meet demand. However, this may not necessarily be the case

operations principle For any capacity plan to meet demand as it occurs, its cumulative production line must always lie above its cumulative demand line.

M11_SLAC8678_08_SE_C11.indd 374 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 375

over the period. Figure 11.13 illustrates an adequate level capacity plan for the chocolate manufacturer, together with the costs of carrying inventory. It is assumed that inventory costs £2 per tonne per day to keep in storage. The average inventory each month is taken to be the average of the beginning- and end-of-month inventory levels, and the inventory carrying cost each month is the product of the average inventory, the inventory cost per day per tonne and the number of days in the month.

comparing plans on a cumulative basis Chase demand plans can also be illustrated on a cumulative representation. Rather than the cumulative production line having a constant gradient, it would have a varying gra- dient representing the production rate at any point in time. If a pure demand chase plan

Figure 11.12 a level capacity plan which produces shortages in spite of meeting demand at the end of the year

1,075 1,425 1,925 2,575 3,025 3,225 3,325

1,7401,4311,122 2,023 2,175 2,469 2,778 3,073 3,325

M11_SLAC8678_08_SE_C11.indd 375 06/02/16 7:20 PM

376 Part tHrEE DELIVER

were adopted, the cumulative production line would match the cumulative demand line. The gap between the two lines would be zero and hence inventory would be zero. Although this would eliminate inventory-carrying costs, as we discussed earlier, there would be costs associated with changing capacity levels. Usually, the marginal cost of making a capacity change increases with the size of the change. For example, if the choc- olate manufacturer wishes to increase capacity by 5 per cent, this can be achieved by requesting its staff to work overtime – a simple, fast and relatively inexpensive option. If the change is 15 per cent, overtime cannot provide sufficient extra capacity and temporary staff will need to be employed – a more expensive solution which also would take more time. Increases in capacity of above 15 per cent might only be achieved by subcontracting

Figure 11.13 a level capacity plan which meets demand at all times during the year

1,075 1,425 1,925 2,575 3,025 3,225 3,325

1,8951,5591,222 2,216 2,368 2,689 3,025 3,346 3,621

2,508 2,5624,120 9,720 7,524

M11_SLAC8678_08_SE_C11.indd 376 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 377

▼

worked example

Suppose the chocolate manufacturer, which has been operating the level capacity plan as shown in Figure 11.14 , is unhappy with the inventory costs of this approach. It decides to explore two alternative plans, both involving some degree of demand chasing.

Plan 1 ● Organize and staff the factory for a ‘normal’ capacity level of 8.7 tonnes per day. ● produce at 8.7 tonnes per day for the first 124 days of the year, then increase capacity to

29 tonnes per day by heavy use of overtime, hiring temporary staff and some subcontracting. ● produce at 29 tonnes per day until day 194, then reduce capacity back to 8.7 tonnes per

day for the rest of the year.

the costs of changing capacity by such a large amount (the ratio of peak to normal capacity is 3.33:1) are calculated by the company as being:

Cost of changing from 8.7 tonnes/day to 29 tonnes/day = £10,000 Cost of changing from 29 tonnes/day to 8.7 tonnes/day = £60,000

Figure 11.14 comparing two alternative capacity plans

some work out. This would be even more expensive. The point from which the change is being made, as well as the direction of the change, will also affect the cost of the change. Usually, it is less expensive to change capacity towards what is regarded as the ‘normal’ capacity level than away from it.

M11_SLAC8678_08_SE_C11.indd 377 06/02/16 7:20 PM

378 Part tHrEE DELIVER

Plan 2 ● Organize and staff the factory for a ‘normal’ capacity level of 12.4 tonnes per day. ● produce at 12.4 tonnes per day for the first 150 days of the year, then increase capacity to

29 tonnes per day by overtime and hiring some temporary staff. ● produce at 29 tonnes/day until day 190, then reduce capacity back to 12.4 tonnes per day

for the rest of the year.

the costs of changing capacity in this plan are smaller because the degree of change is smaller (a peak to normal capacity ratio of 2.34:1), and they are calculated by the company as being:

Cost of changing from 12.4 tonnes/day to 29 tonnes/day = £35,000 Cost of changing from 29 tonnes/day to 12.4 tonnes/day = £15,000

Figure 11.14 illustrates both plans on a cumulative basis. plan 1, which envisaged two drastic changes in capacity, has high capacity change costs but, because its production levels are close to demand levels, it has low inventory-carrying costs. plan 2 sacrifices some of the inventory cost advantage of plan 1 but saves more in terms of capacity change costs.

considering capacity decisions using queuing principles Cumulative representations of capacity plans are useful where the operation has the ability to store its finished goods as inventory. However, for operations where it is not possible to pro- duce products and services before demand for them has occurred, a cumulative representation would tell us relatively little. The cumulative ‘production’ could never be above the cumula- tive demand line. At best, it could show when an operation failed to meet its demand. So the vertical gap between the cumulative demand and production lines would indicate the amount of demand unsatisfied. Some of this demand would look elsewhere to be satisfied, but some would wait. This is why, for operations which, by their nature, cannot store their output, such as most service operations, capacity planning and control is best considered using waiting or queuing theory.

When we were illustrating the use of cumulative representations for capacity manage- ment, our assumption was that, generally, any plan should aim to meet demand at any point in time (the cumulative production line must be above the cumulative demand line). Looking at the issue as a queuing problem (in many parts of the world queuing concepts are referred to as ‘waiting line’ concepts) accepts that, while sometimes demand may be satisfied instantly, at other times customers may have to wait. This is particularly true when the arrival of individual demands on an operation are difficult to predict, or the time to produce a product or service is uncertain, or both. These circumstances make providing adequate capacity at all points in time particularly difficult. Figure 11.15 shows the general form of this capacity issue. Customers arrive according to some probability distribution and wait to be processed (unless part of the operation is idle); when they have reached the front of the queue, they are processed by one of the n parallel ‘servers’ (their processing time also being described by a probability distribution), after which they leave the operation. There are many examples of this kind of system. Table 11.3 illustrates some of these. All of these examples can be described by a common set of elements that define their queuing behaviour:

● The source of customers – Sometimes called the calling population, this is the source of supply of customers. In queue management ‘customers’ are not always human. ‘Customers’ could for example be trucks arriving at a weighbridge, orders arriving to be processed, machines waiting to be serviced, etc. The source of customers for a

M11_SLAC8678_08_SE_C11.indd 378 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 379

queuing system can be either f inite or inf inite. A finite source has a known number of possible customers. For example, if one maintenance person serves four assembly lines, the number of customers for the maintenance person is known, namely four. There will be a certain probability that one of the assembly lines will break down and need repairing. However, if one line really does break down, the probability of another line needing repair is reduced because there are now only three lines to break down. So, with a finite source of customers the probability of a customer arriving depends on the number of customers already being serviced. By contrast, infinite customer sources assume that there are a large number of potential customers so that it is always possi- ble for another customer to arrive no matter how many are being serviced. Most queu- ing systems that deal with outside markets have infinite, or ‘close-to-infinite’, customer sources.

● The arrival rate – This is the rate at which customers needing to be served arrive at the server or servers. Rarely do customers arrive at a steady and predictable rate. Usually there is variability in their arrival rate. Because of this it is necessary to describe arrival rates in terms of probability distributions. The important issue here is that, in queuing systems, it is normal that at times no customers will arrive and at other times many will arrive relatively close together.

● The queue – Customers waiting to be served form the queue or waiting line itself. If there is relatively little limit on how many customers can queue at any time, we can assume that, for all practical purposes, an infinite queue is possible. Sometimes, however, there is a limit to how many customers can be in the queue at any one time.

Figure 11.15 capacity management as a queuing problem

table 11.3 Examples of operations which have parallel processors

operation arrivals Processing capacity

Bank Supermarket Hospital clinic graphic artist Custom cake decorators ambulance service telephone switchboard maintenance department

Customers Shoppers patients Commissions Orders Emergencies Calls Breakdowns

tellers Checkouts Doctors artists Cake decorators ambulances with crews telephonists maintenance staff

M11_SLAC8678_08_SE_C11.indd 379 06/02/16 7:20 PM

380 Part tHrEE DELIVER

● Rejecting – If the number of customers in a queue is already at the maximum number allowed, then the customer could be rejected by the system. For example, during periods of heavy demand some websites will not allow customers to access part of the site until the demand on its services has declined.

● Baulking – When a customer is a human being with free will (and the ability to get annoyed) he or she may refuse to join the queue and wait for service if it is judged to be too long. In queuing terms this is called baulking.

● Reneging – This is similar to baulking but here the customer has queued for a certain length of time and then (perhaps being dissatisfied with the rate of progress) leaves the queue and therefore the chance of being served.

● Queue discipline – This is the set of rules that determine the order in which customers waiting in the queue are served. Most simple queues, such as those in a shop, use a first- come, first-served queue discipline. The various sequencing rules described in Chapter 10 are examples of different queue disciplines.

● Ser vers – A server is the facility that processes the customers in the queue. In any queuing system there may be any number of servers configured in different ways. In Figure 11.15 servers are configured in parallel, but some may have servers in a series arrangement. For example, on entering a self-service restaurant you may queue to col- lect a tray and cutlery, move on to the serving area where you queue again to order and collect a meal, move on to a drinks area where you queue once more to order and collect and drink, and then finally queue to pay for the meal. In this case you have passed through four ser vers (even though the first one was not staffed) in a series arrangement. Of course, many queue systems are complex arrangements of series and parallel connections. There is also likely to be variation in how long it takes to process each customer. Even if customers do not have differing needs, human servers will vary in the time they take to perform repetitive serving tasks. Therefore processing time, like arrival time, is usually described by a probability distribution.

balancing capacity and demand The dilemma in managing the capacity of a queuing system is how many servers to have available at any point in time in order to avoid unacceptably long queuing times or unaccept- ably low utilization of the servers. Because of the probabilistic arrival and processing times, only rarely will the arrival of customers match the ability of the operation to cope with them. Sometimes, if several customers arrive in quick succession and require longer than average processing times, queues will build up in front of the operation. At other times, when custom- ers arrive less frequently than average and also require shorter than average processing times, some of the servers in the system will be idle. So even when the average capacity (processing capability) of the operation matches the average demand (arrival rate) on the system, both queues and idle time will occur.

If the operation has too few servers (that is, capacity is set at too low a level), queues will build up to a level where customers become dissatisfied with the time they are having to wait, although the utilization level of the servers will be high. If too many servers are in place (that is, capacity is set at too high a level), the time which customers can expect to wait will not be long but the utilization of the servers will be low. This is why the capacity planning and control problem for this type of operation is often presented as a trade-off between customer waiting time and system utilization. What is certainly important in making capacity decisions is being able to predict both of these factors for a given queuing system. The supplement to this chapter details some of the more simple mathematical approaches to understanding queue behaviour.

customer perceptions of queuing Few of us like waiting. Yet queuing is something we all have to do. So if you have ever wondered if you are alone in particularly hating queuing, you are not – it is official. According to research involving 45,000 iPhone users who provided regular updates on

M11_SLAC8678_08_SE_C11.indd 380 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 381

their level of happiness via an app, it is one of the activities that most upsets us. 8 In fact, of all the things that make us feel unhappy, queuing is beaten only by being in bed sick. Yet, an important aspect of how people judge the service they receive from a queuing system is how they perceive the time spent queuing. It is well known that if you are told that you will be waiting in a queue for 20 minutes and you are actually serviced in 10 minutes, your perception of the queuing experience will be more positive than if you were told that you would be waiting 10 minutes but the queue actually took 20 min- utes. Because of this, the management of queuing systems usually involves attempting to manage customers’ perceptions and expec- tations in some way. Below is a set of ‘principles’ that can help in evaluating and improving queues (of course, in cases where the queue itself cannot be removed through process improvement):

1. Unoccupied time feels longer than occupied time. 2. Pre-process waits feel longer than in-process waits. 3. Anxiety makes the wait seem longer. 4. Uncertain waits feel longer than known, finite waits. 5. Unexplained waits feel longer than explained waits. 6. Unfair waits feel longer than equitable waits. 7. The more valuable the service, the longer customers will ‘happily’ wait. 8. Solo waiting feels longer than group waiting. 9. Uncomfortable waits feel longer than comfortable waits. 10. New or infrequent users feel they wait longer than frequent users.

considering capacity de.5cisions over time Our emphasis so far has been on the planning aspects of capacity management. In prac- tice, capacity management is a far more dynamic process, which involves controlling and reacting to actual demand and actual capacity as it occurs. The capacity control process can be seen as a sequence of partially reactive capacity decisions. At the beginning of each period, operations management considers its forecasts of demand, its understanding of current capacity and, if appropriate, how much inventory has been carried forward from the previous period. Based on all this information, it makes plans for the following peri- od’s capacity. During the next period, demand might or might not be as forecast and the actual capacity of the operation might or might not turn out as planned. But whatever the actual conditions during that period, at the beginning of the next period the same types of decisions must be made, in the light of the new circumstances.

The success of capacity management is generally measured by some combination of costs, revenue, working capital and cus- tomer satisfaction (which goes on to inf luence revenue). This is inf luenced by the actual capacity available to the operation in any period and the demand for that period. However, capacity man- agement is essentially a for ward-looking activity. Overriding other considerations of what capacity strategy to adopt is usually the difference between the long- and short-term outlook for the volume of demand. If the long-term outlook for demand is ‘good’ (in the sense that it is higher than current capacity can cope with) then it is unlikely that even ‘poor’ (demand less than capacity) short-term demand would cause an operation to make large, or difficult to reverse, cuts in capacity. Conversely if long-term outlook for demand is ‘poor’ (in the sense that it is lower than current capacity) then it is unlikely that even ‘good’ (demand more than capacity) short-term demand would cause an operation to take on large, or difficult to reverse, extra capacity. Figure 11.16 illustrates some appro- priate capacity management strategies depending on the comparison of long- and short- term outlooks.

✽ operations principle Customer reactions to having to queue will be influenced by more factors than waiting time.

✽ operations principle the learning from managing capacity in practice should be captured and used to refine both demand forecasting and capacity planning.

M11_SLAC8678_08_SE_C11.indd 381 06/02/16 7:20 PM

382 Part tHrEE DELIVER

Figure 11.16 capacity management strategies are partly dependent on the long- and short- term outlook for volumes

Short-term outlook for volume

Decreasing below current capacity

Level with current capacity

Increasing above current capacity

Reduce capacity (semi) permanently. For example, reduce sta�ng levels; reduce supply agreements.

Plan to reduce capacity (semi) permanently. For example, freeze recruitment; modify supply agreements.

Increase capacity temporarily. For example, increase working hours, and/or hire temporary sta­; modify supply agreements.

Increase capacity temporarily. For example, increase working hours, and/or hire temporary sta­; modify supply agreements.

Maintain capacity at current level.

Reduce capacity temporarily. For example, reduce sta� working hours; modify supply agreements.

L o

n g

-t e

rm o

u tl

o o

k f

o r

vo lu

m e

Increase capacity (semi) permanently. For example, hire sta�; increase supply agreements.

Reduce capacity temporarily. For example, reduce sta� working hours, but plan to recruit; modify supply agreements.

Plan to increase capacity above current level; plan to increase supply agreements.

D ec

re as

in g

b el

o w

cu rr

en t

ca p

ac ity

L ev

el w

ith c

u rr

en t

ca p

ac ity

In cr

ea si

n g

a b

o ve

cu rr

en t

ca p

ac ity

● the capacity of an operation is the maximum level of value-added activity over a period of time that the process can achieve under normal operating conditions.

● Capacity management is the activity of understanding the nature of demand for products and services, and eff ectively planning and controlling capacity in the short term, medium term and long term.

● Long-term capacity management (or strategy) focuses on introducing or deleting major increments of capacity (see Chapter 5 ) . medium- and short-term capacity management focuses on adjusting capacity and demand within the constraints imposed by long-term capacity decisions.

● the process of managing capacity involves (1) measuring and understanding changes in aggregate demand and capacity (supply); (2) determining the operation’s base level of capacity; (3) identifying and selecting methods of coping with demand–supply mismatches; and (4) understanding the consequences of diff erent capacity decisions.

❯ what is capacity management?

summary answErs to KEy quEstions

M11_SLAC8678_08_SE_C11.indd 382 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 383

❯ How are demand and capacity measured?

● Demand forecasts should be expressed in terms that are useful (for example, units per hour, operatives per month, etc.), be as accurate as possible and give an indication of uncertainty.

● typically, the demand for products and services is not completely stable. Climatic, social, cultural, political and economic factors all act to influence both predictable and unpredict- able volatility in demand.

● Capacity can be measured by the availability of its input resources or by the output that is created. Which of these measures is used partly depends on how stable is the mix of outputs. If it is difficult to aggregate the different types of output from an operation, input measures are usually preferred.

● It is managed either by the availability of its input resources or by the output which is pro- duced. Which of these measures is used partly depends on how stable is the mix of outputs. If it is difficult to aggregate the different types of output from an operation, input measures are usually preferred.

● the usage of capacity is measured by the factors of ‘utilization’ and ‘efficiency ’. a useful measure of capacity leakage is overall operations effectiveness (OEE).

❯ How should the operation’s base capacity be set?

● Capacity planning often involves setting a base level of capacity and then planning capac- ity fluctuations around it. the level at which base capacity is set depends on three main factors: the relative importance of the operation’s performance objectives, the perishability of the operation’s outputs, and the degree of variability in demand or supply.

● High service levels, high perishability of an operation’s outputs and a high degree of varia- bility, either in demand or supply, all indicate a relatively high level of base capacity.

❯ what are the ways of coping with mismatches between demand and capacity?

● Demand–capacity mismatches usually call for some degree of capacity adjustment over time. there are three pure methods of achieving this, although in practice a mixture of all three may be used:

● ‘Level capacity’ plans involve no change in capacity and require that the operation absorb demand–capacity mismatches, usually through under- or over-utilization of its resources, or the use of inventory.

● ‘Chase demand’ plans involve the changing of capacity through such methods as over- time, varying the size of the work force, subcontracting, etc.

● ‘Demand management’ plans involve an attempt to change demand through pricing or promotion methods, or changing product or service mix to reduce fluctuations in activ- ity levels. When outputs cannot be stored, yield management is a common method of coping with mismatches.

❯ How can operations understand the consequences of their capacity decisions?

● presenting demand and output in the form of cumulative representations allows the feasi- bility of alternative capacity plans to be assessed.

● In many operations, especially service operations, a queuing approach can be used to explore the consequences of capacity strategies..

● Using long-term and short-term outlook for demand allows further evaluation of alternative capacity management decisions.

M11_SLAC8678_08_SE_C11.indd 383 06/02/16 7:20 PM

384 Part tHrEE DELIVER

casE study blackberry Hill Farm

‘ Six years ago I had never heard of agri-tourism. As far as I was concerned, I had inherited the farm and I would be a farmer all my life .’ ( Jim Walker, Blackberry Hill Farm)

the ‘agri-tourism’ that Jim was referring to is ‘a commer- cial enterprise at a working farm, or other agricultural cen- tre, conducted for the enjoyment of visitors that generates supplemental income for the owner’. ‘ Farming has become a tough business ’, says Jim. ‘ Low world prices, a reduction in subsidies, and increasingly uncertain weather patterns have made it a far more risky business than when I first inherited the farm. Yet, because of our move into the tourist trade we are flourishing. Also … I’ve never had so much fun in my life .’ But, Jim warns, agri-tourism is not for everyone. ‘ You have to think carefully. Do you really want to do it? What kind of life style do you want? How open-minded are you to new ideas? How business-minded are you? Are you willing to put a lot of effort into marketing your business? Above all, do you like working with people? If you had rather be around cows than people, it isn’t the business for you .’

History Blackberry Hill Farm was a 200-hectare mixed farm in the south of England when Jim and mandy Walker inher- ited it 15 years ago. It was primarily a cereal growing operation with a small dairy herd, some fruit and veg- etable growing and mixed woodland that was protected by local preservation laws. Six years ago it had become evident to Jim and mandy that they may have to rethink how the farm was being managed. ‘ We first started a pick-your-own (PYO) operation because our farm is close to several large centres of population. Also the quantities of fruit and vegetables that we were producing were not large enough to interest the commercial buyers. Entering the PYO market was a reasonable success and in spite of mak- ing some early mistakes, it turned our fruit and vegetable growing operation from making a small loss to making a small profit. Most importantly, it gave us some experience of how to deal with customers face-to-face and of how to cope with unpredictable demand. The biggest variable in PYO sales is weather. Most business occurs at the week- ends between late spring and early autumn. If rain keeps customers away during part of those weekends, nearly all sales have to occur in just a few days .’

Within a year of opening up the pyO operation Jim and mandy had decided to reduce the area devoted to cereals and increase their fruit and vegetable growing capability. at the same time they organized a petting zoo that allowed children to mix with, feed and touch various animals.

‘ We already had our own cattle and poultry but we extended the area and brought in pigs and goats. Later we also introduced some rabbits, ponies and donkeys, and even

a small bee keeping operation .’ at the same time the farm started building up its collection of ‘farm heritage’ exhib- its. these were static displays of old farm implements and ‘recreations’ of farming processes together with information displays. this had always been a personal interest of Jim’s and it allowed him to convert two existing farm outbuild- ings to create a ‘museum of Farming Heritage’.

the year after, they introduced tractor rides for vis- itors around the whole farm and extended the petting zoo and farming tradition exhibits further. But the most significant investment was in the ‘preserving kitchen’. ‘ We had been looking for some way of using the surplus fruits and vegetable that we occasionally accumulated and also for some kind of products that we could sell in a farm shop. We started the Preserving Kitchen to make jams and fruit, vegetables and sauces preserved in jars. The venture was an immediate success. We started making just fifty kilo- grammes of preserves a week , within three months that had grown to three hundred kilogrammes a week and we are now producing around a thousand kilogrammes a week , all under the “Blackberry Hill Farm” label .’ the following year the preserving kitchen was extended and a viewing area added. ‘ It was a great attraction from the beginning ’, says mandy. ‘ We employed ladies from the local village to make the preserves. They are all extrovert characters, so when we asked them to dress up in tradi- tional “farmers wives” type clothing they were happy to do it. The visitors love it, especially the good natured repartee with our ladies. The ladies also enjoy giving informal his- tory lessons when we get school parties visiting us .’

So u

rc e:

S h

u tt

er st

o ck

.c o

m : K

ar el

g al

la s

M11_SLAC8678_08_SE_C11.indd 384 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 385

Within the last two years the farm had further extended its preserving kitchen, farm shop, exhibits and petting zoo. It had also introduced a small adventure playground for the children, a café serving drinks and its own produce, a pic- nic area and a small bakery. the bakery was also open to view by customers and staffed by bakers in traditional dress. ‘It’s a nice little visitor attraction’, says mandy, ‘and it gives us another opportunity to squeeze more value out of our own products.’ table 11.4(a) shows last year ’s visitor numbers, and table 11.4(b) shows the farm’s opening times..

demand the number of visitors to the farm was extremely sea- sonal. From a low point in January and February, when most people just visited the farm shop, the spring and summer months could be very busy, especially on public holidays. the previous year mandy had tracked the num- ber of visitors arriving at the farm each day. ‘It is easy to

record the number of people visiting the farm attractions, because they pay the entrance charge. What we had not done before is include the people who just visited the farm shop and bakery that can be accessed both from within the farm and from the car park. We estimate that the number of people visiting the shop but not the farm ranges from 74 per cent in February down to around 15 per cent in August.’ Figure 11.17 shows the number of visitors in the previous year ’s august. ‘What our figures do not include are those people who visit the shop but don’t buy anything. This is unlikely to be a large number.’ Figure 11.18 shows visitor arrivals on a public holiday in august and a Wednesday in February.

mandy had also estimated the average stay at the farm and/or farm shop. She reckoned that in winter time the aver- age stay was 45 minutes, but in august in climbed to 3.1 hours.

current issues Both Jim and mandy agreed that their lives had funda- mentally changed over the last few years. Income from visitors and from the Blackberry Hill brand of preserves now accounted for 70 per cent of the farm’s revenue. more importantly, the whole enterprise was significantly

table 11.4(a) number of visitors last year

month total visitors

January February march april may June July august September October november December total average

1,006 971

2,874 6,622 8,905

12,304 14,484 15,023 12,938

6,687 2,505 3,777

88,096 7,341.33

table 11.4(b) Farm opening times*

January–mid-march Wednesday–Sunday 10:00–16:00

mid-march–may tuesday–Sunday 09:00–18:00

may–September all week 08:30–19:00

October–november tuesday–Sunday 10:00–16:00

December tuesday–Sunday 09:00–18:00

*Special evening events Easter, summer weekends and Christmas.

Figure 11.17 daily number of visitors in august last year

1 5 201510 3025

500

1000

1500

M11_SLAC8678_08_SE_C11.indd 385 06/02/16 7:20 PM

386 Part tHrEE DELIVER

more profitable than it had ever been. nevertheless, the farm faced a number of issues.

the first was the balance between its different activities. Jim was particularly concerned that the business remained a genuine farm. ‘When you look at the revenue per hectare, visitor and production activities bring in far more revenue than conventional agricultural activities. However, if we push the agri-tourism too far we become no better than a theme park. We represent something more than this to our visitors. They come to us partly because of what we represent as well as what we actually do. I am not sure that we would want to grow much more. Anyway, more visitors would mean that we have to extend the car park. That would be expensive, and although it would be necessary, it does not directly bring in any more revenue. There are already parking problems during peak periods and we have had complaints from the police that our visitors park inappropriately on local roads.

‘There is also the problem of complexity. Every time we introduce a new attraction, the whole business gets that little bit more complex to manage. Although we enjoy it tremen- dously, both Mandy and I are spreading ourselves thinly over an ever widening range of activities [mandy was also con- cerned over this]. I’m starting to feel that my time is being taken up in managing the day-to-day problems of the busi- ness. This does not leave time either for thinking about the overall direction in which we should be going, or spending time talking with the staff. That is why we both see this com- ing year as a time for consolidation and for smoothing out the day-to-day problems of managing the business, particularly the queuing, which is getting excessive at busy times. That is why this year we are limiting ourselves to just one new ven- ture for the business.’

Staff management was also a concern for mandy. the business had grown to over 80 (almost all part-time and seasonal) employees. ‘We have become a significant

employer in the area. Most of our employees are still local people working part-time for extra income but we are also now employing 20 students during the summer period and, last year, 8 agricultural students from Eastern Europe. But now, labour is short in this part of the country and it is becoming more difficult to attract local people, especially to produce Blackberry Hill Farm Preserves. Half of the Preserving Kitchen staff work all year, with the other employed during the summer and autumn periods. But most of them would prefer guaranteed employment throughout the year'

table 11.5 gives more details of some of the issues of managing the facilities at the farm, and table 11.6 shows the preserve demand and production for the previous year.

where next? By the ‘consolidation’ and improvement of ‘day-to-day ’ activities Jim and mandy meant that they wanted to increase their revenue, while at the same time reducing the occasional queues that they knew could irritate their visitors, preferably without any significant investment in extra capacity. they also were concerned to be able to offer more stable employment to the preserving kitchen ‘ladies’ throughout the year, who would produce at a near constant rate. However, they were not sure if this could be done without storing the products for so long that their shelf life would be seriously affected. there was no prob- lem with the supply of produce to keep production level – less than 2 per cent of the fruit and vegetables that go into preserves are actually grown on the farm. the remainder were bought at wholesale markets, although this was not generally understood by customers.

Of the many ideas being discussed as candidates for the ‘one new venture’ for next year, two were emerging as particularly attractive. Jim liked the idea of develop- ing a maize maze, a type of attraction that had become

Figure 11.18 Visitor arrivals, public holiday in august and a wednesday in February

8.00

100

200

10.00 12.00 14.00 16.00 18.00 20.00

300

400

Public holiday in August

Wednesday in February

M11_SLAC8678_08_SE_C11.indd 386 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 387

table 11.5 the farm’s main facilities and some of the issues concerned with managing them

Facility issues

car park ● 85 car parking spaces, 4 * 40-seater tour bus spaces

Fixed exhibits etc. Recreation of old farmhouse kitchen, recreation of barnyard, old-fashioned milking parlour, various small exhibits on farming past and present, adventure playground, ice cream and snack stands

● most exhibits in, or adjacent to, the farm museum ● at peak times have helpers dressed in period costume to entertain

visitors ● Feedback indicates customers find exhibits more interesting than they

thought they would ● Visitors free to look when they wish absorbs demand from busy facilities

tractor rides One tractor towing decorated covered cart with maximum capacity of 30 people, tour takes around 20 minutes on average (including stops). Waits 10 minutes between tours except at peak times when tractor circulates continuously

● tractor acts both as transport and entertainment, approximately 60% of visitors stay on for the whole tour, 40% use it as ‘hop-on, hop-off ’ facility

● Overloaded at peak times, long queues building ● Feedback indicates it is popular, except for queuing ● Jim reluctant to invest in further cart and tractor

Pick-your-own area Largest single facility on the farm. Use local press, dedicated telephone line (answering machine) and website to communicate availability of fruit and vegetables. Checkout and weighing area next to farm shop, also displays picked produce, preserves, etc., for sale

● Very seasonal and weather dependent, both for supply and demand ● Farm plans for a surplus over visitor demand, uses surplus in preserves ● Six weighing/paying stations at undercover checkout area ● Queues develop at peak times. Feedback indicates some

dissatisfaction with this ● Can move staff from farm shop to help with checkout in busy periods,

but farm shop also tends to be busy at the same time ● Considering using packers at pay stations to speed up the process

Petting zoo accommodation for smaller animals including sheep and pigs. Large animals (cattle, horses) brought to viewing area daily. Visitors can view all animals and handle/stroke most animals under supervision

● approximately 50% of visitors view petting zoo ● number of staff in attendance varies between 0 (off-peak) and 5

(peak periods) ● the area can get congested during peak periods ● Staff need to be skilled at managing children

Preserving kitchen Boiling vats, mixing vats, jar sterilizing equipment, etc. Visitor viewing area can hold 15 people comfortably. average length of stay 7 minutes in off-season, 14 minutes in peak season

● Capacity of kitchen is theoretically 4,500 kilograms per month on a 5-day week and 6,000 kilograms on a 7-day week

● In practice, capacity varies with season because of interaction with visitors. Can be as low as 5,000 kilograms on a 7-day week in summer, or up to 5,000 kilograms on a 5-day week in winter

● Shelf life of products is on average 12 months ● Current storage area can hold 16,000 kilograms

bakery Contains mixing and shaping equipment, commercial oven, cooling racks, display stand, etc. Just installed doughnut making machine. all pastries contain farm’s preserved fruit

● Starting to become a bottleneck since doughnut making machine installed – visitors like watching it

● products also on sale at farm shop adjacent to bakery ● Would be difficult to expand this area because of building constraints

Farm shop and café Started by selling farm’s own products exclusively. now sells a range of products from farms in the region and wider. Started selling frozen menu dishes (lasagne, goulash, etc.) produced off-peak in the preserving kitchen

● the most profitable part of the whole enterprise, Jim and mandy would like to extend the retailing and café operation

● Shop includes area for cooking displays, cake decoration, fruit dipping (in chocolate), etc.

● Some congestion in shop at peak times but little visitor dissatisfaction ● more significant queuing for café in peak periods ● Considering allowing customers to place orders before they tour the

farm’s facilities and collect their purchases later ● Retailing more profitable per square metre than café

M11_SLAC8678_08_SE_C11.indd 387 06/02/16 7:20 PM

388 Part tHrEE DELIVER

table 11.6 Preserve demand and production (previous year)

month demand (kg) cumulative demand (kg) Production (kg) cumulative product (kg) inventory (kg)

January February march april may June July august September October november December average demand

682 794

1,106 3,444 4,560 6,014 9,870

13,616 5,040 1,993 2,652 6,148 4,660

682 1,476 2,582 6.026

10,586 16,600 26,470 40,086 45,126 47,119 49,771 55,919

4,900 4,620 4,870 5,590 5,840 5,730 5,710 5,910 5,730 1,570* 2,770* 4,560

4,900 9,520

14,390 19,980 25,820 31,550 37,260 43,170 48,900 50,470 53,240 57,800

average inventory

4,218 8,044

11,808 13,954 15,234 14,950 10,790

3,084 3,774 3,351 3,467 1,881 7,880

*technical problems reduced production level.

increasingly popular in Europe and north america in the last five years. It involved planting a field of maize (corn) and, once grown, cutting through a complex serious of paths in the form of a maze. Evidence from other farms indicated that a maze would be extremely attractive to visitors and Jim reckoned that it could account for up to an extra 10,000 visitors during the summer period. Designed as a separate activity with its own admis- sion charge, it would require an investment of around £20,000, but generate more than twice that in admis- sion charges as well as attracting more visitors to the farm itself.

mandy favoured the alterative idea – that of building up their business in organized school visits. ‘ Last year we joined the National Association of Farms for Schools. Their advice is that we could easily become one of the top school attractions in

this part of England. Educating visitors about farming tradition is already a major part of what we do. And many of our staff have developed the skills to communicate to children exactly what farm life used to be like. We would need to convert and extend one of our existing underused farm outbuildings to make a ‘school room’ and that would cost between and £30,000 and £35,000. And although we would need to discount our admis- sion charge substantially, I think we could break even on the investment within around two years .’

quEstions 1 How could the farm’s day-to-day operations be

improved?

2 what advice would you give Jim and mandy regarding this year ’s ‘new venture’?

a local government office issues hunting licences. Demand for these licences is relatively slow in the first part of the year but then increases after the middle of the year before slowing down again towards the end of the year. the department works a 220-day year on a basis of 5 days a week. Between working days 0 and 100, demand is 25 per cent of demand during the peak period that lasts between day 100 and day 150. after day 150, demand reduces to about 12 per cent of the demand during the peak period. In total, the department processes 10,000 applications per year. the department has two permanent members of staff who are capable of processing 15 licence applications per day. If an untrained temporary member of

1

ProbLEms and aPPLications

M11_SLAC8678_08_SE_C11.indd 388 06/02/16 7:20 PM

cHaPtEr 11 CapaCIty managEmEnt 389

staff can only process 10 licences per day, how many temporary staff should the department recruit between days 100 and 150?

In the problem above, if a new computer system is installed that allows experienced staff to increase their work rate to 20 applications per day, and untrained staff to 15 applications per day, (a) does the department still need two permanent staff, and (b) how many temporary members of staff will be needed between days 100 and 150?

a field service organization repairs and maintains printing equipment for a large number of customers. It offers one level of service to all its customers and employs 30 staff. the oper- ations marketing ’s vice-president has decided that in future the company will offer three standards of service: platinum, gold and silver. It is estimated that platinum service customers will require 50 per cent more time from the company ’s field service engineers than the cur- rent service. the current service is to be called ‘the gold service’. the silver service is likely to require about 80 per cent of the time of the gold service. If future demand is estimated to be 20 per cent platinum, 70 per cent gold and 10 per cent silver service, how many staff will be needed to fulfil demand?

Look again at the principles that govern customers’ perceptions of the queuing experience. For the following operations, apply the principles to minimize the perceived negative effects of queuing: (a) a cinema (b) a doctor’s surgery (c) Waiting to board an aircraft.

Consider how airlines cope with balancing capacity and demand. In particular, consider the role of yield management. Do this by visiting the website of a low-cost airline, and for a number of flights price the fare that is being charged by the airline from tomorrow onwards. In other words, how much would it cost if you needed to fly tomorrow, how much if you needed to fly next week, how much if you needed to fly in 2 weeks, etc.? plot the results for different flights and debate the findings.

Calculate the overall equipment efficiency (OEE) of the following facilities by investigating their use: (a) a lecture theatre (b) a cinema (c) a coffee machine.

Discuss whether it is worth trying to increase the OEE of these facilities and, if it is, how you would go about it.

2

3

4

5

6

sELEctEd FurtHEr rEading

gunther, n. J. (2007) Guerrilla Capacity Planning, springer, new york.

this book provides a tactical approach for planning capacity in both product-based and ser- vice-based contexts. particularly interesting for those new to the ideas of capacity planning as it cov- ers basic and more advanced demand forecasting techniques as well as ‘classic’ capacity responses.

Hansen, r.c. (2005) Overall Equipment Effectiveness (OEE), industrial Press, south norwalk, ct.

If you want to know more about OEE, its origins and application, this is the place to start.

M11_SLAC8678_08_SE_C11.indd 389 06/02/16 7:20 PM

390 Part tHrEE DELIVER

Van mieghem, J. (2003) capacity management, investment, and hedging: review and recent developments, Manufacturing and Service Operations Management, vol. 5, no. 4, 269–302.

an academic article reviewing the literature on strategic capacity management. it does a nice job of covering the different approaches to capacity management under conditions of stability versus volatility (demand change) and of certainty versus uncertainty (that is, the predictability of change).

M11_SLAC8678_08_SE_C11.indd 390 06/02/16 7:20 PM

introduction In the main part of Chapter 11 we described how the queuing approach (in the USa it would be called the ‘waiting line approach’) can be useful in thinking about capacity, especially in service operations. It is useful because it deals with the issue of variability, both of the arrival of customers (or items) at a process and of how long each customer (or item) takes to process. and where variability is present in a process (as it is in most processes, but particularly in service processes) the capacity required by an opera- tion cannot easily be based on averages but must include the effects of the variation. Unfortunately, many of the formulae that can be used to understand queuing are extremely complicated, especially for complex systems, and are beyond the scope of this book. In fact, computer programs are almost always now used to predict the behaviour of queuing systems. However, studying queuing formulae can illustrate some useful char- acteristics of the way queuing systems behave.

notation

Unfortunately there are several different conventions for the notation used for different aspects of queuing system behaviour. It is always advisable to check the notation used by different authors before using their formulae. We will use the following notation:

ta = average time between arrival ra = arrival rate 1items per unit time2 = 1>ta ca = coefficient of variation of arrival times m = number of parallel servers at a station te = mean processing time re = processing rate 1items per unit time2 = m>te ce = coefficient of variation of process time u = utilization of station = ra>re = 1rate2>m

WIP = average work-in-progress (number of items) in the queue WIPq = expected work-in-progress (number of times) in the queue

Wq = expected waiting time in the queue W = expected waiting time in the system 1queue time + processing time2

Some of these factors are explained later.

VariabiLity

The concept of variability is central to understanding the behaviour of queues. If there were no variability there would be no need for queues to occur because the capacity of a process could be relatively easily adjusted to match demand. For example, suppose one member of staff (a server) serves customers at a bank counter who always arrive exactly every five minutes

Supplement to Chapter 11 analytical queuing models

suPPLEmEnt to cHaPtEr 11 anaLytICaL QUEUIng mODELS 391

M11_SLAC8678_08_SE_C11.indd 391 06/02/16 7:20 PM

392 Part tHrEE DELIVER

(that is, 12 per hour). Also suppose that every customer takes exactly five minutes to be served. Then because

(a) the arrival rate is less than or equal to the processing rate, and (b) there is no variation

no customer need ever wait because the next customer will arrive when, or before, the previ- ous customer leaves. That is, WIPq = 0.

Also, in this case, the server is working all the time, again because exactly as one customer leaves the next one is arriving. That is, u = 1.

Even with more than one server, the same may apply. For example, if the arrival time at the counter is five minutes (12 per hour) and the processing time for each customer is now always exactly 10 minutes, the counter would need two servers, and because

(a) arrival rate is less than or equal to processing rate m, and (b) there is no variation

again WIPq = 0 and u = 1. Of course, it is convenient (but unusual) if arrival rate/processing rate = a whole number.

When this is not the case (for this simple example with no variation)

Utilization = Processing rate>1Arrival rate * m2 For example:

if arrival rate, ra = 5 minutes processing rate, re = 8 minutes

number of servers, m = 2 then utilization, u = 8>15 * 22 = 0.8 or 80%

incorporating variability The previous examples were not realistic because the assumption of no variation in arrival or processing times very rarely occurs. We can calculate the average or mean arrival and process times but we also need to take into account the variation around these means. To do that we need to use a probability distribution. Figure S11.1 contrasts two processes with different arrival dis- tributions. The units arriving are shown as people, but they could be jobs arriving at a machine, trucks needing servicing, or any other uncertain event. The upper example shows low variation in arrival time where customers arrive in a relatively predictable manner. The lower example has the same average number of customer arriving but this time they arrive unpredictably with sometimes long gaps between arrivals and at other times two or three customers arriving close together. Of course, we could do a similar analysis to describe processing times. Again, some would have low variation, some higher variation, and others somewhere in between.

In Figure S11.1 high arrival variation has a distribution with a wider spread (called ‘dispersion’) than the distribution describing lower variability. Statistically the usual measure

Figure s11.1 Low and high arrival variation

M11_SLAC8678_08_SE_C11.indd 392 06/02/16 7:20 PM

for indicating the spread of a distribution is its standard deviation, s. But variation does not just depend on standard deviation. For example, a distribution of arrival times may have a standard deviation of 2 minutes. This could indicate very little variation when the average arrival time is 60 minutes. But it would mean a very high degree of variation when the average arrival time is 3 minutes. Therefore to normalize standard deviation, it is divided by the mean of its distribution. This measure is called the coefficient of variation of the distribution. So:

ca = coefficient of variation of arrival times = sa>ta ce = coefficient of variation of processing times = se>te

incorPorating LittLE’s Law

In Chapter 6 we discussed one of the fundamental laws of processes that describes the rela- tionship between the cycle time of a process (how often something emerges from the pro- cess), the work-in-progress in the process and the throughput time of the process (the total time it takes for an item to move through the whole process including waiting time). It was called Little’s law and it was denoted by the following simple relationship:

Throughput time = Throughout time/Cycle time

Therefore: Work-in-progress = Throughout time/Cycle time

or: WIP = T>C

We can make use of Little’s law to help understand queuing behaviour. Consider the queue in front of a station.

Work-in-progress in the queue = Arrival rate at the queue (equivalent to 1/cycle time) * Waiting time in the queue (equivalent to throughput time)

WIPq = ra * Wq and:

Waiting time in the whole system = Waiting time in the queue + Average process time at the station

W = Wq + te We will use this relationship later to investigate queuing behaviour.

tyPEs oF quEuing systEm

Conventionally queuing systems are characterized by four parameters:

A – the distribution of arrival times (or more properly inter-arrival times, the elapsed times between arrivals)

B – the distribution of process times

m – the number of servers at each station

b – the maximum number of items allowed in the system.

The most common distributions used to describe A or B are either:

(a) the exponential (or Markovian) distribution denoted by M; or (b) the general (for example, normal) distribution denoted by G.

So, for example, an M/G/1/5 queuing system would indicate a system with exponentially distributed arrivals, process times described by a general distribution such as a normal distri- bution, with one server and a maximum number of items allowed in the system of five. This type of notation is called Kendall’s notation.

suPPLEmEnt to cHaPtEr 11 anaLytICaL QUEUIng mODELS 393

M11_SLAC8678_08_SE_C11.indd 393 06/02/16 7:20 PM

394 Part tHrEE DELIVER

Queuing theory can help us investigate any type of queuing system, but in order to simplify the mathematics, we will deal here only with the two most common situations. Namely,

M/M/m – the exponential arrival and processing times with m servers and no maximum limit to the queue.

G/G/m – the general arrival and processing distributions with m servers and no limit to the queue.

First we will start by looking at the simple case when m = 1.

For m/m/1 queuing systems The formulae for this type of system are as follows:

WIP = u

1 - u Using Little’s law:

WIP = Cycle time * Throughput time Throughput time = WIP>Cycle time

Then:

Throughput time = u

1 - u *

1 ra

= te

1 - u and since:

Throughput time in the queue = Total throughput time - Average processing time

then: Wq = W - te

= te

1 - u - te

= te - te11 - u2

1 - u =

te - te - ute 1 - u

= u

11 - u2te Again, using Little’s law:

WIPq = ra * Wq = u

11 - u2tera and since:

u = ra re

= rate

ra = u te

then:

WIPq = u

11 - u2 * te * u te

= u2

11 - u2

For m/m/m systems When there are m servers at a station the formula for waiting time in the queue (and there- fore all other formulae) needs to be modified. Again, we will not derive these formulae but just state them:

Wq = u221m + 12- 1 m11 - u2 te

From which the other formulae can be derived as before.

M11_SLAC8678_08_SE_C11.indd 394 06/02/16 7:20 PM

For G/G/1 systems The assumption of exponential arrival and processing times is convenient as far as the math- ematical derivation of various formulae are concerned. However, in practice, process times in particular are rarely truly exponential. This is why it is important to have some idea of how a G/G/1 and G/G/ m queue behaves. However, exact mathematical relationships are not possi- ble with such distributions. Therefore some kind of approximation is needed. The one here is in common use, and although it is not always accurate, it is for practical purposes. For G/G/1 systems the formula for waiting time in the queue is as follows:

Wq = a c2a + c2e

2 b u11 - u2te

There are two points to make about this equation. The first is that it is exactly the same as the equivalent equation for an M/M/1 system but with a factor to take account of the variability of the arrival and process times. The second is that this formula is sometimes known as the VUT formula because it describes the waiting time in a queue as a function of:

V – the variability in the queuing system;

U – the utilization of the queuing system (that is, demand versus capacity); and

T – the processing times at the station.

In other words, we can reach the intuitive conclusion that queuing time will increase as vari- ability, utilization or processing time increases.

For G/G/m systems The same modification applies to queuing systems using general equations and m servers. The formula for waiting time in the queue is now as follows:

Wq = a c2a + c2e

2 bau

221m + 12- 1 m11 - u2 bte

Worked example 1

‘ I can’t understand it. We have worked out our capacity figures and I am sure that one member of staff should be able to cope with the demand. We know that customers arrive at a rate of around 6 per hour and we also know that any trained member of staff can process them at a rate of 8 per hour. So why is the queue so large and the wait so long? Have at look at what is going on there please .’

Sarah knew that it was probably the variation, both in customers arriving and in how long it took each of them to be processed, that was causing the problem. Over a two-day period when she was told that demand was more or less normal, she timed the exact arrival times and processing times of every customer. Her results were as follows:

The coefficient of variation, ca, of customer arrivals = 1

The coefficient of variation, ce, of processing time = 3.5

The average arrival rate of customers, ra = 6 per hour

therefore the average inter-arrival time = 10 minutes The average processing rate, re = 8 per hour

therefore the average processing time = 7.5 minutes

Thus the utilization of the single server, u = 6>8 = 0.75 Using the waiting time formula for a G/G/1 queuing system:

▼

SUPPLEMENT TO CHAPTER 11 ANALYTICAL QUEUING MODELS 395

M11_SLAC8678_08_SE_C11.indd 395 06/02/16 7:44 PM

396 Part tHrEE DELIVER

Wq = a 1 + 12.25

2 ba 0.75

1 - 0.75 b7.5

= 6.625 * 3 * 7.5 = 149.06 minutes

= 2.48 hours

also because:

WIpq = Cycle time * throughput time = 6 * 2.48 = 14.68

So, Sarah had found out that the average wait that customers could expect was 2.48 hours and that there would be an average of 14.68 people in the queue.

‘ Ok, so I see that it’s the very high variation in the processing time that is causing the queue to build up. How about investing in a new computer system that would standardize processing time to a greater degree? I have been talking with our technical people and they reckon that, if we invested in a new system, we could cut the coefficient of variation of processing time down to 1.5. What kind of a difference would this make? ’

Under these conditions with c e = 1.5:

Wq = a 1 + 2.25

2 ba 0.75

1 - 0.75 b7.5

= 1.625 * 3 * 7.5 = 36.56 minutes

= 0.61 hours

therefore: WIpq = 6 * 0.61 = 3.66

In other words, reducing the variation of the process time has reduced average queuing time from 2.48 hours down to 0.61 hours and has reduced the expected number of people in the queue from 14.68 down to 3.66.

worked example 2

a bank wishes to decide how many staff to schedule during its lunch period. During this period customers arrive at a rate of nine per hour and the enquiries that customers have (such as opening new accounts, arranging loans, etc.) take on average 15 minutes to deal with. the bank manager feels that four staff should be on duty during this period but wants to make sure that the customers do not wait more than three minutes on average before they are served. the manager has been told by his small daughter that the dis- tributions that describe both arrival and processing times are likely to be exponential. therefore:

ra = 9 per hour, therefore

ta = 6.67 minutes

re = 4 per hour, therefore

te = 15 minutes

the proposed number of servers, m = 4.

▼

M11_SLAC8678_08_SE_C11.indd 396 06/02/16 7:21 PM

Therefore the utilization of the system is:

u = 9>14 * 42 = 0.5625 From the formula for waiting time for a M/M/ m system:

Wq = u221m + 12- 1 m11 - u2 te

= 0.5625210 - 1

411 - 0.56252 * 0.25

= 0.56252.162

1.75 * 0.25

= 0.042 hours

= 2 .52 minutes

Therefore the average waiting time with four servers would be 2.52 minutes, that is well within the manager ’s acceptable waiting tolerance.

SUPPLEMENT TO CHAPTER 11 ANALYTICAL QUEUING MODELS 397

M11_SLAC8678_08_SE_C11.indd 397 06/02/16 7:45 PM