Discussion: Business Intelligence (300 words)

dk1913

Ch9andCh10.pdf

Home >Information Systems homework help >Discussion: Business Intelligence (300 words)

CHAPTER

9 Business Intelligence and Analytics

Sergey Nivens/Shutterstock.com

Know?Did Yo u

• MetLife is implementing analytical software to identify medical provider, attorney, and repair shop fraud to aid its special investigations unit (SIU).

• Nearly 20 percent of Medicare patients were readmitted to the hospital within 30 days of their initial discharge, running up an additional $17 billion in healthcare costs. Hospitals are now using BI analytics to identify patients are high risk of readmission—especially now that Medicare has begun reducing payments to hospitals with high readmission rates.

• IBM Watson Analytics services, a cloud-based business analytics tool that offers a variety of tools for uncovering trends hidden in large sets of data, uses baseball statistics on every player in Major League Baseball from AriBall to build predictions of player performance. You can use this service to gain an edge over your fantasy baseball league competitors.

Principles Learning Objectives

• Business intelligence (BI) and analytics are used to support improved decision making.

• Define the terms business intelligence (BI) and analytics.

• Provide several real-world examples of BI and analytics being used to improve decision making.

• Identify the key components that must be in place for an organization to get real value from its BI and analytics efforts.

• There are many BI and analytics techniques and tools that can be used in a wide range of problem- solving situations.

• Identify several BI techniques and discuss how they are used.

• Identify several BI tools.

• Define the term self-service analytics and discuss its pros and cons.

Why Learn about Business Intelligence (BI) and Analytics? We are living in the age of big data, with new data flooding us from all directions at the incomprehensible speed of nearly a zettabyte (1 trillion gigabytes or a 1 followed by 21 zeros) per year. What is most exciting about this data is not its amount, but rather the fact that we are gaining the tools and understanding to do something truly meaningful with it. Organizations are learning to analyze large amounts of data not only to measure past and current performance but also to make predictions about the future. These forecasts will drive anticipatory actions to improve business strategies, strengthen business operations, and enrich decision making—enabling the organization to become more competitive.

A wide range of business users can derive benefits from access to data, but most of them lack deep information systems or data science skills. Business users need easier and faster ways to discover relevant patterns and insights into data to better support their decision making and to make their companies more agile. Companies that have access to the same kind of data as their competitors but can analyze it sooner to take action faster will outpace their peers. Providing BI tools and making business analytics more understandable and accessible to these users should be a key strategy of organizations.

Members of financial services organizations use BI and analytics to better understand their customers to enhance service, create new and more appealing products, and better manage risk. Marketing managers analyze data related to the Web-surfing habits, past purchases, and even social media activity of existing and potential customers to create highly effective marketing programs that generate consumer interest and increased sales. Health care professionals who are able to improve the patient experience will reap the benefits of maximized reimbursements, lower costs, and higher market share, and they will ultimately deliver higher quality care for patients. Physicians use business analytics to analyze data in an attempt to identify factors that lead to readmission of hospital patients. Human resources managers use analytics to evaluate job candidates and choose those most likely to be successful. They also analyze the impact of raises and changes in employee-benefit packages on employee retention and long-term costs.

Regardless of your field of study in school and your future career, using BI and analytics, will likely be a significant component of your job. As you read this chapter, pay attention to how different organizations use business analytics. This chapter starts by introducing basic concepts related to BI and analytics. Later in the chapter, several BI and analytics tools and strategies are discussed.

As you read this chapter, consider the following:

• What is business intelligence (BI) and analytics, and how can they be used to improve the operations and results of an organization?

• What are some business intelligence and analytics techniques and tools, and how can they be used?

This chapter begins with a definition of business intelligence (BI) and busi- ness analytics and the components necessary for a successful BI and analytics program. The chapter goes on to describe and provide examples of the use of several BI techniques and tools. It ends with a discussion of some of the issues associated with BI and analytics.

What Are Analytics and Business Intelligence?

Business analytics is the extensive use of data and quantitative analysis to sup- port fact-based decision making within organizations. Business analytics can be used to gain a better understanding of current business performance, reveal new business patterns and relationships, explain why certain results occurred, optimize current operations, and forecast future business results.

384 PART 3 • Business Information Systems

Business intelligence (BI) includes a wide range of applications, prac- tices, and technologies for the extraction, transformation, integration, visuali- zation, analysis, interpretation, and presentation of data to support improved decision making. The data used in BI is often pulled from multiple sources and may come from sources internal or external to the organization. Many organizations use this data to build large collections of data called data ware- houses, data marts, and data lakes, for use in BI applications. Users, including employees, customers, and authorized suppliers and business partners, may access the data and BI applications via the Web or through organizational intranets and extranets—often using mobile devices, such as smartphones and tablets. The goal of business intelligence is to get the most value out of information and present the results of analysis in an easy to understand man- ner that the layman can understand.

Often the data used in BI and analytics must be gathered from a variety of sources. Helse Vest, a regional health authority in Norway, has 26,500 employees who serve 1 million people in 50 healthcare facilities, including 10 hospitals. Helse Vest implemented a BI system to meet the requirements of a government-sponsored national patient safety program. The system collects, visualizes, and shares medical data used to identify quality measures and reporting requirements across all care teams and regional hospi- tals. A major challenge for the project was the need for each of the 10 hospitals to combine data from all the facilities within its region for analysis by the pro- gram’s board and hospital managers. Prior to implementing the new system, it took up to 14 days for employees to produce some reports, making it difficult for hospital staff to assess and act on performance data because it was not cur- rent. With the new system, Helse Vest analysts can easily combine data from different sources and create analytical reports in less than one day. Real-time data enables Helse Vest to act on information much more quickly, while the metrics are still valid for the staff, and a quick response to performance data is more likely to lead to significant improvements in patient safety measures.1

Benefits Achieved from BI and Analytics BI and analytics are used to achieve a number of benefits as illustrated by the following examples:

● Detect fraud. MetLife implemented analytical software to help its special investigations unit (SIU) identify medical provider, attorney, and repair shop fraud. Although an accident claim may not have enough data to be flagged as suspicious when it is first filed, as more claim data is added, a claim is continually rescored by the software. After the first six months of using the software, the number of claims under investigation by the SIU increased 16 percent.2

● Improve forecasting. Kroger serves customers in 2,422 supermarkets and 1,950 in-store pharmacies. The company found that by better predicting pharmacy customer demand, it could reduce the number of prescriptions that it was unable to fill because a drug is out of stock. To do so, Kroger developed a sophisticated inventory management system that could provide employees with a visualization of inventory levels, adapt to user feedback, and support “what-if” analysis. Out-of-stock pre- scriptions have been reduced by 1.5 million per year, with a resulting increase in sales of $80 million per year. In addition, by carrying the right drugs in the right quantities, Kroger was able to reduce its overall inven- tory costs by $120 million per year.3

● Increase Sales. DaimlerChrysler and many other auto manufacturers set their suggested retail and wholesale prices for the year, then adjust pricing through seasonal incentives based on the impact of supply and demand. DaimlerChrysler implemented a price-elasticity model to

business intelligence (BI): A wide range of applications, practices, and technologies for the extraction, transformation, integration, visualiza- tion, analysis, interpretation, and presentation of data to support improved decision making.

CHAPTER 9 • Business Intelligence and Analytics 385

optimize the company’s pricing decisions. The system enables managers to evaluate many potential incentives for each combination of vehicle model (e.g., Jeep Grand Cherokee), acquisition method (cash, finance, or lease), and incentive program (cash back, promotional APR, and a combination of cash back and promotional APR). The firm estimates that use of the system has generated additional annual sales of $500 million.4

● Optimize operations. Chevron is one of the world’s leading integrated energy companies. Its refineries work with crude oil that is used to make a wide range of oil products, including gasoline, jet fuel, diesel fuel, lubricants, and specialty products such as additives. With market prices of crude oil and its various products constantly changing, determining which products to refine at a given time is quite complex. Chevron uses an analytical system called Petro to aid analysts in advising the refineries and oil traders on the mix of products to produce, buy, and sell in order to maximize profit.5

● Reduce costs. Coca-Cola Enterprises is the world’s largest bottler and distributor of Coca Cola products. Its delivery fleet of 54,000 trucks is second in size to only to the U.S. Postal Service. Using analytics software, the firm implemented a vehicle-routing optimization system that resulted in savings of $45 million a year from reduced gas consumption and reduction in the number of drivers required.6

The Role of a Data Scientist Data scientists are individuals who combine strong business acumen, a deep understanding of analytics, and a healthy appreciation of the limitations of their data, tools, and techniques to deliver real improvements in decision mak- ing. Data scientists do not simply collect and report on data; they view a situa- tion from many angles, determine what data and tools can help further an understanding of the situation, and then apply the appropriate data and tools. They often work in a team setting with business managers and specialists from the business area being studied, market research and financial analysts, data stewards, information system resources, and experts highly knowledgeable about the company’s competitors, markets, products, and services. The goal of the data scientist is to uncover valuable insights that will influence organiza- tional decisions and help the organization to achieve competitive advantage.

Data scientists are highly inquisitive, continually asking questions, performing “what-if” analyses, and challenging assumptions and existing processes. Successful data scientists have an ability to communicate their find- ings to organizational leaders so convincingly that they are able to strongly influence how an organization approaches a business opportunity.

The educational requirements for being a data scientist are quite rigorous—requiring a mastery of statistics, math, and computer programming. Most data scientist positions require an advanced degree, such as a master’s degree or a doctorate. Some organizations accept data scientists with under- graduate degrees in an analytical concentration, such as computer science, math and statistics, management information systems, economics, and engi- neering. Colorado Technical University, Syracuse University, and Villanova University are among the many schools that offer online master degree pro- grams related to BI and analytics.

Many schools also offer career-focused courses, degrees, and certificates in analytical-related disciplines such as database management, predictive ana- lytics, BI, big data analysis, and data mining. Such courses provide a great way for current business and information systems professionals to learn data scientist skills. Most data scientists have computer programming skills and are familiar with languages and tools used to process big data, such as Hadoop, Hive, SQL, Python, R, and Java.

386 PART 3 • Business Information Systems

Critical Thinking Exercise

The job outlook for data scientists is extremely bright. The McKinsey Global Institute (the business and economics research arm of the management consulting firm McKinsey & Co.) predicts that by 2018 the United States may face a shortage of 140,000 to 190,000 data scientists.7 The recruitment agency Glassdoor pegs the average salary for a data scientist at $118,709, and highly talented, educated, and experienced data scientists can earn well over $250,000 per year.

Components Required for Effective BI and Analytics A number of components must be in place for an organization to get real value from its BI and analytics efforts. First and foremost is the existence of a solid data management program, including data governance. Recall that data management is an integrated set of functions that defines the processes by which data is obtained, certified fit for use, stored, secured, and processed in such a way as to ensure that the accessibility, reliability, and timeliness of the data meet the needs of the data users within an organization. Data gover- nance is the core component of data management; it defines the roles, respon- sibilities, and processes for ensuring that data can be trusted and used by the entire organization, with people identified and in place who are responsible for fixing and preventing issues with data.

Another key component that an organization needs is creative data scientists—people who understand the business as well as the business ana- lytics technology, while also recognizing the limitations of their data, tools, and techniques. A data scientist puts all of this together to deliver real improvements in decision making with an organization.

Finally, to ensure the success of a BI and analytics program, the manage- ment team within an organization must have a strong commitment to data- driven decision making. Organizations that can put the necessary components in place can act quickly to make superior decisions in uncertain and changing environments to gain a strong competitive advantage.

Argosy Gaming Argosy Gaming Company is the owner and operator of six riverboat gambling casinos and hotels in the United States. Argosy has developed a centralized enter- prise data warehouse to capture the data generated at each property. As part of this effort, Argosy selected an extract-transform-load (ETL) tool to gather and inte- grate the data from six different operational databases to create its data ware- house. The plan is to use the data to help Argosy management make quicker, well-informed decisions based on patrons’ behaviors, purchases, and preferences. Argosy hopes to pack more entertainment value into each patron’s visit by better understanding their gambling preferences and favorite services. The data will also be used to develop targeted direct mail campaigns, customize offers for specific customer segments, and adapt programs for individual casinos.8

Review Questions 1. What are the key components that Argosy must put into place to create an

environment for a successful BI and analytics program? 2. What complications can arise from gathering data from six different opera-

tional databases covering six riverboat gambling casinos and hotels?

Critical Thinking Questions 1. The Argosy BI and analytics program is aimed at boosting revenue not at

reducing costs. Why do you think this is so? 2. What specific actions must Argosy take to have a successful program that will

boost revenue and offset some of the increases in costs?

CHAPTER 9 • Business Intelligence and Analytics 387

Business Intelligence and Analytics Tools

This section introduces and provides examples of many BI and analytics tools, including spreadsheets, reporting and querying tools, data visualization tools, online analytical processing (OLAP), drill-down analysis, linear regression, data mining, and dashboards. It will also cover the strategy of self-service ana- lytics, presenting its pros and cons.

Spreadsheets Business managers often import data into a spreadsheet program, such as Excel, which then can be used to perform operations on the data based on formulas created by the end user. Spreadsheets are also used to create reports and graphs based on that data. End users can employ tools such as the Excel Scenario Manager to perform “what-if” analysis to evaluate various alterna- tives. The Excel Solver Add-in can be used to find the optimal solution to a problem with multiple constraints (e.g., determine a production plan that will maximize profit subject to certain limitations on raw materials).

North Tees and Hartlepool National Health Services Trust provides health- care services and screenings to a population of 400,000 people in the United Kingdom. Professor Philip Dean, head of the Department of Pharmacy and Quality Control Laboratory Services, wanted a way to better understand the clinical use of drugs, the efficacy of treatment, and the associated costs. Dean worked with resources from Ascribe, a BI software and consulting firm, to pilot the use of Microsoft Power BI for Office 365, part of the Microsoft Office 365 cloud-based business productivity suite that works through familiar Excel spreadsheet software (see Figure 9.1). Ascribe developers took an extract of North Tees’s data and imported it into a Power BI model. They then incorpo- rated other data sets of interest to Dean and his colleagues, such as publicly available data on the activity of general practitioners, weather data, and

Sharepoint Distribute

and interact

Power Pivot Link and calculate

Data sources

Excel Analyze, design

and present

Power Map Visualize

Power View Visualize

Greatest sales opportunties

Power Query Extract and transform

T ri ff /S h u tt e rs to ck .c o m

FIGURE 9.1 Components of Microsoft Power BI for Office 365 Microsoft Power BI has been used to better understand the clinical use of drugs, the efficacy of treatment, and the associ- ated costs.

Source: Access Analytics, Power BI for Business, Power Analytics, http://www.accessanalytic.com.au/Power-BI.html.

388 PART 3 • Business Information Systems

treatment data. With all of this new data integrated in the Power BI model, Dean was able to create graphs of his findings, visualize data on regional maps, and even zoom in and around the data to gain various levels of insight. According to Dean, the ability to link disparate data sets for an integrated analysis was “one of the ‘wow’ things” that impressed him most in his use of BI tools. Incorporating additional, external data sets into his analyses comple- mented and helped explain trends, as well as provided useful benchmarks. Use of the weather data helped identify the impact of inclement weather on the frequency of respiratory disease. The treatment data helped Dean and his team to understand which drugs were being prescribed and how prescription patterns varied by locality.9

Reporting and Querying Tools Most organizations have invested in some reporting tools to help their employ- ees get the data they need to solve a problem or identify an opportunity. Reporting and querying tools can present that data in an easy-to-understand fashion—via formatted data, graphs, and charts. Many of the reporting and que- rying tools enable end users to make their own data requests and format the results without the need for additional help from the IT organization.

FFF Enterprises is a supplier of critical-care biopharmaceuticals, plasma products, and vaccines. Its 46,000 customers include over 80 percent of U.S. hospitals.10 The company employs the QlikView query and reporting tool to provide employees with real-time access to data that affects its business and the timely delivery of safe, effective products and services. For example, the company is the largest flu vaccine distributor in the United States, and accu- rately tracking its vaccine shipments is critical to avoiding shortages. As part of those efforts, FFF Enterprises uses QlikView to track and monitor the volume and value of all product transactions, such as the receipt, internal movement, and distribution of products.11

Data Visualization Tools Data visualization is the presentation of data in a pictorial or graphical format. The human brain works such that most people are better able to see significant trends, patterns, and relationships in data that is presented in a graphical format rather than in tabular reports and spreadsheets. As a result, decision makers welcome data visualization software that presents analytical results visually. In addition, representing data in visual form is a recognized technique to bring immediate impact to dull and boring numbers. A wide array of tools and techniques are available for creating visual representations that can immediately reveal otherwise difficult-to-perceive patterns or relation- ships in the underlying data.

Many companies now troll Facebook, Google Plus+, LinkedIn, Pinterest, Tumblr, Twitter, and other social media feeds to monitor any mention of their company or product. Data visualization tools can take that raw data and immediately provide a rich visual that reveals precisely who is talking about the product and what they are saying. Techniques as simple and intuitive as a word cloud can provide a surprisingly effective visual summary of conversa- tions, reviews, and user feedback about a new product. A word cloud is a visual depiction of a set of words that have been grouped together because of the frequency of their occurrence. Word clouds are generated from analy- ses of text documents or a Web page. Using the text from these sources, a simple count is carried out on the number of times a word or phrase appears. Words or phrases that have been mentioned more often than other words or phrases are shown in a larger font size and/or a darker color, as shown in Figure 9.2. ABCya, Image Chef, TagCloud, ToCloud, and ToCloud, Tagul, and Wordle are examples of word cloud generator software.

Data visualization: The presenta- tion of data in a pictorial or graphical format.

word cloud: A visual depiction of a set of words that have been grouped together because of the frequency of their occurrence.

CHAPTER 9 • Business Intelligence and Analytics 389

A conversion funnel is a graphical representation that summarizes the steps a consumer takes in making the decision to buy your product and become a customer. It provides a visual representation of the conversion data between each step and enables decision makers to see what steps are causing customers confusion or trouble. Figure 9.3 shows a conversion funnel for an online sales organization. It shows where visitors to a Web site are dropping off the successful sales path.

FIGURE 9.2 Word cloud This Word cloud shows the topics covered in this chapter.

FIGURE 9.3 The conversion funnel The conversion funnel shows the key steps in converting a consumer to a buyer.

Web site visits 100%

Visit Visit Visit Visit

Visit

Product views 73%

Cart additions 23%

Checkouts 11%

Purchases 3%

conversion funnel: A graphical representation that summarizes the steps a consumer takes in making the decision to buy your product and become a customer.

m in ds ca nn er /S hu tt er st oc k. co m

390 PART 3 • Business Information Systems

Technologia is a business training firm that has trained over 70,000 clients in dozens of technical topics, such as project management, SQL, and Microsoft Windows. One Technologia course can cost $1,000 or more, so most potential customers do careful detailed research and make multiple visits to the com- pany’s Web site before enrolling in a course. Technologia used Multi-Channel Funnels from Google Analytics to determine what factors had the most impact in influencing students to enroll. For the first time, Technologia learned that nearly 18 percent of its sales paths included paid advertising—much higher than previously thought. As a result, it raised its online ad budget by nearly 100 percent, and online conversions shot up 120 percent.12

Dozens of data visualization software products are available for creating various charts, graphs, infographics, and data maps (see Figure 9.4). Some of the more commonly products include Google Charts, iCharts, Infogram, Modest Maps, SAS Visual Statistics, and Tableau. These tools make it easy to visually explore data on the fly, spot patterns, and quickly gain insights.

Online Analytical Processing Online analytical processing (OLAP) is a method to analyze multidimen- sional data from many different perspectives. It enables users to identify issues and opportunities as well as perform trend analysis. Databases built to support OLAP processing consist of data cubes that contain numeric facts called measures, which are categorized by dimensions, such as time and geog- raphy. A simple example would be a data cube that contains the unit sales of a specific product as a measure. This value would be displayed along the metric dimension axis shown in Figure 9.5. The time dimension might be a specific day (e.g., September 30, 2018), whereas the geography dimension might define a specific store (e.g., Krogers in the Cincinnati, Ohio community of Hyde Park).

The key to the quick responsiveness of OLAP processing is the preaggre- gation of detailed data into useful data summaries in anticipation of questions that might be raised. For example, data cubes can be built to summarize unit sales of a specific item on a specific day for a specific store. In addition, the detailed store-level data may be summarized to create data cubes that show

FIGURE 9.4 Data visualization This scatter diagram shows the relationship between MSRP and horsepower. Source: “Data Visualization," SAS, http:// www.sas.com/en_us/insights/big-data/data -visualization.html#m=lightbox5, accessed April 19, 2016.

online analytical processing (OLAP): A method to analyze multidi- mensional data from many different perspectives, enabling users to identify issues and opportunities as well as perform trend analysis.

data cube: A collection of data that contains numeric facts called mea- sures, which are categorized by dimensions, such as time and geography.

CHAPTER 9 • Business Intelligence and Analytics 391

unit sales for a specific item, on a specific day for all stores within each major market (e.g., Boston, New York, Phoenix), for all stores within the United States, or for all stores within North America. In a similar fashion, data cubes can be built in anticipation of queries seeking information on unit sales on a given day, week, month, or fiscal quarter.

It is important to note that if the data within a data cube has been sum- marized at a given level, for example, unit sales by day by store, it is not pos- sible to use that data cube to answer questions at a more detailed level, such as what were the unit sales of this item by hour on a given day.

Data cubes need not be restricted to just three dimensions. Indeed, most OLAP systems can build data cubes with many more dimensions. In the busi- ness world, data cubes are often constructed with many dimensions, but users typically look at just three at a time. For example, a consumer packaged goods manufacturer might build a multidimensional data cube with information about unit sales, shelf space, unit price, promotion price, and level of newspaper advertising—all for a specific product, on a specific date, in a specific store.

In the retail industry, OLAP is used to help firms to predict better cus- tomer demand and maximize sales. Starbucks employs some 149,000 workers in 10,000 retail stores in the United States. The firm built a data warehouse to hold 70 terabytes of point-of-sale and customer loyalty data. This data is com- pressed into data cubes of summarized data to enable users to perform OLAP analysis of store-level sales and operational data.13

Drill-Down Analysis The small things in plans and schemes that don’t go as expected can frequently cause serious problems later on—the devil is in the details. Drill-down analysis is a powerful tool that enables decision makers to gain insight into the details of business data to better understand why something happened.

Drill-down analysis involves the interactive examination of high-level summary data in increasing detail to gain insight into certain elements—sort of like slowly peeling off the layers of an onion. For example, in reviewing the worldwide sales for the past quarter, the vice president of sales might want to drill down to view the sales for each country. Further drilling could be done to view the sales for a specific country (say Germany) for the last quarter. A third level of drill-down analysis could be done to see the sales for a specific country for a specific month of the quarter (e.g., Germany for the month of September). A fourth level of analysis could be accomplished by drilling down to sales by product line for a particular country by month (e.g., each product line sold in Germany for the month of September).

FIGURE 9.5 A data cube The data cube contains numeric facts that are categorized by dimensions, such as time and geography.

Time dimension

Geography dimension

Metric dimension

drill-down analysis: The interactive examination of high-level summary data in increasing detail to gain insight into certain elements—sort of like slowly peeling off the layers of an onion.

392 PART 3 • Business Information Systems

Brisbane is a city on the east coast of Australia that is subject to frequent creek flash flooding from the many streams in the area. One year, particularly heavy rainfall caused many houses to be flooded, brought down power lines, closed roads, and put the city into a state of emergency. Following this disaster, the city installed telemetry gauges across Brisbane to obtain real-time measure- ments of rainfall and water levels. The data is captured and displayed on color- coded maps, which enable staff to quickly spot areas of concern. They can also perform a drill-down analysis to see increasing levels of detail within any critical area. The system enables staff to provide more advanced warnings to the popu- lation of impending flooding and take action to close roads or clean up debris.14

Linear Regression Simple linear regression is a mathematical technique for predicting the value of a dependent variable based on a single independent variable and the linear relationship between the two. Linear regression consists of finding the best- fitting straight line through a set of observations of the dependent and indepen- dent variables. By far, the most commonly used measure for the best-fitting line is the line that minimizes the sum of the squared errors of prediction. This best- fitting line is called the regression line (see Figure 9.6). Linear regression does not mean that one variable causes the other; it simply says that when one value goes up, the other variable also increases or decreases proportionally.

The regression line can be written as Y ¼ a þ bX þ ". In this equation, the following are true:

● X is the value of the independent variable that is observed ● Y is the value of the dependent variable that is being predicted ● a is the value of Y when X is zero, or the Y intercept ● b is the slope of the regression line ● " is the error in predicting the value of Y, given a value of X

The following key assumptions must be satisfied when using linear regression on a set of data:

● A linear relationship between the independent (X) and dependent (Y) variables must exist.

● Errors in the prediction of the value of Y are distributed in a manner that approaches the normal distribution curve.

● Errors in the prediction of the value of Y are all independent of one another.

FIGURE 9.6 Simple linear regression This graph shows a linear regression that predicts students’ final exam scores based on their math aptitude test score.

100

160

140

120

S co

re o

n st

an da

rd a

ch ie

ve m

en t

te st

Score on math aptitude test

100

806040200

Best fit linear regression line

linear regression: A mathematical procedure to predict the value of a dependent variable based on a single independent variable and the linear relationship between the two.

CHAPTER 9 • Business Intelligence and Analytics 393

A number of advanced statistical tests can be used to examine whether or not these assumptions are true for a given set of data and the resulting linear regression equation. For example, the coefficient of determination, denoted r² (and pronounced r squared) is a number that indicates how well data fit a sta- tistical model—sometimes simply a line or a curve. An r² of 1 indicates that the regression line perfectly fits the data, whereas an r² of 0 indicates that the line does not fit the data at all. An r² of 0.92 means that 92 percent of the total variation in Y can be explained by the linear relationship between X and Y as described by the regression equation. The other 8 percent of total variation in Y remains unexplained. A data scientist would always perform several tests to determine the validity of a linear regression to understand how well the model matches to actual data.

Data Mining Data mining is a BI analytics tool used to explore large amounts of data for hidden patterns to predict future trends and behaviors for use in decision making. Used appropriately, data mining tools enable organizations to make predictions about what will happen so that managers can be proactive in capitalizing on opportunities and avoiding potential problems.

Among the three most commonly used data mining techniques are associ- ation analysis (a specialized set of algorithms sorts through data and forms statistical rules about relationships among the items), neural computing (historical data is examined for patterns that are then used to make predic- tions), and case-based reasoning (historical if-then-else cases are used to recognize patterns).

The Cross-Industry Process for Data Mining (CRISP-DM) is a six-phase structured approach for the planning and execution of a data mining project (see Figure 9.7). It is a robust and well-proven methodology, and although it was first conceived in 1999, it remains the most widely used methodology for data mining projects.15 The goals for each step of the process are summarized in Table 9.1.

FIGURE 9.7 The Cross-Industry Process for Data Mining (CRISP-DM) CRISP-DM provides a structured approach for planning and execut- ing a data mining project. Source: Piatetsky, Gregory, “CRISP-DM, Still the Top Methodology for Analytics, Data Mining, or Data Science Projects,” KDNug- gets, October 28, 2014, www.kdnuggets .com/2014/10/crisp-dm-top-methodology -analytics-data-mining-data-science-pro jects.html.

Data

preparation

Evaluation

Deployment

Database

Modeling

Business

understanding Data

understanding

data mining: A BI analytics tool used to explore large amounts of data for hidden patterns to predict future trends and behaviors for use in decision making.

Cross-Industry Process for Data Mining (CRISP-DM): A six-phase structured approach for the planning and execution of a data mining project.

394 PART 3 • Business Information Systems

Here are a few examples showing how data mining can be used:

● Based on past responses to promotional mailings, identify those consu- mers most likely to take advantage of future mailings.

● Examine retail sales data to identify seemingly unrelated products that are frequently purchased together.

● Monitor credit card transactions to identify likely fraudulent requests for authorization.

● Use hotel booking data to adjust room rates so as to maximize revenue. ● Analyze demographic data and behavior data about potential customers to

identify those who would be the most profitable customers to recruit. ● Study demographic data and the characteristics of an organization’s most

valuable employees to help focus future recruiting efforts. ● Recognize how changes in an individual’s DNA sequence affect the risk of

developing common diseases such as Alzheimer’s or cancer.

The average production cost of a Hollywood movie in 2007 was $106 million, with an additional $36 million spent on marketing of the film. With that kind of money being spent, potential investors need to have a good sense of what films will earn a profit and which ones won’t. Researchers from the University of Iowa took data from two online sources (the Internet Movie Database and Box Office Mojo) to build a database of over 14,000 films and 4,000 actors and directors from films released between 2000 and 2010. They calculated return on investment for each film to get an estimate of its profit- ability. The researchers then used a data mining algorithm to discover pat- terns that predict movie profitability. Using the algorithm, the researchers discovered that the factor most strongly correlated with a film’s profitability is the average gross revenue made by the director’s previous films—directors who have generated more revenue in the past are correlated with greater profitability in future. Somewhat surprisingly, while big stars boost box office receipts, they don’t guarantee a profit, because they cost a lot to hire in the first place and they are often involved in higher budget films that require more revenue to generate a profit.16

Dashboards Measures are metrics that track progress in executing chosen strategies to attain organizational objectives and goals. These metrics are also called key performance indicators (KPIs) and consist of a direction, measure, target, and time frame. To enable comparisons over different time periods, it is also important to define the KPIs and to use the same definition from year to

TABLE 9.1 Goals for each phase of CRISP-DM Phase Goal

Business understanding ● Clarify the business goals for the data mining project, convert the goals into a predictive analysis problem, and design a project plan to accomplish these objectives.

Data understanding ● Gather data to be used (may involve multiple sources), become familiar with the data, and identify any data quality problems (lack of data, missing data, data needs adjust- ment, etc.) that must be addressed.

Data preparation ● Select a subset of data to be used, clean data to address quality issues, and transform data into form suitable for analysis.

Modeling ● Apply selected modeling techniques.

Evaluation ● Assess if the model achieves business goals.

Deployment ● Deploy the model into the organization’s decision-making process.

Source: Leaper, Nicole, “A Visual Guide to CRISP-DM Methodology,” https://exde.files.wordpress.com/2009/03/crisp_visualguide.pdf, accessed January 20, 2016.

key performance indicator (KPI): A metric that tracks progress in executing chosen strate- gies to attain organizational objectives and goals and consists of a direction, measure, target, and time frame.

CHAPTER 9 • Business Intelligence and Analytics 395

year. Over time, some existing KPIs may be dropped and new ones added as the organization changes its objectives and goals. Obviously, just as different organizations have different goals, various organizations will have different KPIs. The following are examples of well-defined KPIs:

● For a university. Increase (direction) the five-year graduation rate for incoming freshman (measure) to at least 80 percent (target) starting with the graduating class of 2022 (time frame).

● For a customer service department. Increase (direction) the number of customer phone calls answered within the first four rings (measure) to at least 90 percent (target) within the next three months (time frame).

● For an HR organization. Reduce (direction) the number of voluntary resignations and terminations for performance (measure) to 6 percent or less (target) for the 2018 fiscal year and subsequent years (time frame).

A dashboard presents a set of KPIs about the state of a process at a spe- cific point in time. Dashboards provide rapid access to information, in an easy-to-interpret and concise manner, which helps organizations run more effectively and efficiently.

Options for displaying results in a dashboard can include maps, gauges, bar charts, trend lines, scatter diagrams, and other representations, as shown in Figures 9.8 and Figure 9.9. Often items are color coded (e.g., red ¼ problem; yellow ¼ warning; and green ¼ OK) so that users can see at a glance where attention is needed. Many dashboards are designed in such a manner that users can click on a section of the chart displaying data in one format and drill down into the data to gain insight into more specific areas. For example, Figure 9.9 represents the results of drilling down on the sales region of Figure 9.8.

Dashboards provide users at every level of the organization the informa- tion they need to make improved decisions. Operational dashboards can be designed to draw data in real time from various sources, including corpo- rate databases and spreadsheets, so decision makers can make use of up-to-the-minute data.

FIGURE 9.8 Category management dashboard for total U.S. region This dashboard summarizes a number of sales measures.

Source: www.microstrategy.com/us/analytics/technology.

dashboard: A presentation of a set of KPIs about the state of a process at a specific point in time.

396 PART 3 • Business Information Systems

Widely used BI software comes from many different vendors, including Hewlett Packard, IBM, Information Builders, Microsoft, Oracle, and SAP, as shown in Table 9.2. Vendors such as JasperSoft and Pentaho also provide open-source BI software, which is appealing to some organizations.

Self-Service Analytics Self-service analytics includes training, techniques, and processes that empower end users to work independently to access data from approved sources to perform their own analyses using an endorsed set of tools. In the past, such data analysis could only be performed by data scientists. Self- service analytics encourages nontechnical end users to make decisions based on facts and analyses rather than intuition. Using a self-service analytics appli- cation, end users can gather insights, analyze trends, uncover opportunities and issues, and accelerate decision making by rapidly creating reports, charts, dashboards, and documents from any combination of enterprise information assets. Self-service analytics eliminates decision-making delays that can arise if all requests for data analyses must be made through a limited number of data scientists and/or information system resources. It also frees up these resources to do higher-level analytics work. Ideally, self-service analytics will lead to faster and better decision making.

An organization can take several actions to ensure an effective self-service analytics program. First, to mitigate the risks associated with self-service ana- lytics, data managers should work with business units to determine key metrics, an agreed-upon vocabulary, processes for creating and publishing reports, the privileges required to access confidential data, and how to define and implement security and privacy policies. The information systems organi- zation should help users understand what data is available and recommended for business analytics. One approach to accomplishing this is to provide a data dictionary for use by end users. Training, on both the data and on the use of self-service applications, is critical for getting end workers up to speed

FIGURE 9.9 Category management dashboard for Northwest region This dashboard summarizes a number of revenue measures.

Source: www.microstrategy.com/us/analytics/technology.

self-service analytics: Training, techniques, and processes that empower end users to work indepen- dently to access data from approved sources to perform their own analyses using an endorsed set of tools.

CHAPTER 9 • Business Intelligence and Analytics 397

on how they can use the information in the BI system. Finally, data privacy and security measures should be in place to ensure that the use of the data meets legal, compliance, and regulatory requirements.

A well-managed self-service analytics program allows technology profes- sionals to retain ultimate data control and governance while limiting informa- tion systems staff involvement in routine tasks. Modern data management requires a true balancing act between enabling self-service analysis and pro- tecting sensitive business information, as shown in Figure 9.10.

Table 9.3 presents the pros and cons associated with self-service BI and analytics.

For self-service analytics tools to be effective, they must be intuitive and easy to use. Business users simply don’t have the time to learn how to work with complex tools or sophisticated interfaces. A self-service analytics applica- tion will only be embraced by end users if it allows them to easily access their own customized information, without extensive training. Microstrategy, Power BI, Qlik, SAS Analytics, Tableau, and TIBCO Software are just a few examples of the dozens of software options available for self-service analytics.

Expert Storybooks, a cloud-based, self-service analytics service from IBM’s Watson Analytics line, provides data analysis models that offer connections to a range of data sources, along with secure connections to corporate data. Expert Storybooks are tools for creating sophisticated data visualizations to help users find relevant facts and discover patterns and relationships to make predictive decisions. There are several Expert Storybooks available, including

TABLE 9.2 Widely used BI software Vendor Product Description

HP Autonomy IDOL17 Enables organizations to process unstructured as well as structured data; the software can examine the intricate relationships between data to answer the crucial question “Why has this happened?”

IBM Cognos Business Intelligence18

Turns data into past, present, and future views of an organization’s operations and performance so decision makers can identify opportunities and minimize risks; snapshots of business performance are provided in reports and inde- pendently assembled dashboards.

Information Builders

WebFOCUS19 Produces dashboards and scorecards to display a high-level view of critical indicators and metrics; the software enables users to analyze and manipulate information, with minimal training. It also supports dynamic report distribu- tion, with real-time alerts, and fully automates the scheduling and delivery of vital information.

Microsoft Power BI for Office 36520

Allows users to model and analyze data and query large data sets with pow- erful natural-language queries; it also allows users to easily visualize data in Excel.

Oracle Business Intelligence21

Offers a collection of enterprise BI technology and applications; tools includ- ing an integrated array of query, reporting, analysis, mobile analytics, data integration and management, desktop integration, and financial performance management applications; operational BI applications; and data warehousing.

Oracle Hyperion22 Provides software modules to enable financial management; modules include those for budgeting, planning, and forecasting; financial reporting; database management; financial consolidation; treasury management; and analytics.

SAS Enterprise BI Server23 Provides software modules to support query and analysis, perform OLAP processing, and create customizable dashboards; the software integrates with Microsoft Office.

SAP Business Objects24 Offers a suite of applications that enable users to design and generate reports, create interactive dashboards that contain charts and graphs for visualizing data, and create ad hoc queries and analysis of data; also allows users to search through BI data sources.

398 PART 3 • Business Information Systems

Critical Thinking Exercise

one that uses baseball statistics from AriBall to build predictions of player per- formance, enabling users to gain an edge over their fantasy baseball competi- tors. A variety of other Storybooks help end users incorporate weather data into revenue analysis; analyze social data to measure reputational risk; analyze marketing campaign data; identify and analyze trends in customer profitability; analyze market trends for investment strategy; and examine relationships among pay, performance, and credit risk.25

Fire Department Turns to BI Analytics New York City has nearly 1 million buildings, and each year, more than 3,000 of them experience a major fire. The Fire Department of the City of New York (FDNY) is adding BI analytics to its arsenal of firefighting equipment. It has created a database of over 60 different factors (e.g., building location, age of the building, whether it has electrical issues, the number and location of sprinklers) in an attempt to determine which buildings are more likely to have a fire than others. The values of these parameters for each building are fed into a BI analytics system

TABLE 9.3 Pros and cons associated with self-service BI and analytics Pros Cons

Gets valuable data into the hands of the people who need it the most—end users.

If not well managed, it can create the risk of erroneous analysis and reporting, leading to potentially damaging decisions within an organization.

Encourages nontechnical end users to make decisions based on facts and analyses rather than intuition.

Different analyses can yield inconsistent conclusions, resulting in wasted time trying to explain the differences. Self-service analytics can also result in proliferating “data islands,” with duplications of time and money spent on analyses.

Accelerates and improves decision making. Can lead to over spending on unapproved data sources and business analytics tools.

Business people can access and use the data they need for decision making, without having to go to technology experts each time they have a new question, thus filling the gap caused by a shortage of trained data scientists.

Can exacerbate problems by removing the checks and balances on data preparation and use. Without strong data governance, organizations can end up with lots of silos of information, bad analysis, and extra costs.

FIGURE 9.10 Importance of data management Modern data management requires a true balancing act between enabling self-service analysis and protecting sensitive business information.

Self-service

analytics

Protecting sensitive

business information

CHAPTER 9 • Business Intelligence and Analytics 399

that assigns each of the city’s 330,000 inspectable buildings a risk score. (FDNY doesn’t inspect single and two-family homes.) Fire inspectors then use these risk scores to prioritize which buildings to visit on their weekly inspections.26

Review Questions 1. What kinds of BI analytics tools and techniques is the FDNY likely to use in

sifting through all this data and determining a building’s risk score? 2. Identify three other parameters that ought to be taken into consideration

when setting priorities for building inspections.

Critical Thinking Questions 1. While making investments in BI analytics seems like a good idea, FDNY is

strongly challenged in measuring its success. Officials may be able to cite sta- tistics showing a reduction in the number of fires, but demonstrating that BI analytics tools were the reason behind that decrease may be difficult because it involves proving a negative—that something didn’t happen because of its efforts. Go to the FDNY citywide statistics Web site at www.nyc.gov/html /fdny/html/stats/citywide.shtml. Use those statistics and a data visualization tool of your choice to see if you can discern any change in the number of fires since the BI analytics system was installed in 2014.

2. Can you identify other approaches that would be effective in demonstrating the value of BI analytics in reducing the impact of fires in New York City?

Summary

Principle: The goal of business intelligence (BI) and analytics is to support improved decision making.

Business intelligence (BI) includes a wide range of applications, practices, and technologies for the extraction, transformation, integration, visualization, analysis, and presentation of data to support improved decision making.

Business analytics is the extensive use of data and quantitative analysis to support fact-based decision making within organizations.

A data scientist is an individual who combines strong business acumen, a deep understanding of analytics, and a healthy appreciation of the limitations of their data, tools, and techniques to deliver real improvements in decision making. The educational requirements for a data scientist are quite rigorous, and the job outlook for this profession is extremely good.

A number of components must be in place for an organization to get real value from its BI and analytics efforts: a solid data management program (including a strong data governance element), creative data scientists, and a strong organizational commitment to data-driven decision making.

Principle: There are many BI and analytics techniques and tools that can be used in a wide range of problem-solving situations.

Spreadsheets, reporting and querying tools, data visualization, online ana- lytical processing (OLAP), drill-down analysis, linear regression, data mining, and dashboards are examples of commonly used BI tools.

Business managers and end users often turn to spreadsheets to create use- ful reports and graphs, as well as to employ “what-if” analyses and find the optimal solution to a problem with multiple constraints.

400 PART 3 • Business Information Systems

Reporting and querying tools can present data in an easy-to-understand fashion—via formatted data, graphs, and charts. Many of the reporting and querying tools enable end users to make their own data requests and format the results without the need for additional help from the IT organization.

Data visualization is the presentation of data in a pictorial or graphical format. A wide array of tools and techniques are available for creating visual representations that can immediately reveal otherwise difficult-to-perceive pat- terns or relationships in the underlying data.

Online analytical processing (OLAP) is a method to analyze multidimen- sional data (data cubes) from many different perspectives. Databases built to support OLAP processing consist of data cubes that contain numeric facts called measures, which are categorized by dimensions, such as time and geography.

Drill-down analysis involves the interactive examination of high-level sum- mary data in increasing detail to gain insight into certain elements.

Simple linear regression is a mathematical procedure technique for predict- ing the value of a dependent variable based on a single independent variable and the linear relationship between the two.

Data mining is a BI analytics tools used to explore large amounts of data for hidden patterns to predict future trends and behaviors for use in decision mak- ing. The Cross-Industry Process for Data Mining (CRISP-DM) is a six-phase structured approach used for the planning and execution of a data mining project.

A dashboard presents a set of KPIs about the state of a process at a specific point in time. Dashboards provide rapid access to information, in an easy- to-interpret and concise manner, which helps organizations run more effec- tively and efficiently.

Self-service analytics includes training, techniques, and processes that empower end users to work independently to access data from approved sources to perform their own analyses using an endorsed set of tools. Self- service analytics empowers end users to work independently. A number of measures must be in place to ensure an effective self-serve analytics program and to reduce the risk of invalid analyses leading to poor decisions.

Key Terms

business intelligence (BI)

conversion funnel

Cross-Industry Process for Data Mining (CRISP-DM)

dashboard

data cube

data mining

data visualization

drill-down analysis

key performance indicator (KPI)

linear regression

online analytical processing (OLAP)

self-service analytics

word cloud

Chapter 9: Self-Assessment Test

The goal of business intelligence (BI) is to support improved decision making.

1. Which of the following statements is not true? a. The data used in BI is often pulled from mul-

tiple sources—both internal and external to the organization.

b. Users may access the data and BI applications via the Web or through organizational intra- nets and extranets—often using mobile devices, such as smartphones and tablets.

c. The data used in BI applications can come from data warehouses, data marts, and data lakes.

CHAPTER 9 • Business Intelligence and Analytics 401

d. BI is strictly the realm of data scientists; end users and business managers should be dis- couraged from using these tools and techniques.

2. An individual who combines strong business acumen, a deep understanding of analytics, and healthy appreciation of their data, tools, and techniques to deliver real improvements in deci- sion making is called a _____________. a. data steward b. database administrator c. data scientist d. database manager

3. Which of the following is not an essential component for a highly effective BI program? a. Creative data scientists b. Strong data management program c. Strong management commitment to

data-driven decision making d. The most current and powerful BI and

analytics tools and software 4. Most data scientists have computer programming

skills and are familiar with languages and tools used to process big data, such as: a. Hadoop, Hive, R b. Python, R, and Assembly c. Visual Basic, Java, and Cobol d. PL/1, Hadoop, PHP

Principle: There are many business intelligence (BI) and analytics techniques and tools that can be used in a wide range of problem-solving situations.

5. ____________ is a BI analytics tool that involves the interactive examination of high-level sum- mary data in increasing detail to gain insight. a. OLAP b. Drill-down analysis c. Linear regression d. Dashboard

6. A(n) _______________________ is a measure that tracks progress in executing chosen strategies to attain organizational objectives and goals.

7. Self-service analytics encourages nontechnical end users to make decisions based on facts and analyses rather than intuition. True or False?

8. __________ is a BI analytics tool used to explore large amounts of data for hidden patterns to pre- dict future trends and behaviors for use in deci- sion making. a. Linear regression b. Data mining c. OLAP d. Data visualization

9. A(n) ________________________ is a graphical representation that summarizes the steps a con- sumer takes in making the decision to buy your product and become a customer.

Chapter 9: Self-Assessment Test Answers

1. d 2. c 3. d 4. a 5. b

6. key performance indicator (KPI) 7. True 8. b 9. conversion funnel

Review Questions

1. Provide a definition of business intelligence (BI). 2. Provide a definition of business analytics. 3. Identify and briefly discuss several benefits that

can be gained through the use of BI and analytics.

4. Describe the role of a data scientist. What educa- tion and training are required of data scientists?

5. What is online analytical processing (OLAP), and how is it used?

6. What is a data cube? Define three data dimen- sions of a data cube that might be used to analyze sales by product line.

7. What is drill-down analysis and how is it used?

8. What is linear regression? Identify two variables that have a strong linear relationship.

9. What is r2, and how is it used to evaluate how well data fits a linear regression model?

10. What is data mining? Identify three commonly used data mining techniques.

11. What are key performance indicators (KPIs)? Identify three KPIs that might be used in a doc- tor’s or dentist’s office to measure the patient office visit experience.

12. What is meant by self-service analytics? Identify some of the pros and cons of self-service analytics.

402 PART 3 • Business Information Systems

Discussion Questions

1. Identify and briefly discuss the components that must be in place for an organization to gain real value from its BI and analytics efforts.

2. How would you define BI? Identify and briefly discuss a real-world application of BI that you recently experienced.

3. Imagine that you are the sales manager of a large luxury auto dealer. What sort of data would be useful to you in order to contact potential new car customers and invite them to visit your dealership? Where might you get such data? What sort of BI tools might you need to make sense of this data?

4. You answer your door to find a political activist who asks you to sign a petition to place a propo- sition on the ballot that, if approved, would ban the use of data mining that includes any data about the citizens of your state. What would you do?

5. What is the difference between OLAP analysis and drill-down analysis? Provide an example of the effective use of each technique.

6. Identify at least four well-defined KPIs that could be used by the general manager at a large, full- service health club to track the current state of operations, including availability of trainers; sta- tus of workout equipment; condition of indoor and outdoor swimming pools; use of the spa and salon; utilization of the basketball, handball, and tennis courts; and condition of the showers and locker rooms. Sketch what a dashboard display- ing those KPIs might look like.

7. Your nonprofit organization wishes to increase the efficiency of its fundraising efforts. What sort of data might be useful to achieve this goal? How might BI tools be used to analyze this data?

8. Describe the CRISP-DM model, and explain how it can be used to plan and execute a data mining effort.

9. Must you be a trained statistician to draw valid conclusions from the use of BI tools? Why or why not?

Problem-Solving Exercises

1. You are the sales manager of a software firm that provides BI software for reporting, query, and business data analysis via OLAP and data mining. Write a few paragraphs and create a 3-5 slide presentation that your sales reps can use when calling on potential customers to help them understand the business benefits of BI.

2. Go to the NASA Goddard Institute for Space Studies Web site at http://data.giss.nasa.gov/gis temp and read about the temperature data sets available. Next, use the NASA Global land-sea

temperature data set for the period 1880 to pres- ent at http://data.giss.nasa.gov/gistemp/tableda ta_v3/GLB.Ts+dSST.txt and create a linear regres- sion of annual mean temperature in degrees Fahrenheit versus year. Be sure to adjust the table values based on the notes that accompany this table.

3. Use the Help function of Excel to learn more about the Excel Scenario Manager and the Excel Solver Add-in.

Team Activities

1. A large customer call center for a multinational retailer has been in operation for several months, but has continually failed to meet both customers’ and senior management’s expectations. Your team has been called in to develop a dashboard to help monitor and improve the operations of the call center. What KPIs would you track? Sketch a sample dashboard for this application.

2. Your team has decided to enter a Fantasy Base- ball League. Use IBM’s Watson analytics Story- books to select the best team in the league.

3. Your team has been hired to reapply the NYCFD BI analytics system to the NYCPD in order to prioritize areas of the city for patrol. Describe what data you might need for such an analysis. What BI analytics techniques would you use to become familiar with the data and develop an algorithm for ranking the various area for patrol?

CHAPTER 9 • Business Intelligence and Analytics 403

Web Exercises

1. Do research to learn more about the Goddard Institute of Space Science Surface Temperature Analysis (GISTEMP) temperature measurements used by many scientists to assert that global warming is occurring. One source of such infor- mation can be found at http://data.giss.nasa.gov /gistemp/FAQ.html. Read carefully the sections that explain why adjusted rather than raw data is used. Briefly summarize the kinds of adjustments made to temperature measurements and discuss how this affects your confidence in the data.

2. Do research on the Web to find business analytics applications that can run on a smartphone.

Prepare a brief paragraph summarizing the fea- tures and potential applications of three applica- tions of most interest to you.

3. Do research on the Web to identify examples of a social network allowing a large national retailer to mine its data to learn more about the social network members and to develop targeted direct mailings and emails promoting its products. What are the pros and cons of sharing such data? How do social network members feel when they learn that their data is being shared? In your opinion, under what conditions is this a legiti- mate use of social network member data?

Career Exercises

1. Do research on CareerBuilder, Indeed, LinkedIn, or SimplyHired to identify what you think is an attractive job opening available for a data scien- tist. What does this role entail, and what respon- sibilities must the data scientist perform? What sort of education and experience is required for this position? Is the role of a data scientist of any interest to you? Why or why not?

2. What kinds of decisions are critical for success in a career that interests you? How might you use BI

or analytics to help improve your decision mak- ing? Can you identify and specific techniques or tools that you might use?

3. Copy and paste the text describing a job oppor- tunity in which you are interested into an online word cloud generator. What insights do you gain from the resulting word cloud? How might the results lead you to customize your resume to respond to this job posting?

Case Studies

Case One

Analytics Used to Predict Patients Likely to Be Readmitted Unplanned hospital readmissions are a serious matter for patients and a quality and cost issue for the healthcare system of every country. For example, in the United States, during 2011, nearly 19 percent of Medicare patients were readmitted to the hospital within 30 days of their initial discharge, running up an additional $26 billion in healthcare costs. Hospitals are seeking more effective ways to identify patients at high risk of readmission—especially now that Medicare has begun reducing payments to hospitals with high readmission rates.

Identifying patients at high risk for readmission is important so that hospitals can take a range of preventative measures, including heightened patient education along with medication reconciliation on the day of discharge, increased home services to ensure patient effective at home convalescence, follow-up appointments scheduled for soon after discharge, and follow-up phone calls to ensure an additional level of protection.

Several studies have attempted to identify the key factors that indicate a high risk for unplanned hospital readmission. One study was based on the analysis of the Belgian Hospital Discharge Dataset. This data set contains

patient demographics, data about the hospital stay (date and type of admission and discharge, referral data, admitting department, and destination after discharge), and clinical data (primary and secondary diagnoses). Since 1990, Belgium has required the collection of this data for all inpatients in all acute hospitals. The data is managed by a commission that controls the content and format of patient registration, the data collecting procedures, and the completeness, validity, and reliability of the collected data. In addition, the quality of the data is audited by the Belgium Ministry of Public Health in two ways. First, a software program checks the data for missing, illogical, and outlier values. Second, by regular hospital visits, a random selection of patient records is reviewed to ensure that data were recoded correctly.

Key factors for hospital readmission based on analysis of the Belgian Hospital Discharge Dataset included: (1) chronic cardiovascular disease, (2) patients with chronic pulmonary disease, (3) patients who experienced multiple emergency room visits over the past six months, (4) patients discharged on a Friday, and (5) patients who had a prolonged length of hospital stay. The study also found that patients with short hospital stays were not at high risk for readmission. The research highlighted the need for healthcare providers to work with caregivers and primary care physicians to

404 PART 3 • Business Information Systems

coordinate a smoother transition from hospital to home, especially for patients discharged on Friday, to reduce unplanned readmissions.

The mission of Penn Medicine Center for Evidence-based Practice (CEP) is to support healthcare quality and safety at the University of Pennsylvania Health System (UPHS) through the practice of evidence-based medicine. Established in 2006, Penn Medicine’s CEP is staffed by a hospital director, three research analysts, six physician and nurse liaisons, a health economist, a biostatistician, administrator, and librarians. A study conducted by a team at the CEP examined two years of UPHS discharge data and found that a single variable—prior admission to the hospital two or more times within a span of one year—was the best predictor of being readmitted in the future. This marker was added to UPHS’s EHR, and patient results were tracked for the next year. During that time, patients who triggered the readmission alert were subsequently readmitted 31 percent of the time. When an alert was not triggered, patients were readmitted only 11 percent of the time.

A group of physicians conducted yet a third study using data from a 966-bed, teaching hospital during a five-month period in 2011. Their objective was to determine the association between a composite measure of patient condition at discharge, the Rothman Index (RI), and unplanned readmission within 30 days of discharge. Software employing the Rothman Index tracks the overall state of health of patients by continuously gathering 26 key pieces of data, including vital signs (temperature, blood pressure, heart, blood oxygen saturation, and respiratory rate), nursing assessments, cardiac rhythms, and lab test results from a patient’s EHR to calculate the Rothman Index, a number between 1 and 100. A patient’s Rothman Index is updated continuously throughout the day. A high score indicates a relatively healthy patient, whereas a low score indicates the patient warrants close monitoring or immediate assistance. A physician or nurse can quickly grasp the condition of the patient based on both current score and the trend in the score. The software also draws a graph of the patient’s Rothman Index over time that can be displayed in the patient’s room, on a central nursing station screen, or on a care provider’s mobile device. The software can even send mobile phone alerts to doctors and nurses when a patient’s condition warrants attention.

The Rothman Index study included clinical data from the hospital’s EHR system as well as from a patient activity database for all adult discharges. There was a total of 12,844 such cases. The researchers excluded encounters that were readmissions within 30 days of a previous discharge (2,574), patients who were admitted for observation only (501), patients with length of stay less than 48 hours (3,243), and patients who died during the hospital stay (189)—yielding a sample of 6,337 eligible inpatient discharges. From this sample, 535 additional patients were eliminated due to missing clinical data, for a sample of 5,802 patients, or 92 percent of all eligible inpatient discharges. Sixteen percent of the sample patients had an unplanned readmission within 30 days of discharge. The risk of readmission for a patient in the highest risk category (Rothman Index lower than 70) was more than 1 in 5 while the risk of readmission for patients in the lowest risk category was about 1 in 10.

Critical Thinking Questions 1. Three different analytic studies by three experienced

and highly respected groups of researchers yielded three similar but somewhat different results. Do you believe that the results of these studies are consistent? Why or why not?

2. Do you think the findings of these studies can be applied broadly to all hospitals and medical centers across the United States and around the world? Why or why not?

3. A hospital specializing in the care of patients with various forms of heart disease is attempting to deter- mine the cause of readmission of its patients. Should it rely of the results of general studies such as those described here or should it gather its own data, perform an analysis and draw its own conclusions? Support your recommendation.

SOURCES: Shinkman, Ron, “Readmissions Lead to $41.3B in Additional Hospital Costs,” FierceHealthFinance, April 20, 2014, www.fiercehealth finance.com/story/readmissions-lead-413b-additional-hospital-costs /2014-04-20; Kern, Christine, “5 Risk Factors for Unplanned Read- missions Identified,” Health IT Outcomes, September 8, 2015, www .healthitoutcomes.com/doc/risk-factors-for-unplanned-readmissions -identified-0001; van den Heede, Koen; Sermeus, Walter; Diya, Luwis; Lesaffre, Emmanuel; and Vleugels, Arthur, “Adverse Outcomes in Belgian Acute Hospitals: Retrospective Analysis of the National Hospital Discharge Dataset,” International Journal for Quality in Health Care, vol. 18, no. 3, 2006, pp. 211–219, Advance Access Publication: 23 March 2006, http://intqhc.oxfordjournals.org/content/intqhc/18/3/211.full.pdf; “Center for Evidence-Based Practice,” Penn Medicine, www.uphs.upenn .edu/cep/, accessed January 12, 2015; “ ‘Rothman Index’ May Help to Lower Repeat Hospitalization Risk,” Medical Press, August 15, 2013, http://medicalxpress.com/news/2013-08-rothman-index-hospitaliza tion.html. Bradley, Elizabeth, PhD; Yakusheva, Olga, PhD; Horwitz, Leora, MD; Sipsma, Heather, PhD; and Fletcher, Jason, PhD, “Identifying Patients at Increased Risk for Unplanned Readmission,” US National Library of Medicine National Institutes of Health, September 1, 2014, http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3771868/.

Case Two

Sunny Delight Improves Profitability with a Self-Service BI Solution When implementing a self-service analytics program, information systems staff and end users across an organization often must be willing to give up some control and autonomy in exchange for a cohesive data management strategy. Companies that effectively implement self-service analytics, however, usually find those trade-offs are outweighed by the competitive advantages gained for the organization as a whole.

For Sunny Delight Beverages, a Cincinnati-based producer of juice-based drinks, the payoff from using self- service analytics software has been significant. The company, which generates more than $550 million in annual revenue through sales of its SunnyD, Fruit20, and VeryFine brands, estimates that its newly implemented, self-service analytics program has resulted in a $195,000 annual reduction in staffing costs and a $2 million annual increase in profits.

Getting to these results, however, has not been easy for Sunny Delight. Like many companies, it had developed a patchwork of departmental business analytics applications

CHAPTER 9 • Business Intelligence and Analytics 405

over the years. Sunny Delight’s infrastructure was particularly complex as the company has been bought and sold multiple times since it was founded in 1963. At one point, Sunny Delight’s 480 employees were working with eight different legacy BI applications, resulting in some departments spending up to a week each month producing data that was often not in agreement with the data generated by other departments. Reconciling and rolling up the data was time consuming and left little time for in-depth analysis, much less strategy development and execution.

The data silos also meant that Sunny Delight had no real visibility into its business, which lead to revenue unpredictably, higher-than-necessary inventory levels, and lower margins. The company’s sales efforts were hampered because the sales team did not have a true understanding of the effectiveness and profitability of specific sales promotions. For example, the sales department was unable to correlate the impact of a promotional discount with order volume—a key metric for judging the effectiveness of a promotional program. The company was also unable to tie shipping costs directly to specific promotions, which was significant since the timing of many promotions required shipping products to stores on weekends, when shipping and warehouse labor costs were higher.

When the company made the decision to revamp its analytics efforts, the company’s CIO and CFO pulled together a cross-functional team of managers from sales, marketing, logistics, warehousing, and accounting who were responsible for developing a comprehensive picture of the required BI functionality—which ranged from simple, canned reports to complex, ad hoc data analysis tools. Working to understand each department’s needs built credibility for the project team and helped them choose the solution that would be most effective across the company, which they did after evaluating 17 different options.

The team selected Birst, a cloud-based, self-service BI solution that offers an end to data silos with what it refers to as “local execution with global governance.” Because the project team understood that a centrally managed data source was critical to ensuring consistent user-generated data and analysis across the company, they also opted to implement a data warehouse at the same time Birst was rolled out to employees.

According to John Gordos, Sunny Delight’s associate director of application development, Birst provides Sunny Delight with a single, networked source of data, which employees at all levels can access quickly and easily, regardless of where they work. Birst’s data governance features mean that Sunny Delight’s IS team maintains final control over all data, while the user-friendly interface, which is the same whether users are accessing data on a PC, laptop, or smartphone, makes it easy for nontechnical users to access and customize the system’s departmental dashboards.

With the data from the new system, Sunny Delight was able to create a more efficient production schedule that

allowed it to cut back on production, decrease inventory levels, and reduce plant overtime costs by 90 percent—all without impacting order fulfillment. And with a clearer picture of overall costs, the sales and distribution teams worked together to revise shipping schedules, resulting in a 7 percent drop in the transportation costs tied to promotions.

According to Gordos, “Birst helps [Sunny Delight] employees to think fast because they no longer have to worry about building and aggregating the data. They just get the data, and then they think about it—instead of accumulating it.”

Critical Thinking Questions 1. Is it surprising to you that a relatively small company

like Sunny Delight could end up with so many differ- ent analytics tools? How might the fact that Sunny Delight has changed ownership multiple times have impacted the number and variety of BI tools being used?

2. What are some of the trade-offs of a move to an enterprise-level analytics solution for individual end users who might have grown accustomed to working with their own customized solutions for generating data?

3. According to a recent report by Gartner, most busi- ness users will have access to some sort of self- service BI tool within the next few years; however, Gartner estimates that less than 10 percent of com- panies will have sufficient data governance practices in place to prevent data inconsistencies across the organization. Why do you think so many companies continue to invest in new analytics tools without implementing governance programs that ensure data consistency?

SOURCES: “Sunny Delight Beverages Co,” Sunny Delight Beverages, Co., ww2.sunnyd.com/company/overview.shtml, accessed March 16, 2016; Boulton, Clint, “How Sunny Delight Juices up Sales with Cloud-Based Analytics,” CIO, September 14, 2015, www.cio.com/article/2983624 /business-analytics/how-sunny-delight-juices-up-sales-with-cloud-based -analytics.html; “Birst Customer Testimonial: John Gordos - Associate Director, Application Development,” YouTube video, posted by BirstBI, August 14, 2014, www.youtube.com/watch?v=d3AjCIzWO5Y; “SunnyD Case Study,” Birst, February 1, 2016, www.google.com/url?sa=t&rct =j&q=&esrc=s&source=Web&cd=4&cad=rja&uact=8&ved=0ahUKEwi nieHr78XLAhXFwj4KHa-5CzMQFggtMAM&url=https%3A%2F%2Fwww .birst.com%2Fwp-content%2Fuploads%2F2016%2F02%2FBirst _CaseStudy_SunnyD_NetworkedBI.pdf&usg=AFQjCNFwbr-mGBWeu5zU mu_k40JAPOO_Mw&bvm=bv.116954456,d.amc; “Birst Sunny D Testi- monial,” YouTube video, posted by BirstBI, October 9, 2015, www .youtube.com/watch?v=tEuHH4IGHLU; Roberts, Shawn, “How Analytics Saved Sunny Delight $1M,” CIO Insight, October 14, 2015, www .cioinsight.com/it-strategy/big-data/how-analytics-saved-sunny-delight -1m.html; “Networked BI,” Birst, www.birst.com/product, accessed March 16, 2016.

Notes

1. “Helse Vest,” Microsoft, February 6, 2014, https://custo mers.microsoft.com/Pages/CustomerStory.aspx? recid=2223.

2. “MetLife Auto & Home Puts Brakes on Fraud with CSC’s Fraud Evaluator,” CSC, www.csc.com/p_and_c_genera l_insurance/success_stories/45406-metlife_auto_and_ho

406 PART 3 • Business Information Systems

me_puts_brakes_on_fraud_with_csc_s_fraud_evaluator, accessed January 8, 2016.

3. “Getting Started with Analytics: Kroger Uses Simulation Optimization to Improve Pharmacy Inventory Manage- ment,” INFORMS, www.informs.org/Sites/Getting -Started-With-Analytics/Analytics-Success-Stories/Case -Studies/Kroger, accessed January 8, 2016.

4. “Getting Started with Analytics: DaimlerChrysler: Using a Decision Support System for Promotional Pricing at the Major Auto Manufacturer, INFORMS, www.informs.org /Sites/Getting-Started-With-Analytics/Analytics-Success -Stories/Case-Studies/DaimlerChrysler, accessed January 8, 2016.

5. “Getting Started with Analytics: Optimizing Chevron’s Refineries,” INFORMS, https://www.informs.org/Sites /Getting-Started-With-Analytics/Analytics-Success-Stories /Case-Studies/Chevron, accessed January 8, 2016.

6. “Getting Started with Analytics: Coca-Cola Enterprises: Optimizing Product Delivery of 42 Billon Soft Drinks a Year,” INFORMS, www.informs.org/Sites/Getting -Started-With-Analytics/Analytics-Success-Stories/Case -Studies/Coca-Cola-Enterprises, accessed January 8, 2016.

7. Violino, Bob, “The Hottest Jobs in IT: Training Tomor- row’s Data Scientists,” Forbes, June 26, 2014, www.for bes.com/sites/emc/2014/06/26/the-hottest-jobs-in-it-train ing-tomorrows-data-scientists/#131733cd4b63.

8. “Argosy Hits the Jackpot with OpenText and Teradata,” Open Text, http://connectivity.opentext.com/resource -centre/success-stories/Success_Story_Argosy_Hits_the _ Jackpot_with_OTIC_and_Teradata.pdf.pdf, accessed January 19, 2015.

9. “UK Hospital Sees Cloud-Based BI Service as a Tool to Boost Clinical Outcomes and Efficiency,” Microsoft, http://blogs.msdn.com/b/powerbi/archive/2014/04/16 /uk-hospital-sees-cloud-based-bi-service-as-a-tool-to -boost-clinical-outcomes-and-efficiency.aspx, accessed February 8, 2015.

10. “Who We Are,” FFF Enterprises, www.fffenterprises.com /company/who-we-are.html, accessed January 20, 2014.

11. “At FFF Enterprises Collaboration Is Key to Success with QlikView,” Qlik, www.qlik.com/us/explore/customers /customer-listing/f/fff-enterprises, accessed January 20, 2015.

12. “Adviso and Technologia Use the Power of Multi- Channel Funnels to Discover the True Paths to Conver- sion,” Google Analytics, https://static.googleusercontent .com/media/www.google.com/en/us/analytics/customers /pdfs/technologia.pdf, accessed January 25, 2016.

13. “Starbucks Coffee Company Delivers Daily, Actionable Information to Store Managers, Improves Business Insight with High Performance Data Warehouse,” Ora- cle, www.oracle.com/us/corporate/customers/customer

search/starbucks-coffee-co-1-exadata-ss-1907993.html, accessed January 20, 2014.

14. Misson, Chris, “AQUARIUS WebPortal—a Flash Flooding Emergency Management Success Story,” Hydrology Cor- ner Blog, October 21, 2014, http://aquaticinformatics .com/blog/aquarius-webportal-flash-flooding-emergency -management.

15. Piatetsky, Gregory, “CRISP-DM, Still the Top Methodol- ogy for Analytics, Data Mining, or Data Science Projects,” KDNuggets, October 28, 2014, www.kdnuggets.com /2014/10/crisp-dm-top-methodology-analytics-data-min ing-data-science-projects.html.

16. “Data Mining Reveals the Surprising Factors behind Successful Movies,” MIT Technology Review, June 22, 2015, www.technologyreview.com/view/538701/data -mining-reveals-the-surprising-factors-behind-successful -movies.

17. McNulty, Eileen, “HP Rolls Out BI and Analytics Soft- ware Bundle,” dataconomy, June 10, 2014, http://datac onomy.com/hp-rolls-bi-analytics-software-bundle.

18. “Cognos Business Intelligence: Coming Soon to the Cloud,” IBM, www-03.ibm.com/software/products/en /business-intelligence, accessed January 19, 2015.

19. “Business Intelligence for Everyone,” Information Builders, www.informationbuilders.com/products/webfo cus, accessed January 19, 2015.

20. Lardinois, Frederic, “Microsoft’s Power BI for Office 365 Comes out of Preview, Simplifies Data Analysis and Visualizations,” Tech Crunch, February 10, 2014, http:// techcrunch.com/2014/02/10/microsofts-power-bi-for -office-365-comes-out-of-preview-simplifies-data-analy sis-and-visualizations.

21. “Oracle Business Intelligence,” Oracle, www.oracle.com /technetwork/middleware/index-084205.html, accessed January 19, 2015.

22. Rouse, Margaret, “Oracle Hyperion,” TechTarget, http:// searchfinancialapplications.techtarget.com/definition /Oracle-Hyperion, accessed January 19, 2015.

23. “SAS Enterprise BI Server,” SAS, www.sas.com/en_us /software/business-intelligence/enterprise-bi-server.html, accessed January 30, 2015.

24. Rouse, Margaret, “SAP Business Objects BI,” TechTarget, http://searchsap.techtarget.com/definition/SAP-Busines sObjects-BI, accessed January 19, 2015.

25. Ferranti, Marc, “IBM’s Watson Analytics Offers New Data Discovery Tools for Everyday Business Users,” PC World, October 13, 2015, www.pcworld.com/article /2992124/ibms-watson-analytics-offers-new-data-discov ery-tools-for-everyday-business-users.html.

26. Dwoskin, Elizabeth, “How New York’s Fire Department Uses Data Mining,” Digits, January 24, 2014, http://blogs .wsj.com/digits/2014/01/24/how-new-yorks-fire-depart ment-uses-data-mining.

CHAPTER 9 • Business Intelligence and Analytics 407

CHAPTER

10 Knowledge Management and Specialized Information Systems

docstockmedia/Shutterstock.com

Know?Did Yo u

• Watson, a supercomputer developed by IBM with artifi- cial intelligence capabilities, was able to soundly defeat two prior champions of the popular TV game show, Jeopardy! Now a cloud-based Watson system is being used by doctors to develop treatment options for a wide range of diseases.

• Some experts have predicted that every home will have a robot of some sort by 2025. One area of particular interest is the use of robots as companions and care- givers for people who are sick, elderly, or physically challenged.

Principles Learning Objectives

• Knowledge management allows organizations to share knowledge and experience among its workers.

• Discuss the relationships between data, informa- tion, and knowledge.

• Identify the benefits associated with a sound knowledge management program.

• List some of the tools and techniques used in knowledge management.

• Artificial intelligence systems form a broad and diverse set of systems that can replicate human decision making for certain types of well-defined problems.

• Define the term "artificial intelligence" and state the objective of developing artificial intelligence systems.

• List the characteristics of intelligent behavior and compare the performance of natural and artificial intelligence systems for each of these characteristics.

• Identify the major components of the artificial intelligence field and provide one example of each type of system.

• Multimedia and virtual reality systems can reshape the interface between people and infor- mation technology by offering new ways to com- municate information, visualize processes, and express ideas creatively.

• Discuss the use of multimedia in a business setting.

• Define the terms “virtual reality” and “augmented reality” and provide three examples of these applications.

• Specialized systems can help organizations and individuals achieve their goals.

• Discuss examples of specialized systems for organizational and individual use.

Why Learn about Knowledge Management and Specialized Information Systems? Knowledge management and specialized information systems are used in almost every industry. As a manager, you might use a knowledge management system to obtain advice on how to approach a problem that others in your organization have already encountered. As an executive at an automotive company, you might oversee robots that attach windshields to cars or paint body panels. As a stock trader, you might use a special system called a neural network to uncover patterns to make investment decisions. As a new car sales manager you might incorporate virtual reality into a company Web site to show potential customers the features of your various models. As a member of the military, you might use computer simulation as a training tool to prepare you for combat. As an employee of a petroleum company, you might use an expert system to determine where to drill for oil and gas. This chapter provides many additional examples of these types of specialized information systems. Learning about these systems will help you discover new ways to use information systems in your day-to-day work.

As you read this chapter, consider the following:

• What are the many uses of information systems designed to collect knowledge and provide expertise?

• Can specialized information systems and devices provide expertise superior to that which can be obtained through human effort?

This chapter identifies the challenges associated with knowledge manage- ment, provides guidance to overcome these challenges, presents best practices for selling and implementing a successful knowledge management project, and outlines various technologies that support knowledge management. We begin with a definition of knowledge management and identify several knowledge management applications and their associated benefits.

What Is Knowledge Management?

Knowledge management (KM) comprises a range of practices concerned with increasing awareness, fostering learning, speeding collaboration and innovation, and exchanging insights. Knowledge management is used by organizations to enable individuals, teams, and entire organizations to collec- tively and systematically create, share, and apply knowledge in order to achieve their objectives. Globalization, the expansion of the services sector, and the emergence of new information technologies have caused many orga- nizations to establish KM programs in their IT or human resource manage- ment departments. The goal is to improve the creation, retention, sharing, and reuse of knowledge. As already discussed, a knowledge management sys- tem is an organized collection of people, procedures, software, databases, and devices that creates, captures, refines, stores, manages, and disseminates knowledge, as shown in Figure 10.1.

An organization’s knowledge assets often are classified as either explicit or tacit (see Table 10.1). Explicit knowledge is knowledge that is documen- ted, stored, and codified—such as standard procedures, product formulas, customer contact lists, market research results, and patents. Tacit knowledge is the know-how that someone has developed as a result of personal experi- ence; it involves intangible factors such as beliefs, perspective, and a value system. Examples include how to ride a bike, the decision-making process used by an experienced coach to make adjustments when her team is down at halftime of a big game, a physician’s technique for diagnosing a rare illness and prescribing a course of treatment, and an engineer’s approach to cutting

knowledge management (KM): A range of practices concerned with increasing awareness, fostering learn- ing, speeding collaboration and inno- vation, and exchanging insights.

explicit knowledge: Knowledge that is documented, stored, and codified—such as standard proce- dures, product formulas, customer contact lists, market research results, and patents.

tacit knowledge: The know-how that someone has developed as a result of personal experience; it involves intangible factors such as beliefs, perspective, and a value system.

410 PART 3 • Business Information Systems

costs for a project that is over budget. This knowledge cannot be documented easily; yet, tacit knowledge is key to high performance and competitive advantage because it’s difficult for others to copy.

Much of the tacit knowledge that people carry with them is extremely useful but cannot be shared with others easily. This means that new employ- ees might spend weeks, months, or even years learning things on their own that more experienced coworkers might have been able to convey to them. In some cases, these nuggets of valuable knowledge are lost forever when experienced employees retire, and others never learn them.

A major goal of knowledge management is to somehow capture and doc- ument the valuable work-related tacit knowledge of others and to turn it into explicit knowledge that can be shared with others. This is much easier said than done, however. Over time, experts develop their own processes for their areas of expertise. Their processes become second nature and are so internal- ized that they are sometimes unable to write down step-by-step instructions to document the processes.

Two processes are frequently used to capture tacit knowledge— shadowing and joint problem solving. Shadowing involves a novice observ- ing an expert executing her job to learn how she performs. This technique is often used in the medical field to help young interns learn from experienced physicians. With joint problem solving, the novice and the expert work side by side to solve a problem so that the expert’s approach is slowly revealed to the observant novice. Thus a plumber trainee will work with a master plumber to learn the trade.

Create

Disseminate Refine

StoreManage

Capture

Knowledge base

TABLE 10.1 Explicit and tacit knowledge Asset Type Description Examples

Explicit knowledge Knowledge that is documented, stored, and codified

Customer lists, product data, price lists, a database for telemarketing and direct mail, patents, best practices, standard procedures, and market research results

Tacit knowledge Personal knowledge is not documented but embedded in individual experience

Expertise and skills unique to individual employ- ees, such as how to close a sale or troubleshoot a complex piece of equipment

shadowing: A process used to cap- ture tacit knowledge that involves a novice observing an expert executing her job to learn how she performs.

joint problem solving: A process used to capture tacit knowledge where the novice and the expert work side by side to solve a problem so that the expert’s approach is slowly revealed to the observant novice.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 411

The next section discusses how KM is used in organizations and illustrates how these applications lead to real business benefits.

Knowledge Management Applications and Associated Benefits Organizations employ KM to foster innovation, leverage the expertise of people across the organization, and capture the expertise of key individuals before they retire. Examples of knowledge management efforts that led to these results and their associated benefits are discussed in the following sections.

Foster Innovation by Encouraging the Free Flow of Ideas Organizations must continuously innovate to evolve, grow, and prosper. Organizations that fail to innovate will soon fall behind their competition. Many organizations implement knowledge management projects to foster innovation by encouraging the free flow of ideas among employees, contrac- tors, suppliers, and other business partners. Such collaboration can lead to the discovery of a wealth of new opportunities, which, after evaluation and testing, may lead to an increase in revenue, a decrease in costs, or the crea- tion of new products and services.

TMW Systems, a provider of logistics operations and fleet management systems, has experienced rapid growth over the past several years. The com- pany now has more than 700 employees in its Cleveland headquarters and in satellite offices across North America.1 As the company grew, it became more difficult to ensure that all employees had access to the most current knowl- edge available within the company because most information was shared via email or shared network drives. To ensure that its base of institutional knowl- edge was being preserved and to encourage collaboration and innovation across all of its offices, TMW implemented a knowledge management system with a strong social learning feature, which places an emphasis on decentra- lized information sharing. The system, which is part of an intentional cultural shift at TMW toward a more open and collaborative working environment, allows employees to easily share new ideas, expertise, and best practices with other employees no matter where they are based.2

Leverage the Expertise of People across the Organization It is critical that an organization enables its employees to share and build on one another’s experience and expertise. In this manner, new employees or employees moving into new positions are able to get up to speed more quickly. Workers can share thoughts and experiences about what works well and what does not, thus preventing new employees from repeating some of the mistakes of others. Employees facing new (to them) challenges can get help from coworkers in other parts of the organization whom they have never even met to avoid a costly and time-consuming “reinvention of the wheel.” All of this enables employees to deliver valuable results more quickly, improve their productivity, and get products and new ideas to market faster.

White & Case, an international law firm headquartered in New York City, represents well-known organizations around the world through its offices in more than 20 countries in Africa, Asia, Europe, Latin America, the Middle East, and North America. The firm’s employees have diverse backgrounds and speak more than 60 different languages.3 One strength of the firm is that the lawyers truly operate as a team by constantly sharing know-how, experience, and market and client information. Thus, a client anywhere in the world receives the full benefit of White & Case’s global knowledge. The firm’s knowl- edge management system pulls relevant information from the firm’s document management, CRM, case management, and billing and financial management systems as well as from lawyers’ work histories to create a context for all this information. The system enables lawyers to find all relevant knowledge within

412 PART 3 • Business Information Systems

the firm about a case or subject, often within a matter of minutes. As a result of its ability to leverage its global knowledge, White & Case has been able to win new business, such as that of a major manufacturing company that approached the firm’s New York office to find out if it had expertise in privatizing an Eastern European company. Using the enterprise search software, an attorney in New York quickly determined that the company had experience in this area and that the best lawyer for the job was working out of the firm’s Germany office.4

Capture the Expertise of Key Individuals before They Retire In the United States, 3 to 4 million employees will retire each year for the next 20 years or so. Add to that a 5 to 7 percent employee turnover as workers move to different companies, and it is clear that organizations are facing a tremendous challenge in trying to avoid the loss of valuable experience and expertise. Many organizations are using knowledge management to capture this valuable exper- tise before it simply walks out the door and is lost forever. The permanent loss of expertise related to the core operations of an organization can result in a sig- nificant loss of productivity and a decrease in the quality of service over time.

The state of New Hampshire has developed a knowledge management and transfer model to prevent critical knowledge loss as state employees retire. The process begins by identifying what critical tasks the individual per- forms and assessing whether others can perform these tasks. To do this, the employee is asked to answer questions such as the following:

● If you left your position today, what wouldn’t get done because no one else knows how to do it?

● How important is this work? What is the impact of it not getting done? ● If this work is important, what resources exist to help others learn this

task? ● If this work is important, how should we plan to address this knowledge

gap? Who will learn this? How and when?

Following this discussion, the employee and his manager define appropri- ate methods to transfer any critical knowledge. This could include transferring the knowledge to others, creating job aids, providing on-the-job training for a replacement, and so on.5

Best Practices for Selling and Implementing a KM Project Establishing a successful KM program is challenging, but most of the chal- lenges involved have nothing to do with the technologies or vendors employed. Instead they are challenges associated with human nature and the manner in which people are accustomed to working together. A set of best practices for selling and implementing a KM project are summarized in Figure 10.2 and are discussed in the following sections.

Connect KM to goals and objectives

Start with small pilot and enthusiastic

participants

Identify valuable tacit knowledge

Get employee buy in

FIGURE 10.2 Steps in selling and implementing a knowledge management project The key challenges have to do with human nature and how people work together. Source: From Reynolds, Information Technology for Managers, 2E. © 2016 Cengage Learning.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 413

Connect the KM Effort to Organizational Goals and Objectives When starting a KM effort, just as with any other project, you must clearly define how that effort will support specific organizational goals and objec- tives, such as increasing revenue, reducing costs, improving customer service, or speeding up the time to bring a product to market. Doing so will help you sell the project to others and elicit their support and enthusiasm; it will also help you determine if the project is worthwhile before the organization com- mits resources to it. Although many people may intuitively believe that shar- ing knowledge and best practices is a worthy idea, there must be an underlying business reason to do so. The fundamental business case for implementing knowledge management must be clearly defined.

Start with a Small Pilot Involving Enthusiasts Containing the scope of a project to impact only a small part of the organiza- tion and a few employees is definitely less risky than trying to take on a proj- ect very large in scope. With a small-scale project, you have more control over the outcome, and if the outcome is not successful, the organization will not be seriously impacted. Indeed, failure on a small scale can be considered a learn- ing experience on which to build future KM efforts. In addition, obtaining the resources (people, dollars, etc.) for a series of small, successful projects is typ- ically much easier than getting large amounts of resources for a major organization-wide project.

Furthermore, defining a pilot project to address the business needs of a group of people who are somewhat informed about KM and are enthusiastic about its potential can greatly improve the odds of success. Targeting such a group of users reduces the problem of trying to overcome skepticism and unwillingness to change, which have doomed many a project. Also, such a group of users, once the pilot has demonstrated some degree of success, can serve as strong advocates who communicate the positive business benefits of KM to others.

Identify Valuable Tacit Knowledge Not all tacit knowledge is equally valuable, and priorities must be set in terms of what knowledge to go after. The intent of a KM program is to identify, cap- ture, and disseminate knowledge gems from a sea of information. Within the scope of the initial pilot project, an organization should identify and prioritize the tacit knowledge it wants to include in its KM system.

As Toyota Financial Services (TFS) was preparing to move its North Amer- ican headquarters from California to Texas, the company implemented a cloud-based knowledge management system, with the goal of ensuring that the important tacit knowledge of its employees was not lost during the transi- tion. In the two-year lead-up to the move, TFS employees were encouraged to use the platform to post questions and share expertise, allowing the company to develop a repository of curated subject matter expertise that the company will use to train new employees in Texas.6

Get Employees to Buy In Managers must create a work culture that places a high value on tacit knowl- edge and that strongly encourages people to share it. In a highly competitive work environment, it can be especially difficult to get workers to surrender their knowledge and experience as these traits make the employees more valuable as individual contributors. For example, it would be extremely diffi- cult to get a highly successful mutual fund manager to share her stock- picking technique with other fund managers. Such sharing of information would tend to put all fund managers on a similar level of performance and would also tend to level the amount of their annual compensation.

414 PART 3 • Business Information Systems

Some organizations believe that the most powerful incentive for experts to share their knowledge is to receive public recognition from senior man- agers and their peers. For example, some organizations provide recognition by mentioning the accomplishments of contributors in a company email or newsletter, or during a meeting. Other companies identify knowledge sharing as a key expectation for all employees and even build this expectation into the employees’ formal job performance reviews. Many organizations provide incentives in a combination of ways—linking KM directly to job performance, creating a work environment where sharing knowledge seems like a safe and natural thing to do, and recognizing people who contribute.

Technologies That Support KM We are living in a period of unprecedented change where the amount of available knowledge is expanding rapidly. As a result, there is an increasing need for knowledge to be quality filtered and distributed to people in a more specific task relevant and timely manner. Technology is needed to acquire, produce, store, distribute, integrate, and manage this knowledge. Organiza- tions interested in piloting KM should be aware of the wide range of technol- ogies that can support KM efforts. These include communities of practice, organizational network analysis, a variety of Web 2.0 technologies, business rules management systems, and enterprise search tools. These technologies are discussed in the following sections.

Communities of Practice A community of practice (CoP) is a group whose members share a common set of goals and interests and regularly engage in sharing and learning as they strive to meet those goals. A community of practice develops around topics that are important to its members. Over time, a CoP typically develops resources such as models, tools, documents, processes, and terminology that represent the accumulated knowledge of the community. It is not uncommon for a CoP to include members from many different organizations. CoP has become associated with knowledge management because participation in a CoP is one means of developing new knowledge, stimulating innovation, or sharing existing tacit knowledge within an organization.

The origins and structures of CoPs vary widely. Some may start up and organize of their own accord; in other cases, there may be some sort of orga- nizational stimulus that leads to their creation. Members of an informal CoP typically meet with little advanced planning or formality to discuss problems of interest, share ideas, and provide advice and counsel to one another. Mem- bers of a more formal CoP meet on a regularly scheduled basis with a planned agenda and identified speakers.

The General Services Administration (GSA) was established in 1949 to streamline the administrative work of the federal government.7 The GSA’s Office of Citizen Services and Innovative Technologies supports eight differ- ent communities of practice that focus on topics such as crowdsourcing, citi- zen science, and mobile government.8 Recently, the agency founded an interagency CoP focused on improving the delivery of government services. The Customer Experience Community of Practice (CX-COP), which has close to 600 members from more than 140 federal, state, and local U.S. government offices and agencies, provides government customer experience professionals a hub where they can gather—both online and in person—to ask questions, share best practices, and collaborate.9,10

Organizational Network Analysis Organizational network analysis (ONA) is a technique used for document- ing and measuring flows of information among individuals, workgroups,

community of practice (CoP): A group whose members share a com- mon set of goals and interests and regularly engage in sharing and learn- ing as they strive to meet those goals.

organizational network analy- sis (ONA): A technique used for documenting and measuring flows of information among individuals, work- groups, organizations, computers, Web sites, and other information sources.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 415

organizations, computers, Web sites, and other information sources (see Figure 10.3). Each node in the diagram represents a knowledge source; each link represents a flow of information between two nodes. Many software tools support organizational network analysis, including Cytoscape, Gephi, GraphChi, NetDraw, NetMiner, NetworkX, and UCINET.

In analyzing social media communications from sources such as text, video, and chat as well as “likes” and “shares,” many experts agree that the most sig- nificant data isn’t the content itself, but rather the metadata that connects vari- ous pieces of content to form a complete picture. Metadata is data that describes other data. For instance, metadata about social media use could relate to properties of the message (e.g., whether the message is comical, sarcastic, genuine, or phony) and of the author (e.g., sex, age, interests, political view- point, ideological beliefs, and degree of influence on the audience). The meta- data enables analysts to make judgments about how to interpret and value the content of the message. Without the important metadata, it is not possible to know the value of the communications and how to take effective action.11

ONA has many knowledge management applications, ranging from map- ping knowledge flows and identifying knowledge gaps within organizations to helping establish collaborative networks. ONA provides a clear picture of how geographically dispersed employees and organizational units collaborate (or don’t collaborate). Organizations frequently employ ONA as part of a larger organizational network analysis to identify subject experts and then set up mechanisms (e.g., communities of practice) to facilitate the passing of knowledge from those experts to colleagues. Software programs that track email and other kinds of electronic communications may be used to identify in-house experts.

Westwood Professional Services, Inc., a Minnesota-based engineering and survey firm, recently conducted an organizational network analysis of one of its business units, which had been experiencing extensive change due to rapid growth. Through the analysis, the company was able to determine which employees within the business unit were connecting most frequently and the extent to which members of different subteams were collaborating.

FIGURE 10.3 Organizational network analysis Each node in the diagram represents a knowledge source; each link represents a flow of information between two nodes.

Steve Alan

Johnny

Carson

Laura

George Carlin

Rodney

Bob

Hope

Frank Jim

Sara

Jane

Stevie

Emma

Alice Susan

Aaron

metadata: Data that describes other data.

416 PART 3 • Business Information Systems

The company used the results of the analysis to develop targeted team- building initiatives and reevaluate its organizational structures to ensure they continued to support information sharing and creative problem solving—both high priorities for the company.12

Web 2.0 Technologies “Web 2.0” is a term describing changes in technology and Web site design to enhance information sharing, collaboration, and functionality on the Web. Major corporations have integrated Web 2.0 technologies such as blogs, for- ums, podcasts, RSS newsfeeds, and wikis to support knowledge management to improve collaboration, encourage knowledge sharing, and build a corpo- rate memory. For example, many organizations are using Web 2.0 technolo- gies such as podcasts and wikis to capture the knowledge of longtime employees, provide answers to cover frequently asked questions, and save time and effort in training new hires.

Business Rules Management Systems Change is occurring all the time and at a faster and faster pace—changes in economic conditions, new government and industry rules and regulations, new competitors, product improvements, new pricing and promotion strate- gies, and on and on. Organizations must be able to react to these changes quickly to remain competitive. The decision logic of the operational systems that support the organization—systems such as order processing, pricing, inventory control, and customer relationship management (CRM)—must con- tinually be modified to reflect these business changes. Decision logic, also called business rules, includes policies, requirements, and conditional state- ments that govern how the systems work.

The traditional method of modifying the decision logic of information sys- tems involves heavy interaction between business users and IT analysts work- ing together over a period of weeks, or even months, to define new systems requirements and then to design, implement, and test the new decision logic. Unfortunately, this approach to handling system changes is often too slow, and in some cases, results in incorrect system changes.

A business rule management system (BRMS) is software used to define, execute, monitor, and maintain the decision logic that is used by the operational systems and processes that run the organization. A BRMS enables business users to define, deploy, monitor, and maintain organizational poli- cies and the decisions flowing from those policies—such as claim approvals, credit approvals, cross-sell offer selection, and eligibility determinations— without requiring involvement from IT resources. This process avoids a potential bottleneck and lengthy delays in implementing changes and improves the accuracy of the changes.

BRMS components include a business rule engine that determines which rules need to be executed and in what order. Other BRMS components include an enterprise rules repository for storing all rules, software to manage the various versions of rules as they are modified, and additional software for reporting and multiplatform deployment. Thus, a BRMS can become a reposi- tory of important knowledge and decision-making processes that includes the learnings and experiences of experts in the field. The creation and mainte- nance of a BRMS can become an important part of an organization’s knowl- edge management program.

BRMS is increasingly used to manage the changes in decision logic in applications that support credit applications, underwriting, complex order processing, and difficult scheduling. The use of BRMS leads to faster and more accurate implementation of necessary changes to organizations’ policies and procedures. Table 10.2 lists several business rule management software vendors and their products.

business rule management system (BRMS): Software used to define, execute, monitor, and maintain the decision logic that is used by the operational systems and processes that run the organization.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 417

HanseMerkur Krankenversicherung is a German health insurance com- pany. The firm developed a BRMS to replace the time-consuming manual pro- cesses required to confirm insurance coverage with an automatic reconciliation of the information extracted from invoices (contract type, ser- vice submitted, insured party, billing amount, etc.). Automation of this and many of its other billing processes enabled HanseMerkur to maintain its level of service with no increase in staff even though the number of customers tri- pled from 366,000 to 1.2 million over the course of six years.13

Adobe is a digital marketing and digital media solutions provider whose products include Adobe Creative Cloud, a cloud-based subscription service; Adobe Digital Publishing Suite, which enables users to create, distribute, and optimize content for tablets; Adobe Photoshop for working with digital images; and Adobe Acrobat, which supports communication and collaboration on documents and other content both inside and outside an organization.14

Maintaining the rules needed for effective and efficient territory assignment and sales-lead distribution was a significant challenge for Adobe given its size, number of customers, personnel turnover, and the geographic distribu- tion of its sales force and product lines. To enable the company to react quickly to changes within its sales organization, Adobe implemented a BRMS system that includes tools that allow for the routine shifting of assignments due to personnel changes as well as more complex annual go-to-market terri- tory changes.15

DBS (formerly known as the Development Bank of Singapore) is a lead- ing financial services group in Asia and a leading consumer bank in Singapore and Hong Kong. The bank has a growing presence across Asia, and it serves more than 4 million customers, including 1 million retail customers through 250 branches.16 Assessing the risk and creditworthiness of individuals and businesses is a critical activity for DBS. However, until recently, this was an error-prone, labor-intensive process built around the completion of a ques- tionnaire during an interview between a relationship manager and an appli- cant. During the interview, applicants could provide any answer they chose as the process was not linked directly to any data. To revamp its inadequate credit reporting system, DBS implemented a BRMS that relies on verifiable customer and credit data rather than on unsubstantiated information supplied by the applicant. The BRMS supports eight different scoring models, each with hundreds of rules and hundreds of factors that go into a score. The rules are derived from a combination of regulatory sources, such as the Monetary Authority of Singapore, and statistical analysis performed by the bank’s credit portfolio analytics department. DBS has greatly improved its credit model with a resulting reduction in risk. As a result, the amount of financial reserves required to cover unanticipated losses has been reduced. The BRMS also allows the bank to quickly adapt the rules and factors inherent in its credit

TABLE 10.2 Business rule management software Software Manufacturer Product

Appian Corporation Business Process Management (BPM) Software Suite

Bosch Software Innovations inubit BPM

CA Aion Business Rules Expert

IBM Operational Decision Manager

Open Source Process Maker BPM

Oracle Business Rules

Pegasystems Pega Business Rules Platform

Progress Corticon

418 PART 3 • Business Information Systems

reporting process to respond to new opportunities and changing business conditions.17

Enterprise Search Software Enterprise search is the application of search technology to find information within an organization. Enterprise search software matches a user’s query to many sources of information in an attempt to identify the most important con- tent and the most reliable and relevant source.

Enterprise search software indexes documents from a variety of sources— such as corporate databases, departmental files, email, corporate wikis, and document repositories. When a search is executed, the software uses the index to present a list of relevance-ranked documents from these various sources. The software must be capable of implementing access controls so users are restricted to viewing only documents to which they have been granted access. Enterprise search software may also allow employees to move selected information to a new storage repository and apply controls to ensure that the files cannot be changed or deleted. Table 10.3 lists a number of enter- prise search products.18

Members of IT and human resources organizations may use enterprise search software to enforce corporate guidelines on the storage of confidential data on laptops that leave the office, and governance officials may use it to ensure that all guidelines for the storage of information are being followed.

Founded in the 1890s as a printing press manufacturer, Harris Corpora- tion now generates $1 billion in annual revenue through the sale of communi- cation services and systems to clients in a wide range of industries, including avionics, defense, energy, government, health care, and transportation.19

Harris employs more than 15,000 people across 125 countries, including 3,000 engineers in the Government Communications Systems Division (GCSD). The engineers in this division work in 12 different offices and need a more effective way to access existing information assets in a variety of repositories—all while maintaining very high levels of security. GCSD imple- mented a unified enterprise search platform that allows engineers to search various company databases, file shares, SharePoint sites, and the company intranet using one search engine. Access permissions are granted through the system on a “need-to-know” basis. The search engine indexes document files as well as videos, which the company uses extensively to record meetings and training sessions. The software has cut down on the time engineers spend looking for relevant information; enabled faster, incremental innovation that is built on knowledge accrued through previous projects; and increased collaboration among employees in different offices.20

Enterprise search software can also be used to support Web site visitor searches. It is critical that such software returns meaningful results to ensure

TABLE 10.3 Enterprise search solutions Software Manufacturer Software Product

Attivio Active Intelligence Engine

BA Insight Knowledge Integration Platform

Coveo Enterprise Search & Relevance

Dassault Systemes Exalead CloudView

Google Google Search Appliance

HP HP Autonomy

Mark Logic Corporation Mark Logic

enterprise search: The application of search technology to find information within an organization.

enterprise search software: Software that matches a user’s query to many sources of information in an attempt to identify the most important content and the most reliable and rele- vant source.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 419

Critical Thinking Exercise

that visitors get search results that meet their needs, thus increasing the rate at which Web site visitors convert to paying customers and are encouraged to spend more time at your site.

Electronic discovery is another important application of enterprise search software. Electronic discovery (e-discovery) refers to any process in which electronic data is sought, located, secured, and searched with the intent of using it as evidence in a civil or criminal legal case. The Federal Rules of Civil Procedure governs the processes and requirements of parties in federal civil suits and sets the rules regarding e-discovery. These rules compel civil litigants to both preserve and produce electronic documents and data related to a case, such as email, voice mail, texts, graphics, photographs, contents of databases, spreadsheets, Web pages, and so on. “We can’t find it” is no lon- ger an acceptable excuse for not producing information relevant to a lawsuit.

Effective e-discovery software solutions preserve and destroy data based on approved organizational policies through processes that cannot be altered by unauthorized users. To be useful, this software must also allow users to locate all of the information pertinent to a lawsuit quickly, with a minimum amount of manual effort. Furthermore, the solution must work for all data types across dissimilar data sources and systems and operate at a reasonable cost. The legal departments of many organizations are collaborating with their IT organization and technology vendors to identify and implement a solution that meets these e-discovery requirements.

A recent development in the area of e-discovery is the use of “predictive coding,” which involves the use of computer algorithms rather than a full manual review to determine the relevance of electronic documents in a law- suit. The algorithms are developed by having attorneys and other people knowledgeable about the case documents review a subset of documents; the results of that review are used to “teach” the software what to search for.21

Ideally, predictive coding would greatly reduce the costs and time required to complete e-discovery in large-scale litigations. In practice, however, this pro- cess can be another area of contention in a lawsuit. In Rio Tinto v. Vale, S.A. et al., a much-watched case in which predictive coding was used, the parties to the lawsuit spent months arguing over the protocols being used to develop the coding algorithms.22

KM Experiment You are a talent scout for a professional sports team. Over the years, the players you have recommended have had outstanding performance records for your team. Indeed, although you are only in your late thirties, you are frequently cited as one of the top talent recruiters in the league.

You have read and reread the study guide on knowledge management your general manager provided you two weeks ago. In addition to some basic defini- tions and discussion of KM, it includes several examples of successful applications of KM to the selection of top recruits for academic and athletic scholarships. Now you are sitting in your hotel room staring at the email from the general manager. He wants you to become the subject of a KM experiment for the team. The goal is to train the other three talent scouts for the team in your approach.

Review Questions 1. Do you think that shadowing or joint problem solving would be a more effec-

tive way to share your tacit knowledge with the other scouts? Why? 2. Would the formation of a community of practice be an effective method for

sharing tacit knowledge among all the scouts? How might such a CoP operate?

electronic discovery (e-discovery): Any process in which electronic data is sought, located, secured, and searched with the intent of using it as evidence in a civil or criminal legal case.

420 PART 3 • Business Information Systems

Critical Thinking Questions 1. For years, you have competed against the other scouts to sign the most pro-

ductive new talent. How do you feel about being asked to share your insights and expertise with the other scouts?

2. What sort of bonus or incentive could management offer to make you strongly motivated to participate in this experiment and make it a success?

Overview of Artificial Intelligence

At a Dartmouth College conference in 1956, John McCarthy proposed the use of the term artificial intelligence (AI) to describe computers with the ability to mimic or duplicate the functions of the human brain. A paper was pre- sented at the conference proposing a study of AI based on the conjecture that “every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it.”23 Many AI pioneers attended this first conference; a few predicted that computers would be as “smart” as people by the 1960s. The prediction has not yet been realized, but many applications of artificial intelligence can be seen today, and research continues.

Watson is the first commercially available cognitive computing capability, a computer capable of processing information like a human. As such, Watson represents a new era in computing. The system, delivered through the cloud, analyzes large volumes of data, understands complex questions posed in natu- ral language, and proposes evidence-based answers. Watson continuously learns, gaining in value and knowledge over time, from previous interactions. Watson learns in three ways: by being taught by its users, by learning from prior interactions, and by being presented with new information.24 An early version of Watson was able to soundly defeat prior champions of the popular TV game show, Jeopardy! The artificial intelligence computer can process human speech, search its vast databases for possible responses, and reply in a human voice. See Figure 10.4.

FIGURE 10.4 IBM Watson IBM Watson is being used to develop treatment options for cancer patients based on the DNA of their disease Ca

ro ly n Co le /G et ty Im ag es

artificial intelligence: The ability to mimic or duplicate the functions of the human brain.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 421

Now a cloud-based Watson system is being used by doctors to develop treatment options for a wide range of diseases, including glioblastoma, an aggressive brain cancer that kills over 13,000 people in the United States each year. Watson will correlate data from the DNA associated with each glioblas- toma patient’s disease to the latest findings from medical journals, new stud- ies, medical images, and clinical records to develop a highly personalized treatment regimen. The goal is for Watson to increase the number of patients who can benefit from care options uniquely tailored to their disease’s DNA. Watson will continually learn and improve as it deals with each new patient scenario and new medical research becomes available.

Artificial Intelligence in Perspective Computers were originally designed to perform simple mathematical opera- tions, using fixed programmed rules and eventually operating at millions of computations per second. When it comes to performing mathematical opera- tions quickly and accurately, computers beat humans’ hands down. However, computers still have trouble recognizing patterns, adapting to new situations, and drawing conclusions when not provided complete information—all activi- ties that humans can perform quite well. Artificial intelligence systems tackle these sorts of problems. Artificial intelligence systems include the people, procedures, hardware, software, data, and knowledge needed to develop computer systems and machines that can simulate human intelligence pro- cesses, including learning (the acquisition of information and rules for using the information), reasoning (using rules to reach conclusions), and self- correction (using the outcome from one scenario to improve its performance on future scenarios).

AI is a complex and interdisciplinary field that involves several specialties, including biology, computer science, linguistics, mathematics, neuroscience, philosophy, and psychology. The study of AI systems causes one to ponder philosophical issues such as the nature of the human mind and the ethics of creating objects gifted with human-like intelligence. Today, artificial intelli- gence systems are used in many industries and applications. Researchers, scientists, and experts on how human beings think are often involved in developing these systems.

Nature of Intelligence From its earliest stages, the emphasis of much AI research has been on developing machines with the ability to “learn” from experiences and apply knowledge acquired from those experiences; to handle complex situations; to solve problems when important information is missing; to determine what is important and to react quickly and correctly to a new situation; to understand visual images, process and manipulate symbols, and be creative and imaginative; and to use heuristics—all of which together is considered intelligent behavior.

The Turing Test, designed by Alan Turing, a British mathematician, attempts to determine whether a computer can successfully impersonate a human. Human judges are connected to the computer and to another human via an instant messaging system and the only information flowing between the contestants is text. The judges pose questions on any topic from the arts to zool- ogy, even questions about personal history and social relationships. To pass the test, the computer must communicate via this medium so competently that the judges cannot tell the difference between the computer’s responses and the human’s responses.25 No computer has yet passed the Turing Test, although many computer scientists believe it may happen in the next few years.26 The Loebner Prize is an annual competition in artificial intelligence that awards prizes to the computer system designed to simulate an intelligent conversation

artificial intelligence system: The people, procedures, hardware, software, data, and knowledge needed to develop computer systems and machines that can simulate human intelligence processes, including learn- ing (the acquisition of information and rules for using the information), rea- soning (using rules to reach conclu- sions), and self-correction (using the outcome from one scenario to improve its performance on future scenarios).

intelligent behavior: The ability to learn from experiences and apply knowledge acquired from those experi- ences; to handle complex situations; to solve problems when important infor- mation is missing; to determine what is important and to react quickly and cor- rectly to a new situation; to understand visual images, process and manipulate symbols, and be creative and imagina- tive; and to use heuristics.

422 PART 3 • Business Information Systems

that is considered by the judges to be the most humanlike. Although no one has yet won the $100,000 top prize for passing the Turing Test, each year a prize of $4,000 and a bronze medal is awarded for the computer that is con- sidered by the judges to be the most humanlike.27

Some of the specific characteristics of intelligent behavior include the abil- ity to do the following:

● Learn from experience and apply the knowledge acquired from expe- rience. Learning from past situations and events is a key component of intelligent behavior and is a natural ability of humans, who learn by trial and error. This ability, however, must be carefully programmed into a computer system. Today, researchers are developing systems that can “learn” from experience. The 20 Questions (20Q) Web site, www.20q.net (see Figure 10.5), is an example of a system that learns.28 The Web site is an artificial intelligence game that learns as people play.

● Handle complex situations. In a business setting, top-level managers and executives must handle a complex market, challenging competitors, intricate government regulations, and a demanding workforce. Even human experts make mistakes in dealing with these matters. Very careful planning and elaborate computer programming are necessary to develop systems that can handle complex situations.

● Solve problems when important information is missing. An integral part of decision making is dealing with uncertainty. Often, decisions must be made with little or inaccurate information because obtaining complete information is too costly or impossible. Today, AI systems can make important calculations, comparisons, and decisions even when informa- tion is missing.

● Determine what is important. Knowing what is truly important is the mark of a good decision maker. Developing programs and approaches to allow computer systems and machines to identify important information is not a simple task.

● React quickly and correctly to a new situation. A small child, for exam- ple, can look over an edge and know not to venture too close. The child reacts quickly and correctly to a new situation. On the other hand, with- out complex programming, computers do not have this ability.

FIGURE 10.5 The 20Q Web site 20Q is a game where users play the popular game, 20 Questions, against an artificial intelligence foe. Source: www.20q.net

CHAPTER 10 • Knowledge Management and Specialized Information Systems 423

● Understand visual images. Interpreting visual images can be extremely difficult, even for sophisticated computers. Moving through a room of chairs, tables, and other objects can be trivial for people but extremely complex for machines, robots, and computers. Such machines require an extension of understanding visual images, called a perceptive system. Having a perceptive system allows a machine to approximate the way a person sees, hears, and feels objects.

● Process and manipulate symbols. People see, manipulate, and process symbols every day. Visual images provide a constant stream of information to our brains. By contrast, computers have difficulty handling symbolic pro- cessing and reasoning. Although computers excel at numerical calculations, they aren’t as good at dealing with symbols and three-dimensional objects. Recent developments in machine-vision hardware and software, however, allow some computers to process and manipulate certain symbols.

● Be creative and imaginative. Throughout history, some people have turned difficult situations into advantages by being creative and imagina- tive. For instance, when defective mints with holes in the middle arrived at a candy factory, an enterprising entrepreneur decided to market these new mints as LifeSavers instead of returning them to the manufacturer. Ice cream cones were invented at the St. Louis World’s Fair when an imaginative store owner decided to wrap ice cream with a waffle from his grill for portability. Developing new products and services from an exist- ing (perhaps negative) situation is a human characteristic. While software has been developed to enable a computer to write short stories, few com- puters can be imaginative or creative in this way.

● Use heuristics. For some decisions, people use heuristics (rules of thumb arising from experience) or even guesses. Some computer systems obtain good solutions to complex problems (e.g., scheduling the flight crews for a large airline) based on heuristics rather than trying to search for an optimal solution, which might be technically difficult or too time consuming.

Brain-Computer Interface Developing a link between the human brain and the computer is another excit- ing aspect of artificial intelligence research. The idea behind a brain-computer interface (BCI) is to directly connect the human brain to a computer so that human thought can control the activities of the computer. One potential use of BCI technology would be to give people without the ability to speak or move (a condition called locked-in syndrome) the capability to communicate, control a computer, and move artificial limbs. Honda Motors has developed a BCI sys- tem that allows a person to complete certain operations, such as bending a leg, with 90 percent accuracy. See Figure 10.6. The new system uses a special helmet that can measure and transmit brain activity to a computer.

AI is a broad field that includes several specialty areas, such as expert sys- tems, robotics, vision systems, natural language processing, learning systems, and neural networks. See Figure 10.7. Many of these areas are related; advances in one can occur simultaneously with or result in advances in others.

Expert Systems An expert system consists of hardware and software that stores knowledge and makes inferences, enabling a novice to perform at the level of an expert. Like human experts, computerized expert systems use heuristics, or rules of thumb, to arrive at conclusions or make suggestions. Since expert systems can be difficult, expensive, and time consuming to develop, they should be developed when there is a high potential payoff or when they have the poten- tial to significantly reduce downside risk and the organization wants to cap- ture and preserve irreplaceable human expertise.

perceptive system: A system that approximates the way a person sees, hears, and feels objects.

expert system: A system that con- sists of hardware and software that stores knowledge and makes infer- ences, enabling a novice to perform at the level of an expert.

424 PART 3 • Business Information Systems

Components of Expert Systems An expert system is made up of a collection of integrated and related compo- nents, including a knowledge base, an inference engine, an explanation facil- ity, a knowledge base acquisition facility, and a user interface. A diagram of a typical expert system is shown in Figure 10.8.

As shown in the figure, the user interacts with the user interface, which interacts with the inference engine. The inference engine interacts with the

FIGURE 10.6 Brain-machine interface Honda Motors has developed a brain-machine interface that mea- sures electrical current and blood flow change in the brain and uses the data to control ASIMO, the Honda robot. YO

SH IK A ZU

TS U N O /A FP /G et ty Im ag es

Natural language processing

Neural networks

Vision systems Learningsystems

Robotics

Artificial intelligence

Expert systems

FIGURE 10.7 Conceptual model of artificial intelligence AI is a broad field that includes several specialty areas.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 425

other expert system components to provide expertise. This figure also shows the inference engine coordinating the flow of knowledge to other components of the expert system.

Knowledge Base The knowledge base stores all relevant information, data, rules, cases, and relationships that the expert system uses. As shown in Figure 10.9, a knowledge base is a natural extension of a database and an information and decision support system. A knowledge base must be devel- oped for each unique expert system supplication. Rules and cases are fre- quently used to create a knowledge base.

A rule is a conditional statement that links conditions to actions or out- comes. In many instances, these rules are stored as IF-THEN statements, which are rules that suggest certain conclusions. The FICO Blaze Advisor sys- tem is a rules-based platform that allows business users to develop and test rule-based decision applications used by clients for benefits eligibility determi- nation, insurance underwriting, regulatory compliance monitoring, and per- sonal and commercial lending—among other uses.29

A case-based system can also be used to develop a solution to a current problem or situation. In such a system, each case typically contains a descrip- tion of the problem, plus a solution and/or the outcome. The case-based solu- tion process involves (1) finding cases stored in the knowledge base that are similar to the problem or situation at hand, (2) reusing the case in an attempt to solve the problem at hand, (3) revising the proposed solution if necessary, and (4) retaining the new solution as part of a new case. A washing machine repairman who fixes a washer recalling another washer that presented similar

FIGURE 10.8 Components of an expert system An expert system includes a knowl- edge base, an inference engine, an explanation facility, a knowledge base acquisition facility, and a user interface.

Knowledge base

acquisition facility

Inference engine

Explanation facility

Experts User

User interface

FIGURE 10.9 Relationships between data, information, and knowledge A knowledge base stores all relevant information, data, rules, cases, and relationships that an expert system uses.

Database raw facts

Information and decision support information

Knowledge base patterns and relationships

Increasing understanding

rule: A conditional statement that links conditions to actions or outcomes.

IF-THEN statement: A rule that suggests certain conclusions.

426 PART 3 • Business Information Systems

symptoms is using case-based reasoning, so is the lawyer who advocates a particular outcome in a trial based on legal precedents.

Inference Engine The main purpose of an inference engine is to seek infor- mation and relationships from the knowledge base and to provide answers, predictions, and suggestions similar to the way a human expert would. In other words, the inference engine is the component that delivers the expert advice. Consider the expert system that forecasts future sales for a product. One approach is to start with a fact such as “The demand for the product last month was 20,000 units.” The expert system searches for rules that con- tain a reference to product demand. For example, “IF product demand is over 15,000 units, THEN check the demand for competing products.” As a result of this process, the expert system might use information on the demand for competitive products. Next, after searching additional rules, the expert system might use information on personal income or national infla- tion rates. This process continues until the expert system can reach a conclu- sion using the data supplied by the user and the rules that apply in the knowledge base.

Explanation Facility An important part of an expert system is the explanation facility, which allows a user or decision maker to understand how the expert system arrived at certain conclusions or results. A medical expert system, for example, might reach the conclusion that a patient has a defective heart valve given certain symptoms and the results of tests on the patient. The explana- tion facility allows a doctor to find out the logic or rationale of the diagnosis made by the expert system. The expert system, using the explanation facility, can indicate all the facts and rules that were used in reaching the conclusion, which the doctors can look at to determine whether the expert system is pro- cessing the data and information correctly and logically.

Knowledge Acquisition Facility A challenging aspect of developing a useful expert system is the creation and updating of the knowledge base. In the past, when more traditional programming languages were used, developing a knowledge base was tedious and time consuming. Each fact, relationship, and rule had to be programmed—usually by an experienced programmer.

Today, specialized software allows users and decision makers to create and modify their own knowledge bases through the knowledge acquisition facility, using user-friendly menus. The purpose of the knowledge acquisition facility is to provide a convenient and efficient means of capturing and storing all components of the knowledge base. The knowledge acquisition facility acts as an interface between experts and the knowledge base.

User Interface The main purpose of the user interface is to make an expert system easier for users and decision makers to develop and use. At one time, skilled computer personnel created and operated most expert systems; today, simplified user interfaces permit decision makers to develop and use their own expert systems.

Participants in Developing and Using Expert Systems Typically, several people are involved in developing and using an expert sys- tem. The domain expert is the person or group with the expertise or knowl- edge the expert system is trying to capture (domain). In most cases, the domain expert is a group of human experts. A knowledge engineer is a per- son who has training or experience in the design, development, implementa- tion, and maintenance of an expert system, including training or experience with expert system shells. Knowledge engineers can help transfer the knowl- edge from the expert system to the knowledge user. The knowledge user

inference engine: Part of the expert system that seeks information and relationships from the knowledge base and provides answers, predictions, and suggestions similar to the way a human expert would.

explanation facility: Component of an expert system that allows a user or decision maker to understand how the expert system arrived at certain conclusions or results.

knowledge acquisition facility: Part of the expert system that provides a convenient and efficient means of capturing and storing all the compo- nents of the knowledge base.

knowledge user: The person or group who uses and benefits from the expert system.

knowledge engineer: A person who has training or experience in the design, development, implementation, and maintenance of an expert system.

knowledge user: The person or group who uses and benefits from the expert system.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 427

is the person or group who uses and benefits from the expert system. Knowledge users do not need any previous training in computers or expert systems.

Expert System Shells and Products An expert system shell is a suite of software that allows construction of a knowledge base and interaction with this knowledge base through the use of an inference engine. Expert system shells are available for both personal com- puters and mainframe systems, with some shells being inexpensive, costing less than $500. In addition, off-the-shelf expert system shells are complete and ready to run. The user enters the appropriate data or parameters, and the expert system provides output to the problem or situation.

Robotics Robotics is a branch of engineering that involves development and manufac- ture of mechanical or computer devices that can perform tasks that require a high degree of precision or are tedious or hazardous for human beings, such as painting cars or making make precision welds. Karel Capek introduced the word “robot” in his 1921 play, R.U.R. (an abbreviation of Rostrum’s Universal Robots). The play was about an island factory that produced artificial people called “robots” who are consigned to do drudgery work and eventually rebel and overthrow their creators, causing the extinction of human beings.30 Orga- nizations today do indeed use robots to perform dull, dirty, and/or dangerous jobs. They are often used to lift and move heavy pallets in warehouses, per- form welding operations, and provide a way to view radioactively contami- nated areas of power plants inaccessible by people.

However, the use of robots has expanded and is likely to continue to grow. Robots are increasingly being used in surgical procedures ranging from prostrate removal to open-heart surgery. See Figure 10.10. Robots can provide doctors with enhanced precision, improved dexterity, and better visualization. In 2014, the U.S. Navy’s Bluefin 21 robotic submarine made several trips below the Indian Ocean’s surface to scan the seabed for any trace of the miss- ing Malaysia Airlines Flight 370. iRobot (www.irobot.com) is a company that builds a variety of robots, including the Roomba for vacuuming floors, the

FIGURE 10.10 Robotic surgery The arms of the Da Vinci robot assist in a kidney transplant. A surgeon controls the robot remotely from a corner of the operating room. Ma

st er

Vi de o/ Sh ut te rs to ck .c om

robotics: A branch of engineering that involves the development and manufacture of mechanical or com- puter devices that can perform tasks requiring a high degree of precision or that are tedious or hazardous for humans.

428 PART 3 • Business Information Systems

Looj for cleaning gutters, and the PackBot, an unmanned vehicle used to assist and protect soldiers.31

Some robots, such as the ER series by Intelitek, can be used for training or entertainment.32 Wonder Workshop created robots Dash and Dot to help young children (ages 5+) to learn programming concepts and creative prob- lem solving.33

Some people fear that robots will increasingly take jobs from human employees. For example, the use of autonomous vehicles may place millions of truck drivers, chauffeurs, and cab drivers out of work. Some experts have predicted that every home will have a robot of some sort by 2025. One area of particular interest is the use of robots as companions and caregivers for people who are sick, elderly, or physically challenged.34,35

Vision Systems Another area of AI involves vision systems, which include hardware and software that permit computers to capture, store, and process visual images. 3D machine-vision systems are used to increase the accuracy and speed of industrial inspections of parts. Automated fruit-picking machines use a unique vacuum gripper combined with a vision system to pick fruit. Facebook is developing an AI vision system called DeepFace, which creates 3D models of the faces in photos. The technology represents a vast improvement over cur- rent facial recognition software. DeepFace can correctly tell if two photos show the same person with 97.25 percent accuracy. This nearly matches humans, who are correct 97.53 percent of the time. The technology, which uses more than 120 million parameters, can also recognize people whose faces are not showing with 83 percent accuracy, using features such as body shape, posture, hairstyle, and clothing. 36,37

Natural Language Processing Natural language processing is an aspect of artificial intelligence that involves technology that allows computers to understand, analyze, manipu- late, and/or generate “natural” languages, such as English. Many companies provide natural language processing help over the phone. When you call a help phone number, you are typically given a menu of options and asked to speak your responses. Many people, however, become easily frustrated talk- ing to a machine instead of a human. The Naturally Speaking application from Dragon Systems uses continuous voice recognition, or natural speech, that allows the user to speak to the computer at a normal pace without paus- ing between words. The spoken words are transcribed immediately onto the computer screen. See Figure 10.11.

FIGURE 10.11 Voice recognition software With the NaturallySpeaking applica- tion from Dragon Systems, computer users can speak and have their words transcribed into text for input to software such as Microsoft Word. Source: Nuance Communications.

vision system: The hardware and software that permit computers to cap- ture, store, and manipulate visual images.

natural language processing: An aspect of artificial intelligence that involves technology that allows com- puters to understand, analyze, manip- ulate, and/or generate “natural” languages, such as English.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 429

After converting sounds into words, natural language-processing systems react to the words or commands by performing a variety of tasks. Brokerage services are a perfect fit for voice recognition and natural language-processing technology to replace the existing “press 1 to buy or sell a stock” touchpad tele- phone menu system. Using voice recognition to convert recordings into text is also possible. Some companies claim that voice recognition and natural language-processing software is so good that customers forget they are talking to a computer and start discussing the weather or sports scores.

Learning Systems Another aspect of AI deals with learning systems, a combination of software and hardware that allows a computer to change how it functions or how it reacts to situations based on feedback it receives. For example, some comput- erized games have learning abilities. If the computer does not win a game, it remembers not to make the same moves under the same conditions again. Some learning systems utilize reinforcement learning, which involves the use of sequential decisions—with learning taking place between each decision. Reinforcement learning often involves sophisticated computer programming and optimization techniques. The computer makes a decision, analyzes the results, and then makes a better decision based on the analysis. The process, often called dynamic programming, is repeated until it is impossible to make improvements in the decision.

Learning systems software requires feedback on the results of actions or decisions. At a minimum, the feedback needs to indicate whether the results are desirable (winning a game) or undesirable (losing a game). The feedback is then used to alter what the system will do in the future.

After Google combined natural language processing with learning systems in its Android smartphone operating system, it reduced word-recognition errors by 25 percent.38 With this new technology, the voice assistant is also able to ask questions to clarify what a user is searching for.

Neural Networks An increasingly important aspect of AI involves neural networks, also called neural nets. A neural network is a computer system that can recognize and act on patterns or trends that it detects in large sets of data. A neural network employs massively parallel processors in an architecture that is based on the human brain’s own meshlike structure. As a result, neural networks can pro- cess many pieces of data at the same time and learn to recognize patterns.

AI Trilogy, available from the Ward Systems Group (www.wardsystems .com), is a neural network software program that can run on a standard PC. The software package’s NeuroShell Predictor component can be used to make predictions, such as agricultural production estimates and call volume forecasts. The NeuroShell Classifier can be used to aid in classification and decision-making tasks, such as alarm system malfunction diagnosis and sales prospect selection. See Figure 10.12. The software package also contains Gen- eHunter, which uses a special type of algorithm called a genetic algorithm to get the best result from the neural network system. (Genetic algorithms are discussed next.) Some pattern recognition software uses neural networks to make credit lending decisions by predicting the likelihood a new borrower will pay back a loan. Neural networks are also used to identify bank or credit card transactions likely to be fraudulent. Large call centers use neural net- works to create staffing strategies by predicting call volumes.

Dr. José R. Iglesias-Rozas at the Katharinenhospital in Stuttgart, Germany, is a leader in researching the use of neural networks to diagnose the degree of malignancy of tumors. In his early research, microscopic sections of 786 different human brain tumors were collected. A neural network tool called

learning system: A combination of software and hardware that allows a computer to change how it functions or how it reacts to situations based on feedback it receives.

neural network: A computer system that can recognize and act on patterns or trends that it detects in large sets of data.

430 PART 3 • Business Information Systems

NeuralTools was then used to predict the degree of malignancy based on the presence of 10 histological characteristics. The neural network accurately pre- dicted over 95 percent of the sample cases. Dr. Iglesias-Rozas plans to expand his research to analyze over 30 years of data from more than 8,000 patients with brain tumors.39

Other Artificial Intelligence Applications Other artificial intelligence applications include genetic algorithms, which solve problems based on the theory of evolution—using the concept of survival of the fittest as a problem-solving strategy. The genetic algorithm uses a fitness function that quantitatively evaluates a set of initial candidate solutions. The highest- scoring candidate solutions are allowed to “reproduce,” with random changes introduced to create new candidate solutions. These digital offspring are sub- jected to a second round of fitness evaluation. Again, the most promising candi- date solutions are selected and used to create a new generation with random changes. The process repeats for hundreds or even thousands of rounds. The expectation is that the average fitness of the population will increase each round and that eventually very good solutions to the problem will be discovered.

Genetic algorithms have been used to solve large, complex scheduling problems, such as scheduling airline crews to meet flight requirements while minimizing total costs and staying within federal guidelines on maximum crew flight hours and required hours of rest. Genetic algorithms have also

FIGURE 10.12 Neural network software NeuroShell Predictor uses recognized forecasting methods to look for future trends in data. Source: Ward Systems Group, Inc.

genetic algorithm: An approach to solving problems based on the theory of evolution; uses the concept of sur- vival of the fittest as a problem-solving strategy.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 431

Critical Thinking Exercise

been used to design mirrors that funnel sunlight to a solar collection and radio antenna that pick up signals from space.

Another artificial intelligence application, intelligent agent (also called an intelligent robot or bot), consists of programs and a knowledge base used to perform a specific task for a person, a process, or another program. Like a sports agent who searches for the best endorsement deals for a top athlete, an intelligent agent is often used to search for the best price, schedule, or solution to a problem. The programs used by an intelligent agent can search large amounts of data as the knowledge base refines the search or accommo- dates user preferences. Often used to search the vast resources of the Internet, intelligent agents can help people find information on any topic, such as the best price for a new camera or used car.

IBM Watson Offers Advice to Cancer Patients IBM and the American Cancer Society (ACS) are working in partnership to develop a Watson-based advisor to support cancer patients. They will create this robust resource by drawing upon massive sources of data from both organiza- tions, and then train Watson to use the data to understand and anticipate indivi- duals’ needs. In addition, the advisor will learn about the patient and that patient’s planned treatment regimen, allowing it to offer personalized advice that matches the patient’s individual characteristics, preferences, and treatment plan.40

The goal is for the Watson-based advisor to anticipate the needs of people with different types of cancers, at different stages of disease, and at various points in treatment. For example, a person with lung cancer experiencing unusual levels of pain could ask what might be causing pain. The advisor would be designed to respond with information on symptoms and self-management options associated with that persons’ current and future phases of treatment, based on the experi- ences of people with similar characteristics.

Review Questions 1. How might visual systems and natural language processing be incorporated

into the Watson cancer adviser system? 2. What other major branches of artificial intelligence are employed in the

Watson cancer patient adviser?

Critical Thinking Questions 1. One of the challenges of a cognitive computing capability such as the Watson

cancer adviser is keeping the information that Watson draws on as current as possible. Over time, new approaches, courses of treatment, medicines, and ideas will be discovered that are improvement over the old way of doing things. How might the Watson cancer adviser be kept as current as possible?

2. Cancer patients frequently suffer from depression. Do you think it is possible for Watson to recognize symptoms of depression and provide encouragement and advice to the patient? How might this be accomplished?

Multimedia and Virtual Reality

The use of multimedia and virtual reality has helped many companies achieve a competitive advantage and increase profits. The approach and technology used in multimedia is often the foundation of virtual reality systems, discussed later in this section. While these specialized information systems are not used by all organizations, they can play a key role for many. We begin with a dis- cussion of multimedia.

intelligent agent: Programs and a knowledge base used to perform a specific task for a person, a process, or another program; also called an intelli- gent robot or bot.

432 PART 3 • Business Information Systems

Overview of Multimedia Multimedia is content that uses more than one form of communication—such as text, graphics, video, animation, audio, and other media. Multimedia tools can be used to help an organization achieve its goals through the production of compelling brochures, presentations, reports, and documents. Many compa- nies use multimedia approaches to develop animations and video games to help advertise products and services. For example, insurance company Geico uses an animated gecko in some of its TV ads. Animation software, such as Blender, GoAnimate, and Powtoon, offer tools for individuals and businesses looking to develop these types of animations. Although not all organizations use the full capabilities of multimedia, most use text and graphics capabilities.

Text and Graphics Most organizations use text and graphics to develop reports, financial state- ments, advertising pieces, and other documents for internal and external use. Internally, organizations use text and graphics to communicate policies, guidelines, procedures, and much more to employees. Externally, they use text and graphics to communicate to suppliers, customers, federal and state governmental agencies, and a variety of other stakeholders. Different sizes, fonts, and colors can be used to create different effects with text. Graphics such as photographs, illustrations, drawings, a variety of charts, and other still images can be used to create interest and illustrate a message. Some pop- ular digital image formats include EPS (Encapsulated PostScript), GIF (Graphic Interchange Format), JPEG (Joint Photographic Experts Group for- mat), PNG (Portable Network Graphics), TIFF (Tagged Image File Format), and Raw image files.

While standard word-processing programs are an inexpensive and simple way to develop documents and reports that require text and graphics, most organizations use specialized software. Adobe Illustrator, for example, can be used to create typography and vector graphics, such as illustrations and logos. Other graphics programs include CorelDraw by Corel Corporation and Serif DrawPlus. Software products such as Adobe InDesign can be used for page design, layout, and publishing of digital and print manuals, brochures, and reports. Adobe Photoshop is a sophisticated and popular software package that can be used to edit photographs and other visual images. Once created, these documents and reports can be saved in an Adobe PDF file, which can be used for printing or for digital delivery. Other photo-editing software includes Apple’s iPhoto, Corel PaintShop Pro, and Pixelmator—among many others.

Microsoft PowerPoint can be used to develop presentations with sound and animation that can be displayed on a large viewing screen. Prezi and Swipe are both Web-based presentation software alternatives to PowerPoint.

Many graphics programs can also create 3D images. Once used primarily in movies, 3D technology can be employed by companies to design products, such as motorcycles, jet engines, and bridges. Autodesk, for example, makes exciting 3D software that companies can use to design everything from flat fruit-packing machines for Sunkist to large skyscrapers and other buildings for architectural firms.41 The technology used to produce 3D movies is also available with some TV programs. Nintendo developed the Nintendo 3DS, one of the first portable gaming devices that displays images in 3D.

Audio Audio, which includes music, human voices, recorded sounds, and a variety of computer-generated sounds, can be stored in a variety of file formats, including AAC (Advanced Audio Coding), AIFF (Audio Interchange File For- mat), ALAC (Apple Lossless Audio Codec), FLAC (Free Lossless Audio Codec), MIDI (Musical Instrument Digital Interface), MP3 (Motion Picture Experts

multimedia: Content that uses more than one form of communication—such as text, graphics, video, animation, audio, and other media.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 433

Group Audio Layer 3), and WAV (wave format). The term “streaming audio” refers to audio files that are played while they are being downloaded from the Internet.

Input to audio software includes audio recording devices, microphones, imported music or sound from CDs or audio files, MIDI instruments that can create music and sounds directly, and other audio sources. Once stored, audio files can be edited and augmented using audio software, such as Apple Quick- Time, Microsoft Sound Recorder, Adobe Audition, and SourceForge Audacity. See Figure 10.13. Once edited, audio files can also be used to enhance presen- tations, create music, broadcast satellite radio signals, develop audio books, record podcasts, add realism to movies, and enrich video and animation.

Video and Animation The moving images of video and animation are typically created by rapidly displaying one still image after another. Video and animation can be stored in a variety of file formats, including AVI (Audio Video Interleave), FLV/FV4 (Flash Video), MOV (QuickTime format) files, MPEG (Motion Picture Experts Group format), QTFF (QuickTime File Format), and WMV (Windows Media Video). When video files are played while they are being downloaded from the Internet, it’s called streaming video. For example, Amazon Prime, Hulu, and Netflix are just three of the many sites through which users can stream movies and TV programs. On the Internet, Java applets (small downloadable programs) and animated GIF files can be used to animate or create “moving” images.

A number of video and animation software products can be used to create and edit video and animation files. Many video and animation programs can create realistic 3D moving images. James Cameron’s movie Avatar used sophisticated computers and 3D imaging to create one of the most profitable movies in history. Adobe’s Premiere and After Effects and Apple’s Final Cut Pro can be used to edit video images taken from cameras and other sources. Final Cut Pro, for example, has been used to edit and produce full-length motion pictures shown in movie theaters. Adobe Flash and LiveMotion can be used to add motion and animation to Web pages.

Video and animation have many business uses. Companies that develop computer-based or Internet training materials often use video and audio soft- ware. An information kiosk at an airport or shopping mall can use animation to help customers check in for a flight or get information.

FIGURE 10.13 Audio-editing software Audacity provides tools for editing and producing audio files in a variety of formats. Source: Audacity.

434 PART 3 • Business Information Systems

Visual effects involve the integration of live-action footage and generated imagery to create settings that look completely lifelike, but would be danger- ous or extremely expensive to capture on film. RenderMan is Pixar’s technical specification for a standard communications interface between 3D computer graphics programs and rendering programs. (Rendering is the final step in the animation process and provides the final appearance to the animation with visual effects such as shading, texture mapping, shadows, reflections, and motion blurs.) RenderMan software is used widely to create outstanding graphics for feature films and broadcast television. Indeed, it was used on every Visual Effects Academy Award Winner for 15 straight years.42

File Conversion and Compression Most multimedia applications are created, edited, and distributed in a digital file format, such as the ones discussed earlier. Older inputs to these applications, however, can be in an analog format—from old home movies, magnetic tapes, vinyl records, or similar sources. In addition, some older digital formats are no longer popular or used. In order to edit and process analog and older digital formats using current multimedia software, the content must be converted into a newer digital format. Files can be converted using software or specialized hardware. Some of the multimedia software, such as Adobe Premiere, Adobe Audition, and others, have this analog-to-digital conversion capability. Stand- alone software and specialized hardware can also be used. Grass Valley, for example, is a hardware device that can be used to convert analog video to digi- tal video or digital video to analog video. With this device, you can convert old VHS tapes to digital video files or digital video files to an analog format.

Because multimedia files can be large, it’s sometimes necessary to com- press files to make them easier to download from the Internet or send as email attachments. Many of the multimedia software programs discussed earlier can be used to compress multimedia files. In addition, standalone file conversion programs, such as PNGGauntlet, ScriptPNG, WinZip, and Wonder- share Video Converter Ultimate, can be used to compress many file formats.

Designing a Multimedia Application Designing multimedia applications requires careful thought and a systematic approach. Multimedia applications can be printed in brochures, placed into corporate reports, uploaded to the Internet, or displayed on large screens for viewing. Because these applications are typically more expensive than prepar- ing documents and files in a word-processing program, it is important to spend time designing the best possible multimedia application. Designing a multimedia application requires that the end use of the document or file be carefully considered. For example, some text styles and fonts are designed for Internet display. Because different computers and Web browsers display infor- mation differently, it is a good idea to select styles, fonts, and presentations based on computers and browsers that are likely to display the multimedia application. Because large files can take much longer to load into a Web page, smaller files are usually preferred for Web-based multimedia applications.

Overview of Virtual Reality The term “virtual reality” was initially coined in 1989 by Jaron Lanier, founder of VPL Research. Originally, the term referred to immersive virtual reality in which the user becomes fully immersed in an artificial, 3D world that is completely generated by a computer. Through immersion, the user can gain a deeper understanding of the virtual world’s behavior and functionality.

virtual reality system: A system that enables one or more users to move and react in a computer-simulated environment.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 435

simulated world to the user. These devices can also record and send the speech and movements of the participants to the simulation program, enabling users to sense and manipulate virtual objects much as they would real objects. This natural style of interaction gives the participants the feeling that they are immersed in the simulated world. For example, an auto manu- facturer can use virtual reality to simulate and design automobiles and pro- duction lines in factories.

In justifying Facebook’s $2 billion acquisition of virtual reality company Oculus VR, Mark Zuckerberg said that “while mobile is the key platform for today, virtual reality will be one of the major platforms for tomorrow. Imagine enjoying a courtside seat at a game, studying in a classroom of students and teachers all over the world or consulting with a doctor face-to-face, just by putting on goggles in your home.”43 In late 2015, Oculus VR released one of the first social virtual reality applications—an app that allows users of the Samsung Gear VR headset to go to a virtual movie theater (in the form of a selected avatar), watch a movie, and chat with other people who are also in the theater.44

Interface Devices To see in a virtual world, the user often wears a head-mounted display (HMD) with screens directed at each eye. The HMD also contains a position tracker to monitor the location of the user’s head and the direction in which the user is looking. Employing this information, a computer generates images of the vir- tual world—a slightly different view for each eye—to match the direction in which the user is looking and displays these images on the HMD. In addition to Oculus VR (Oculus Rift) and Samsung (Gear VR), other big players in the emerging virtual reality interface industry include Google (Cardboard), HTC (Vive), Microsoft (Hololens), and Sony (PlayStation VR).

The Electronic Visualization Laboratory (EVL) at the University of Illinois at Chicago introduced a room constructed of large screens on three walls and a floor on which the graphics are projected. The CAVE (Cave Automatic Virtual Environment), as the room is called, provides the illusion of immer- sion by projecting stereo images on the walls and floor of a room-sized cube (www.evl.uic.edu). Several people wearing lightweight stereo glasses can enter and walk freely inside the CAVE. A head-tracking system continu- ously adjusts the stereo projection to the current position of the leading viewer. The most recent version of the CAVE is called CAVE2 and features realistic high-resolution graphics that respond to user interactions. See Figure 10.14.

FIGURE 10.14 Large-scale virtual reality environment The CAVE2 virtual reality system has 72 stereoscopic LCD panels encir- cling the viewer 320 degrees and creates a 3D environment that can simulate the bridge of the Starship U.S.S. Enterprise, a flyover of the planet Mars, or a journey through the blood vessels of the brain. AP

Im ag es /C ha rle s Re x A rb og as t

436 PART 3 • Business Information Systems

In a virtual world, users hear sounds through earphones, and informa- tion reported by the position tracker is used to update audio signals. When a sound source in virtual space is not directly in front of or behind the user, the computer transmits sounds to arrive at one ear a little earlier or later than at the other and to be a little louder or softer and slightly different in pitch.

The haptic interface, which relays the sense of touch and other physical sensations in the virtual world, is the least developed and perhaps the most challenging virtual reality component to create. One virtual reality company has developed a haptic interface device that can be placed on a person’s fin- gertips to give an accurate feel for game players, surgeons, and others. With the use of a glove and position tracker, the computer locates the user’s hand and measures finger movements. The user can reach into the virtual world and handle objects; still, it is difficult to generate the sensations of a person tapping a hard surface, picking up an object, or running a finger across a tex- tured surface. Touch sensations also have to be synchronized with the sights and sounds users experience. Today, some virtual reality developers are even trying to incorporate taste and smell into virtual reality applications.

Forms of Virtual Reality Aside from immersive virtual reality, virtual reality can also refer to applica- tions that are not fully immersive, such as mouse-controlled navigation through a 3D environment on a graphics monitor, stereo viewing from the monitor via stereo glasses, and stereo projection systems. Augmented reality, a newer form of virtual reality, has the ability to superimpose digital data over real photos or images. Augmented reality is being used in a variety of settings. Some luxury car manufacturers, for example, display dashboard information, such as speed and remaining fuel, on windshields. The technol- ogy is also used in some military aircraft and is often called heads-up display. The use of lines (typically yellow and blue) that are superimposed onto a football field during broadcasted games to indicate the first down marker and the line of scrimmage is another example of augmented reality. GPS maps can be combined with real pictures of stores and streets to help you locate your position or find your way to a new destination. Using augmented reality, you could point a smartphone camera at a historic landmark, such as a castle, museum, or other building, and have information about the landmark appear on your screen, including a brief description of the landmark, admission price, and hours of operation. Although still in its early phases of implementa- tion, augmented reality has the potential to become an important feature of tomorrow’s smartphones and similar mobile devices.

Virtual Reality Applications Thousands of applications of virtual reality are available, with more being developed as the cost of hardware and software declines and as people’s ima- ginations are opened to the potential of virtual reality. Virtual reality applica- tions are being used in medicine, education and training, business, and entertainment, among other fields.

Medicine Virtual reality has been successful in treating children with autism by helping them pick up on social cues, refine their motor skills, and acquire real-life skills, such as looking both ways before crossing the street. Some children with autism interact well with technology because of its predictability, control- lability, and incredible patience. Virtual reality has also been used to help train medical students with simulations for many forms of surgery from brain surgery to delivery of a baby.45

CHAPTER 10 • Knowledge Management and Specialized Information Systems 437

Education and Training Virtual environments are used in education to bring new resources into the classroom. Thousands of administrators, faculty, researchers, staff, and stu- dents are members of the Immersive Education Initiative (IEI), a nonprofit international partnership of colleges, companies, research institutes, and universities working together to define and develop open standards, best practices, platforms, and communities of support for virtual reality and game-based learning and training systems. The IEI sponsors immersive edu- cation summits as well as clubs and camps for children, college students, and corporate professionals.

Google’s Expeditions Pioneer Program is a pilot immersive education ini- tiative that brings virtual reality kits into the classroom so that students can take part in expeditions that help them learn about locations around world. Using cheap cardboard headsets, Android phones, and a teacher-operated tab- let, Google Expeditions lets students experience 360-degree views of outer space, the White House, caves in Slovakia, Buckingham Palace, Antarctica, the Amazon rainforest, the Great Barrier Reef, and 100 other locations.46,47

Virtual technology is also being used to train members of the military. To help with aircraft maintenance, a virtual reality system has been developed to simulate an aircraft and give a user a sense of touch, while computer graphics provide a sense of sight and sound. The user sees, touches, and manipulates the various parts of the virtual aircraft during training. Also, the Pentagon is using a virtual reality training lab to prepare for a military crisis. The virtual reality system simulates various war scenarios.

Business and Commerce Virtual reality is being used in business for many purposes—to provide virtual tours of plants and buildings, enable 360-degree viewing of a product or machine, and train employees. For example, Ford uses virtual reality technol- ogy to refine its auto designs. Designers and engineers are able to scrutinize the interior and exterior of a car design. Because the virtual reality technology is tied directly into Ford’s Autodesk computer aided design (CAD) system, workers can even inspect a particular component to see exactly how it is designed.48

Several teams within the National Football League (NFL) have begun using virtual reality applications to train players. The teams are able to load all of their offensive and defensive plays into the system, which uses a 360-degree high-definition capture of each position on the football field. Players training with the system are able to study the plays from a first- person perspective, rewinding plays as necessary. While the player is training, coaches can see what the player is seeing and doing, providing ongoing feed- back. The VR systems are an enhancement to the previous off-field training programs that focused on players learning plays by reading large playbooks and watching DVDs of games.49,50

Swedish furniture chain IKEA recently launched a virtual reality app that allows users to experiment with different kitchen design features, such as cab- inet configurations, countertop materials, and drawer pulls. The app is intended to help consumers make design choices and visualize space in three dimensions by allowing them to quickly make changes and view their cus- tomized kitchens from different perspectives. The company is using this pilot program to explore the possibilities of virtual reality technology in terms of connecting with and empowering its customers.51

Microsoft is developing its Oculus virtual reality headset for enterprise workers to interact with office productivity software.52 With this headset and associated software, users can see Microsoft Excel PivotTables that actually pivot in 3D, send weekly status reports that appear in a fully immersive

438 PART 3 • Business Information Systems

Critical Thinking Exercise

environment, and generate PowerPoint presentations that support full posi- tional head tracking and provide a 3D stereo sound sensation for the listener. Virtual reality headsets such as those by Oculus Rift can be used to walk through the simulated environment.

Launching Google Expeditions You have been asked to help a local elementary school explore the feasibility of implementing the Google Expeditions Pioneer Program. This program enables students to experience over 100 locations around the world using virtual reality.

Review Questions 1. Do research to find out how the school can sign up for this program and what

hardware and software is required. 2. What training is necessary for teachers to lead this program?

Critical Thinking Questions 1. Prepare a recommendation for the school outlining the steps and resources

necessary to take advantage of this program. 2. What potential barriers to implementing this program could arise? How might

these be overcome?

Other Specialized Systems

In addition to artificial intelligence, expert systems, and virtual reality, other interesting specialized systems continue to be developed, including assistive technology systems, game theory, and informatics.

Assistive Technology Systems Assistive technology systems includes a wide range of assistive, adaptive, and rehabilitative devices to help people with disabilities perform tasks that they were formerly unable to accomplish or had great difficulty accomplishing.

Many assistive technology products are designed to enhance the human- computer interface. Electronic pointing devices are available that enable users to control the pointer on the screen without the use of hands, using ultrasound, infrared beams, eye movements, and even nerve signals and brain waves. Sip- and-puff systems are activated by inhaling or exhaling. Braille embossers can translate text into embossed Braille output. Screen readers can be used to speak everything displayed on the computer screen, including text, graphics, control buttons, and menus. Speech recognition software enables users to give commands and enter data using their voices rather than a mouse or keyboard. Text-to-speech synthesizers can “speak” all data entered to the computer to allow users who are visually impaired or who have learning difficulties to hear what they are typing.53 Stephen Hawking is an English theoretical physicist and cosmologist considered by many to be the most intelligent man alive today. Hawking is almost entirely paralyzed and uses assistive technology systems to communicate his thoughts and to interact with computers. See Figure 10.15.

Personal assistive listening devices help people understand speech in dif- ficult situations. They separate the speech that a person wants to hear from background noise by improving what is known as the “speech to noise ratio.” A personal assistive learning device typically has at least three compo- nents: a microphone, a transmission technology, and a device for receiving the signal and bringing the sound to the ear.54

assistive technology system: An assistive, adaptive, or rehabilitative device designed to help people with disabilities perform tasks that they were formerly unable to accomplish or had great difficulty accomplishing.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 439

Personal emergency response systems use electronic sensors connected to an alarm system to help maintain security, independence, and peace of mind for anyone who is living alone, at risk for falls, or recuperating from an illness or surgery. These systems include fall detectors, heart monitors, and unlit gas sensors. When an alert is triggered, a message is sent to a caregiver or contact center who can respond appropriately.

Game Theory Game theory is a mathematical theory for developing strategies that maximize gains and minimize losses while adhering to a given set of rules and constraints. Game theory is frequently applied to solve various decision-making problems in which two or more participants are faced with choices of action, by which each may gain or lose, depending on what others choose to do or not to do. Thus, the final outcome of a game is determined jointly by the strategies chosen by all participants. Such decisions involve a degree of uncertainty because no partici- pant knows for sure what course of action the other participants will take. In zero-sum games, the fortunes of the players are inversely related so that one par- ticipant’s gain is the other participant’s loss. In non-zero-sum games, it is wise for the participants to cooperate so that the action taken by one participant may ben- efit both participants. Two-person zero-sum games are used by military strate- gists. Many-person non-zero-sum games are used in many business decision- making settings. Game theory Explorer and Gambit are collections of software tools for building, analyzing, and exploring game models.55

In the TV game show Jeopardy!, contestants typically select a single cate- gory and progressively move down from the top question (easiest and lowest dollar value) to the bottom (hardest and highest dollar value). This provides the contestants and viewers with an easy-to-understand escalation of diffi- culty. Recently, however, one player used a much different strategy, employ- ing the fundamentals of game theory. The player sought out the Daily Double questions, which are usually hidden in one of the three highest-paying and most difficult questions in the categories. Thus rather than selecting a single category and increasing the degree of difficulty, he began with the two most

FIGURE 10.15 Stephen Hawking Stephen Hawking employs a number of assistive technology systems to support his activities. Th

e W or ld in H D R/ Sh ut te rs to ck .c om

game theory: A mathematical theory for developing strategies that maximize gains and minimize losses while adhering to a given set of rules and constraints.

440 PART 3 • Business Information Systems

difficult questions in the category. Once the two most difficult questions were taken off the board in one category, he skipped to another category in search of the Daily Doubles. This strategy proved highly successful.56

The U.S. Coast Guard employs a game theory system called PROTECT (Port Resiliency for Operational/Tactical Enforcement to Combat Terrorism) to randomize patrols while still achieving a very high level of security that provides maximum deterrence. There are insufficient resources to provide full security coverage around the clock at all high-value potential targets in the 361 shipping ports in the United States. This means that enemies can observe patrol and monitor activities and take actions in an attempt to avoid patrols. PROTECT generates patrol and monitoring schedules that take into account the importance of different targets at each port and the enemy’s likely surveil- lance and anticipated reaction to those patrols.57

Informatics Informatics is the combination of information technology with traditional dis- ciplines, such as medicine or science, while considering the impact on indivi- duals, organizations, and society. Informatics places a strong emphasis on the interaction between humans and technology—with the goal of engineering information systems that provide users with the best possible user experience. Indeed, informatics represents the intersection of people, information, and technology. See Figure 10.16. The field of informatics has great breadth and encompasses many individual specializations such as biomedical, health, nurs- ing, medical, and pharmacy informatics. Those who study informatics learn how to build new computing tools and applications. They gain an understand- ing of how people interact with information technology and how information technology shapes our relationships, our organizations, and our world.

Biomedical informatics (or bioinformatics) develops, studies, and applies theories, methods, and processes for the generation, storage, retrieval, use, and sharing of biomedical data, information, and knowledge. Bioinformatics has been used to help map the human genome and conduct research on bio- logical organisms. Using sophisticated databases and artificial intelligence, bioinformatics helps unlock the secrets of the human genome, with the objec- tive of preventing diseases and saving lives. Many universities have courses on bioinformatics and offer bioinformatics certification.

Healthcare informatics is the science of how to use data, information, and technology to improve human health and the delivery of healthcare services. Healthcare informatics applies principles of computer and information science to the advancement of patient care, life sciences research, health professional education, and public health. Journals, such as Healthcare Informatics, report current research on applying computer systems and technology to increase efficiency, reduce medical errors, and improve health care.

Boston Medical Center (BMC) has implemented a healthcare informatics program to develop and measure several key operational metrics to improve

FIGURE 10.16 Informatics Informatics represents the intersec- tion of people, information, and technology.

People

Technology

Informatics

Information

informatics: The combination of information technology with traditional disciplines, such as medicine or sci- ence, while considering the impact on individuals, organizations, and society.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 441

Critical Thinking Exercise

efficiency in its operating room procedures. The system is being used to optimize surgical schedules, wait times, postsurgical unit availability, and patient-discharge processes. By optimizing the use of its operating rooms, the hospital is able to ensure that surgeons are fully utilizing their assigned operating room time, and—even more important for patient care—reduce the time it takes for a patient to have surgery after diagnosis.58

Allina Health Employs Healthcare Informatics Allina Health is a not-for-profit healthcare system that owns or operates 14 hospitals and over 90 clinics throughout Minnesota and western Wisconsin. The organization created an enterprise-wide data warehouse of clinical, financial, operational, patient satisfaction, and other data. The data is used to calculate and track various healthcare measures that measure performance and can identify opportunities for improvement.

Review Questions 1. What sort of training and experience is needed by the individuals who built

and operate this system? What about by the individuals who use this system? 2. Identify several sources from which the data for this system may be obtained.

Critical Thinking Questions 1. Provide several examples of key patient healthcare measures that are tracked

by this system. 2. Provide an example or two of how this system might be used to identify an

opportunity for improvement in the treatment of patients.

Summary

Principle: Knowledge management allows organizations to share knowledge and experience among its workers.

Knowledge management is a range of practices concerned with increasing awareness, fostering learning, speeding collaboration and innovation, and exchanging insights. A knowledge management system is an organized collec- tion of people, procedures, software, databases, and devices used to create, capture, store, share, and use the organization’s knowledge and experience.

Explicit knowledge is objective and can be measured and documented in reports, papers, and rules. Tacit knowledge is hard to measure and document and is typically not objective or formalized.

A major goal of knowledge management is to somehow capture and docu- ment the valuable work-related tacit knowledge of others and to turn it into explicit knowledge that can be shared with others.

Two processes are frequently used to capture tacit knowledge—shadowing and joint problem solving. Shadowing involves a novice observing an expert exe- cuting her job to learn how she performs. This technique often is used in the medi- cal field to help young interns learn from experienced physicians. With joint problem solving, the novice and the expert work side by side to solve a problem so that the expert’s approach is slowly revealed to the observant novice.

Organizations employ KM to foster innovation, leverage the expertise of people across the organization, and capture the expertise of key individuals before they retire.

Some organizations and professions use communities of practice (CoP), which are groups of people with common interests who come together to cre- ate, store, and share knowledge on a specific topic.

442 PART 3 • Business Information Systems

Obtaining, storing, sharing, and using knowledge is the key to any knowl- edge management system, which often leads to additional knowledge creation, storage, sharing, and usage. Many tools and techniques can be used to create, store, and use knowledge. These tools and techniques are available from IBM, Microsoft, and other companies and organizations.

Establishing a successful KM program is challenging, but most of the chal- lenges involved have nothing to do with the technologies or vendors employed. Instead they are challenges associated with human nature and the manner in which people are accustomed to working together. Best practices for selling and implementing a KM project include connecting KM to goals and objectives, starting with a small pilot and enthusiastic participants, identifying tacit knowledge, and getting employee buy in.

Organizations interested in piloting KM should be aware of the wide range of technologies that can support KM efforts. These include communities of practice, organizational network analysis, a variety of Web 2.0 technologies, business rules management systems, and enterprise search tools.

Principle: Artificial intelligence systems form a broad and diverse set of systems that can replicate human decision making for certain types of well-defined problems.

The term artificial intelligence (AI) is used to describe computers with the ability to mimic or duplicate the functions of the human brain. The objective of building AI systems is not to replace human decision making but to replicate it for certain types of well-defined problems.

Artificial intelligence systems include the people, procedures, hardware, software, data, and knowledge needed to develop computer systems and machines that can simulate human intelligence processes, including learning (the acquisition of information and rules for using the information), reasoning (using rules to reach conclusions), and self-correction (using the outcome from one scenario to improve its performance on future scenarios).

Intelligent behavior encompasses several characteristics, including the abil- ities to learn from experience and apply this knowledge to new experiences, handle complex situations and solve problems for which pieces of information might be missing, determine relevant information in a given situation, think in a logical and rational manner and give a quick and correct response, and understand visual images and process symbols. Computers are better than peo- ple at transferring information, making a series of calculations rapidly and accurately, and making complex calculations, but human beings are better than computers at all other attributes of intelligence.

Artificial intelligence is a broad field that includes several key components, such as expert systems, robotics, vision systems, natural language processing, learning systems, and neural networks.

An expert system consists of hardware and software that stores knowledge and makes inferences, enabling a novice to perform at the level of an expert. An expert system is made up of a collection of integrated and related compo- nents, including a knowledge base, an inference engine, an explanation facility, a knowledge acquisition facility, and a user interface.

Robotics is a branch of engineering that involves development and manu- facture of mechanical or computer devices that can perform tasks that require a high degree of precision or are tedious or hazardous for human beings, such as painting cars or making make precision welds.

Vision systems include hardware and software that permit computers to capture, store, and manipulate images and pictures (e.g., face-recognition software).

CHAPTER 10 • Knowledge Management and Specialized Information Systems 443

Natural language processing allows the computer to understand and react to statements and commands made in a “natural” language, such as English.

Learning systems use a combination of software and hardware to allow a computer to change how it functions or reacts to situations based on feedback it receives (e.g., a computerized chess game).

A neural network is a computer system that can recognize and act on pat- terns or trends that it detects in large sets of data.

A genetic algorithm is an approach to solving problems based on the the- ory of evolution and the concept of survival of the fittest. Intelligent agents consist of programs and a knowledge base used to perform a specific task for a person, a process, or another program.

Principle: Multimedia and virtual reality systems can reshape the interface between people and information technology by offering new ways to communicate information, visualize processes, and express ideas creatively.

Multimedia is content that uses more than one form of communication— such as text, graphics, video, animation, audio, and other media that can be used to help an organization efficiently and effectively achieve its goals. Multi- media tools can be used to help an organization achieve its goals through the production of compelling brochures, presentations, reports, and documents. Although not all organizations use the full capabilities of multimedia, most use text and graphics capabilities. Presentation software such as Microsoft PowerPoint, Prezi, and Swipe can be used to develop presentations with sound and animation that can be displayed on a large screen. Other applica- tions of multimedia include audio, video, and animation. File compression and conversion are often needed in multimedia applications to import or export analog files and to reduce file size when storing multimedia files and sending them to others. Designing a multimedia application requires careful thought to get the best results and achieve corporate goals.

A virtual reality system enables one or more users to move and react in a computer-simulated environment. Virtual reality simulations require special interface devices that transmit the sights, sounds, and sensations of the simu- lated world to the user. These devices can also record and send the speech and movements of the participants to the simulation program. Thus, users can sense and manipulate virtual objects much as they would real objects. This natural style of interaction gives the participants the feeling that they are immersed in the simulated world.

Virtual reality can also refer to applications that are not fully immersive, such as mouse-controlled navigation through a three-dimensional environment on a graphics monitor, stereo viewing from the monitor via stereo glasses, and stereo projection systems. Some virtual reality applications allow views of real environments with superimposed virtual objects. Augmented reality, a newer form of virtual reality, can superimpose digital data over real photos or images. Virtual reality applications are found in medicine, education and training, real estate and tourism, and entertainment.

Principle: Specialized systems can help organizations and individuals achieve their goals.

A number of specialized systems have recently appeared to assist organiza- tions and individuals in new and exciting ways. Assistive technology systems include a wide range of assistive, adaptive, and rehabilitative devices to help people with disabilities perform tasks that they were formerly unable to accom- plish or had great difficulty accomplishing. Game theory is a mathematical

444 PART 3 • Business Information Systems

theory that helps to develop strategies for maximizing gains and minimizing losses while adhering to a given set of rules and constraints. Informatics is the combination of information technology with traditional disciplines such as medicine or science, while considering the impact on individuals, organiza- tions, and society. It represents the intersection of people, information, and technology.

Key Terms

artificial intelligence

artificial intelligence system

assistive technology system

business rule management system (BRMS)

community of practice (CoP)

domain expert

electronic discovery (e-discovery)

enterprise search

enterprise search software

expert system

explanation facility

explicit knowledge

game theory

genetic algorithm

IF-THEN statement

inference engine

informatics

intelligent agent

intelligent behavior

joint problem solving

knowledge acquisition facility

knowledge engineer

knowledge management (KM)

knowledge user

learning systems

metadata

multimedia

natural language processing

neural network

perceptive system

robotics

rule

shadowing

organizational network analysis (ONA)

tacit knowledge

virtual reality system

vision system

Chapter 10: Self-Assessment Test

Knowledge management allows organizations to share knowledge and experience among its workers.

1. knowledge is knowledge that is documented, stored, and codified.

2. Tacit knowledge is extremely useful and can be easily shared with others. True or False?

3. Which of the following is not a benefit associated with knowledge management? a. It helps leverage the expertise of people

across the organization. b. It helps capture the expertise of key indivi-

duals before they retire. c. It helps contain the specialized knowledge of

experts to a few people on a need-to-know basis.

d. It fosters innovation by encouraging the free flow of ideas.

4. The initial step in selling and implementing a knowledge management project is . a. get employee buy in b. identify valuable tacit knowledge c. start with a small pilot project and enthusiastic

participants d. connect knowledge management to goals and

objectives

Artificial intelligence systems form a broad and diverse set of systems that can replicate human decision making for certain types of well-defined problems.

5. The use of enables humans and computers to find a good solution, although not optimal, to a complex problem. a. perceptive systems b. heuristics c. visual systems d. robotics

CHAPTER 10 • Knowledge Management and Specialized Information Systems 445

6. The component of an expert sys- tem seeks information and relationships to pro- vide answers, predictions, and suggestions similar to the way a human expert would. a. knowledge base b. explanation facility c. inference engine d. user interface

7. Robots are increasingly being used in surgical procedures including open-heart surgery. True or False?

8. Facebook’s AI vision system called DeepFace can correctly tell if two photos show the same person with accuracy. a. less than 50 percent b. between 50 percent and 75 percent c. between 75 percent and 95 percent d. over 95 percent

9. is a type of artificial system that enables a computer to change how it functions or reacts to situations based on feedback it receives. a. Neural network b. Voice recognition system c. Learning system d. Vision system

10. is a computer system that can rec- ognize and act on patterns or trends that it detects in large sets of data. a. Neural network b. Voice recognition system c. Learning system d. Vision system

11. algorithms solve problems using the concept of survival of the fittest as a problem- solving strategy.

Multimedia and virtual reality systems can reshape the interface between people and information

technology by offering new ways to communicate information, visualize processes, and express ideas creatively.

12. are some of the file formats that can be used to store graphic images. a. MP3, WAV, and MIDI b. AVI, MPEG, and MOV c. DOC and DOCX d. EPS, GIF, JPEG, PNG, TIFF, and Raw

13. When video files are played while they are being downloaded from the Internet, it is called video.

14. software is used widely to create outstanding graphics for feature films and has been used on every Visual Effects Academy Award for over 15 years. a. Apple’s Final Cut Pro b. Adobe Audition c. Prezi d. Pixar’s RenderMan

15. acquired the virtual reality com- pany Oculus VR. a. Apple b. Microsoft c. Facebook d. Google

Specialized systems can help organizations and individuals achieve their goals.

16. systems include a wide range of devices that help people with disabilities to per- form tasks that they were formerly unable to accomplish or had great difficulty accomplishing.

17. involves the use of information systems to develop competitive strategies for people, organizations, or even countries.

Chapter 10: Self-Assessment Test Answers

1. Explicit 2. False 3. c 4. d 5. b 6. c 7. True 8. d 9. c

10. a 11. Genetic 12. d 13. streaming 14. d 15. c 16. Assistive technology 17. Game theory

Review Questions

1. Identify and briefly discuss the six processes that comprise knowledge management.

2. Briefly explain the difference between explicit and tacit knowledge. Give an example of each.

446 PART 3 • Business Information Systems

3. Identify and briefly describe two processes fre- quently used to capture explicit knowledge.

4. Name four key steps in selling and implementing a knowledge management project.

5. What is a community of practice (CoP)? Give an example of a CoP. What are some of the advan- tages of participating in a CoP?

6. What is organizational network analysis? How is it used?

7. What is enterprise search software and how is it used?

8. How would you define artificial intelligence? 9. Identify several specific characteristics of intelli-

gent behavior. 10. Identify six major branches of artificial

intelligence. 11. What are the fundamental components of an

expert system and what function does each perform?

12. Give several examples of robots being used in the real world. What advantages do robots provide?

13. What is DeepFace and how is it used? 14. What is natural language processing? 15. What is a learning system? Give an example of a

learning system. 16. What is a neural network? Give an example of a

neural network. 17. What is a genetic algorithm? Give an example of

the use of a genetic algorithm. 18. Identify and briefly describe five forms of media

that can be used to help an organization achieve its goals. Identify at least two software programs that are used to work with each of these media types.

19. What is the difference between file conversion and file compression?

20. What is a virtual reality system? Identify three areas of virtual reality application.

21. What is an assistive technology system? 22. What is game theory? Identify two applications of

game theory. 23. What is informatics?

Discussion Questions

1. You are an entry-level manager for the customer service desk of a telecommunications firm that provides telephone, Internet access, and cable TV services. A knowledge management system would be useful to capture, store, and retrieve much of the explicit and tacit knowledge needed to provide excellent service. An expert system would prove valuable in helping customer service reps to handle common, reoccurring problems. The organization only has the time and resources to develop one of these two systems. What factors must you consider in making the choice of which system to develop?

2. We are capable of building computers that exhibit human-level intelligence. Are there cer- tain areas of application where we should push to accelerate the building of such computers? Why these application areas? Are there certain areas of application we should avoid? Why these applica- tion areas?

3. Many of us use heuristics each day in completing ordinary activities—such as planning our meals, executing our workout routine, or determining what route to drive to school or work. Imagine that you are developing a set of heuristics for deciding which social invitations to accept. What rules or heuristics would you include?

4. How could you use a community of practice to help you in your work or studies? How would you go about identifying who to invite to join the CoP?

5. A bank is considering implementing a business rules management system for assessing the risk

and creditworthiness of individuals as part of the loan approval process. What might be the bene- fits of such a system? What are some of the factors that must be weighed in this decision? What potential legal or ethical issues might arise in the use of such a system?

6. Describe a situation in which the use of an expert system would be highly practical as well as very beneficial.

7. What are some of the routine day-to-day house- hold tasks that robots are able to accomplish today? What additional tasks might they be able to do in the near future? Are there certain limita- tions that restrict the kind of household chores robots can do?

8. Describe how a combination expert system and natural language processing could be used to provide student counseling for registering for classes. How successful do you think such a sys- tem would be? Explain.

9. Discuss the similarities and differences between learning systems and neural systems. Give an example of how each technology might be used.

10. What is the differences and similarities between a database and a knowledge base?

11. Describe how game theory might be used in a business setting.

12. Describe how augmented reality can be used in a classroom. How could it be used in a work setting?

13. Describe how assistive living systems might ben- efit the residents of a nursing home.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 447

Problem-Solving Exercises

1. You are investigating the use of automated robots to replace some of the tasks performed by waiters and waitresses in your neighborhood restaurant. The robots are capable of clearing tables and delivering food to customers. Customer interac- tions to order food would continue to be done by humans. You estimate that you could reduce your staff from 15 workers making $15/hour to around 7 workers. In addition, you are sure that use of the robots would attract additional customers perhaps increasing sales by $100,000/year. The robots cost $75,000 each and you figure you will need two or three depending on how fast busi- ness increases. Perform an analysis to determine

if hiring robots and letting humans go makes good economic sense.

2. Shoot a brief video of you and some friends. Download the video to a computer in your school’s computer lab that has video editing software. Use this software to experiment making edits to the video. Write a brief report summariz- ing your experience learning and using this software.

3. Capture a large text file and use a data compres- sion program to reduce the size of the file and email it to a friend. Verify that your friend is able to open and decompress the file. Repeat this exercise with a large video file.

Team Activities

1. You and your team are challenged to visit the local entertainment centers in search of the most realistic virtual reality game that requires a player to wear a head-mounted display. What makes this game so realistic? Is there any way it could be improved? Talk to the manager and try to identify the hardware and software employed. Summarize your findings in a brief report.

2. Work with your team to design an expert system to predict how many years it will take a typical student to graduate from your college or univer- sity. Some factors to consider include the major

the student selects, the student’s SAT score, the number of courses taken each semester, and the number of parties or social activities the student attends each month. Identify six other factors that should be considered. Develop six IF-THEN rules or cases to be used in the expert system.

3. Have your team members explore the use of assistive technology systems by recent combat veterans. Write a short paper summarizing your findings and the advantages and disadvantages of these type systems.

Web Exercises

1. NelNet is one of the nation’s largest student loan servicing companies. The U.S. Department of Education contracts with companies like NelNet to provide servicing options for their student loans including:

● Take in and apply payments to student loans

● Administer the transfer of student loans ● Process student loan programs, such as for-

giveness, forbearance, and deferment ● Report to credit agencies past due payments

● Seek repayment from defaulted loans

NelNet chose to deploy a knowledge manage- ment system called OpenText Process Suite. Go online and investigate the features and

capabilities of this suite of software products. What functions does OpenText provide that can augment and assist customer relationship man- agement (CRM) systems? Research and briefly document students’ recent experience with NelNet.

2. Do research to identify instances in which robots have taken jobs away from human employees. Do you believe that the increasing use of robots should be encouraged? Why or why not? Write a one-page summary of your findings and opinions.

3. Use the Internet to identify several applications of neural networks. Write a brief summary of these applications.

Career Exercises

1. Identify several career fields that are likely to be negatively impacted by artificial intelligence

applications. Can you identify any career fields that might be positively impacted?

448 PART 3 • Business Information Systems

2. Do research to explore career opportunities in healthcare informatics. What sort of training and experience is necessary for such a career? What sort of role does this prepare one to fulfill? What is the forecast for the number of job openings and starting salaries? Does this career field inter- est you?

3. Imagine that you are forming a community of practice to deal with the issues associated with transitioning to a new career. Identify people from your experience who you would like to include as members. Why factors did you con- sider when you selected these people? Identify three key topics you would like the community to address.

Case Studies

Case One

The NASA Knowledge Map At 11:38 a.m. on January 28, 1986, the space shuttle orbiter Challenger launched from Cape Canaveral, Florida. Less than a second later, gray smoke streamed out from a hot flare burning in the rocket motor. The flare ignited liquid hydrogen and nitrogen inside the fuel tank, which exploded 73 seconds after liftoff. The Challenger was torn apart, and all seven astronauts were killed.

In the days and weeks following the disaster, it became clear that two O-ring seals within the rocket booster had failed. Engineers working for the space agency had warned of just such a failure. In particular, they had expressed concerns that the O-ring seals could fail when outside temperatures dropped below 53 degrees Fahrenheit. On the morning of January 28, the temperature was 36 degrees. The launch pad was covered with solid ice.

In response to the Challenger disaster, NASA established the Program and Project Initiative whose purpose was to improve individual competency for NASA employees—and to prevent another catastrophe. The Challenger, however, was followed by the failure of three expensive Mars missions. The software system used for the Mars Climate Orbiter mission erred when one part of the software used pound-force units to calculate thrust, whereas another part used the newton metric unit. Less than a month later, the Mars Polar Lander crashed into the surface of the planet at too high velocity— triggering the failure of a concurrent mission, the Mars Deep Space 2 probes. A review of the Deep Space 2 mission revealed that NASA engineers had decided to skip a complete system impact test in order to meet the project’s tight deadline. In the wake of these failures, NASA sought to improve communication and collaboration among teams. Yet in 2003, a large piece of insulation foam broke off from the Columbia space shuttle during launch, creating a hole in its wing, ultimately causing a catastrophic breach of the shuttle during reentry; again, all seven astronauts on board were killed.

These terrible losses brought about a fundamental change in NASA’s approach to knowledge management. In 1976, NASA had created the Office of the Chief Engineer (OCE), which was initially staffed by only one employee whose job was to offer advice and expertise on NASA’s administration. In response to the Challenger disaster, NASA established the Academy of Project/Program and Engineering Leadership (APPEL) as a resource for developing NASA’s technical staff. In 2004, the agency moved APPEL to the OCE in order to promote talent development through the analysis

of lessons learned and through knowledge capture—the codification of knowledge. The purpose was to improve not only individual but also team performance and to overcome the disconnect between the different engineering and decision-making teams across the huge organization. The overarching goal was to create an organization that learns from its mistakes. APPEL emphasized not only technical training curriculum but also the sharing of practitioner experience, storytelling, and reflective activities. In 2012, NASA furthered this initiative and established the role of chief knowledge officer whose mission is to capture implicit and explicit knowledge. Today, the agency has an extensive knowledge management system called NASA Knowledge Map, which is a tool that helps employees navigate the enormous collection of knowledge within NASA. The map encompasses six major categories: (1) Case Studies and Publications, (2) Face-to-Face Knowledge Services, (3) Online Tools, (4) Knowledge Networks, (5) Lessons Learned and Knowledge Processes, and (6) Search/Tag/Taxonomy Tools.

Fifteen organizations within NASA contribute to Case Studies and Publications. The Goddard Space Flight Center, for example, publishes studies that range from analysis of the Challenger disaster to an analysis of a protest submitted by a NASA contractor who lost a follow-up contract. The latter case may not seem critical, but in one such case, the Office of Inspector General had to launch a formal investigation that cost NASA time, money, and energy. This case study was then integrated into the APPEL curriculum with the goal of avoiding the mistakes that led to the protest. The Johnson Space Center issues oral history transcripts, as well as newsletters, case studies, and reports. The Jet Propulsion Laboratory publishes conference papers and a Flight Anatomy wiki that tracks prelaunch and in-flight anomalies.

Face-to-Face Knowledge Services comprise programs that are conducted in person at many locations, including, for example, workshops presented by the NASA Engineering and Safety Center. Within the Online Tools category are video libraries, portals, document repositories, and synchronous and asynchronous collaboration and sharing sites. Some of these tools are quite sophisticated. For example, Human Exploration and Operations (HEO) deploys a GroupSystems Think Tank decision support tool to improve group decision making. The Knowledge Networks category includes information about formal and informal communities of practice, mass collaborative activities, and methods for locating and accessing experts, and group workspaces for projects such as static code analysis.

Twenty organizations within NASA contribute data to the Lessons Learned and Knowledge Processes databases, which

CHAPTER 10 • Knowledge Management and Specialized Information Systems 449

capture and store knowledge, lessons learned, and best practices. These include, for example, HEO’s knowledge- based risks library with topics covering project management, design and development, systems engineering, and integration and testing. HEO also sponsors lessons-learned workshops and forums on topics such as solar array deployment, shuttle transition and retirement, system safety, and risk management.

Finally, the system’s Search/Tag/Taxonomy Tools allow individuals to access organization-specific sites as well as the abundance of materials offered through the five other KM programs. This final category within the KM system may be the most important, as NASA’s own inspector general issued a report indicating that the tremendous wealth of KM resources is still significantly underutilized. For instance, NASA managers rarely consult the Lessons Learned Information System (LLIS) despite NASA requirements that they do so. The Glenn Research Center received $470,000 over two years to support LLIS activities, but contributed only five reports to the system during that time. Moreover, the inspector general concluded that inconsistent policy direction, disparate KM project development, and insufficient coordination marginalize the system.

NASA is clearly at the bleeding edge of large-scale KM system development, creating the tools of the future. APPEL and other NASA teams are able to make use of some amazing tools that are being developed within the agency. It may be, however, that NASA’s KM system suffers from the same disjointed development and communication barriers that led to the space shuttle disasters and the failures of the Mars missions. Yet, it is vital that NASA learn to make use of its state-of-the-art KM system as the success of every NASA mission requires that thousands of employees are able to make the most of NASA’s vast collection of knowledge.

Critical Thinking Questions 1. How is the KM system at NASA different from other

KM systems that you have studied within the chapter? How is it similar?

2. What steps can NASA take to make sure that the KM system is better utilized by individuals and teams?

3. What can NASA do to ensure that individuals and teams can find what they need within the mountain of data residing within the KM system?

4. Is NASA’s KM system, as it exists now, a good way to combat the type of failures the agency has experienced in the past? If not, how could the KM system be chan- ged to support mission success?

5. Are there other measures that NASA should take in addition to or in conjunction with the development of its KM system?

SOURCES: Oberg, James, “7 Myths about the Challenger Shuttle Disaster,” NBC News, January 25, 2011, www.nbcnews.com/id /11031097/ns/technology_and_science-space/t/myths-about-challenger -shuttle-disaster/#.U2AsyIFdUrU; Atkinson, Joe, “Engineer Who Opposed Challenger Launch Offers Personal Look at Tragedy,” NASA Researcher News, October 5, 2012, www.nasa.gov/centers/langley/news/researcher news/rn_Colloquium1012.html; “Challenger Disaster,” History Channel, www.history.com/topics/challenger-disaster, accessed April 29, 2014; “Failure as a Design Criteria,” Plymouth University, www.tech.plym.ac .uk/sme/interactive_resources/tutorials/failurecases/hs1.html, accessed

April 29, 2014; Lipowicz, Alice, “Is NASA’s Knowledge Management Program Obsolete?,” GCN Technology, Tools and Tactics for Public Sector IT, March 19, 2012, http://gcn.com/Articles/2012/03/15/NASA -knowledge-management-IG.aspx; Luttrell, Anne, “NASA’s PMO: Building and Sustaining a Learning Organization,” Project Management Institute, www.pmi.org/Learning/articles/nasa.aspx, accessed February 9, 2015; Hoffman, Edward J. and Boyle, Jon, “Tapping Agency Culture to Advance Knowledge Services at NASA,” ATD, September 15, 2013, www.td.org /Publications/Magazines/The-Public-Manager/Archives/2013/Fall/Tap ping-Agency-Culture-to-Advance-Knowledge-Services-at-NASA; “Knowledge Map,” NASA, http://km.nasa.gov/knowledge-map/, accessed February 9, 2015.

Case Two

Doctor on Demand Enables Physicians to Make House Calls In addition to cost, provider availability and travel time are barriers for many Americans seeking access to healthcare services. In fact, a recent study of 4,000 patients determined that, on average, patients spend 38 minutes on travel time to and from outpatient appointments. Improving patient’s access to care continues to be a priority for healthcare providers and government agencies across the United States, and an increasing number of companies have begun offering telemedicine services, such as video-based doctors’ appointments, as a potential solution.

Founded in 2013, Doctor on Demand, offers the possibility of increasing access to health care through video visits with doctors who can diagnose and treat a range of noncritical symptoms for patients who are unable or unwilling to visit a clinic. Using the Doctor on Demand services, patients can connect with one of more than 1,400 licensed physicians through the company’s Web site using a Chrome, Firefox, or Safari browser or via an Android or iOS app. In addition to video conferences, the Doctor on Demand app allows patients to upload high-resolution images so that doctors can better assess certain conditions.

The top conditions treated by the service are cold and flu symptoms, sore throats, urinary tract infections, skin rashes, diarrhea and vomiting, eye issues, sports injuries, and travel- related illnesses. The site also offers video visits with board- certified lactation consultants for women who are breastfeeding. In addition, patients who need psychological or psychiatric services can consult with mental health professionals via the service.

According to Adam Jackson, CEO of Doctor on Demand, the most frequent users of the company’s services are working mothers, who often have questions about their children’s health but aren’t always able (or willing) to take time off to get every question answered. According to Jackson, 92 percent of video consultations require no in- person follow-up.

Although Doctor on Demand suggests that patients have access to Wi-Fi to ensure the highest quality appointment, the company promises a smooth experience as long as patients have a 4G or LTE connection. Patients who have connection problems can also switch to audio only to complete a visit, if necessary. The Doctor on Demand network runs on a cloud- based platform run by Amazon Web Services. Due to the nature of the communication, the company had to go through several steps to ensure that all of its infrastructure

450 PART 3 • Business Information Systems

was compliant with HIPAA (Health Insurance Portability and Accountability Act) requirements.

Most experts predict a shift to telemedicine, including video doctors’ visits, will continue. In fact, a report by analytics company IHS Technology predicts that video consultations will increase from 2 million in 2015 to 5.4 million by 2020. For some patients, however, technology limitations will continue to impede their ability to access health care through telemedicine services. A grainy connection or one that cuts out in the middle of an appointment is unlikely to result in a high quality of care. Other hurdles that will also need to be overcome include patient’s privacy concerns, patients’ uncertainty about when a video appointment is appropriate in terms of symptoms, and patients’ lack of trust that a virtual provider can accurately diagnose and treat them. That concern was reinforced by a recent study published in the JAMA Internal Medicine that found significant variations in the quality of care provided by different companies offering virtual visits for the diagnosis and treatment of common acute illnesses.

Critical Thinking Questions 1. Would you consider using Doctor on Demand or a

similar service to access treatment for a minor health- care issue? If not, which aspects of the service are most concerning to you (privacy, quality of care, security or other technology issues, etc.)?

2. Do more research online about Doctor on Demand and two of its competitors (such as Amwell, MDLive, and Teladoc). What information does each company pro- vide on its Web site that is designed to ease patients’ concerns about privacy, quality, and technology-

related issues? Which company does the best job of convincing you that their service is safe and secure?

3. In the study on patient travel time, researchers found that minority patients and those who were unem- ployed faced longer travel times when visiting a doc- tor. Rural Americans also often have more difficulty accessing health care. Is a video-based telemedicine app likely to improve access for those populations? How might this technology be used in a way that would be more likely to improve healthcare access for those populations?

SOURCES: Doyle, Kathryn, “Study: How Long You Wait to See a Doctor Is Linked to Race, Employment,” Huffington Post, October 6, 2015, www.huffingtonpost.com/entry/study-how-long-you-wait-to-see-a-doc tor-is-linked-to-race-employment_us_5613b0cbe4b0baa355ad2621; “Troubleshooting,” Doctor on Demand, https://doctorondemand.zen desk.com/hc/en-us/sections/200218868-Troubleshooting, accessed April 9, 2016; “Our Mission,” Doctor on Demand, www.doctorondemand.com /our-mission, April 8, 2016; Lapowsky, Issie, “Video Is about to Become the Way We All Visit the Doctor,” Wired, www.wired.com/2015/04 /united-healthcare-telemedicine; Van Thoen, Lindsay, “Healthcare IT is Failing (And It Needs AWS), Logicworks (blog), July 20, 2015, www .logicworks.net/blog/2015/07/healthcare-cloud-saas-aws; Japsen, Bruce, “Doctors’ Virtual Consults with Patients to Double by 2020,” Forbes, August 9, 2015, www.forbes.com/sites/brucejapsen/2015/08/09/as-tele health-booms-doctor-video-consults-to-double-by-2020/#639cbc4e5d66; “Press Release: “39% of Tech-Savvy Consumers Have Not Heard of Telemedicine: HealthMine Survey,” HealthMine, March 27, 2016, www .prnewswire.com/news-releases/39-of-tech-savvy-consumers-have-not -heard-of-telemedicine-healthmine-survey-300241737.html#continue -jump; Schoenfield, Adam J., et al., “Variation in Quality of Urgent Health Care Provided during Commercial Virtual Visits,” JAMA Internal Medicine, April 4, 2016, http://archinte.jamanetwork.com/article.aspx? articleid=2511324.

Notes

1. “About TMW,” TMW, www.tmwsystems.com/about-tmw, accessed March 22, 2016.

2. “TMW Deploys Bloomfire for Social Learning,” Bloom- fire (blog), April 16, 2015, https://bloomfire.com /blog/tmw-deploys-bloomfire-for-social-learning.

3. “About the Firm,” White & Case, www.whitecase.com /about/, accessed February 5, 2015.

4. Britt, Phil, “Creating a More Knowledgeable, Nimble Organization,” KM World, January 30, 2015, www .kmworld.com/Articles/Editorial/Features/Creating-a- more-knowledgeable-nimble-organization-101536.aspx.

5. “Knowledge Management & Transfer Model {Techniques and Forms},” Division of Personnel, Department of Administrative Service, State of New Hampshire, www .admin.state.nh.us/hr/documents/Workforce_Develop ment/Knowledge%20Management%20&%20Transfer% 20Model.doc, accessed February 26, 2015.

6. “Kaleo Software Secures $7 Million to Accelerate Inno- vation and Growth,” Kaleo, October 15, 2015, www .kaleosoftware.com/press/seriesa.

7. “This is GSA,” U.S. General Services Administration (GSA), www.gsa.gov/thisisgsa/#/welcome, accessed April 2, 2016.

8. “Communities,” DigitalGov, www.digitalgov.gov/commu nities, accessed April 3, 2016.

9. “GSA Leads Customer Service Community of Practice,” FedScoop, October 7, 2015, http://fedscoop.com /gsa-leads-customer-experience-community-of-practice.

10. “Customer Experience Community,” U.S. General Ser- vices Administration (GSA), www.digitalgov.gov/commu nities/customer-experience-community, accessed April 2, 2015.

11. Grimes, Seth, “Metadata, Connection, and the Big Data Story,” Breakthrough Analysis, April 26, 2014, http:// breakthroughanalysis.com/2014/04/26/metadata-con nection-and-the-big-data-story/.

12. “Using Organizational Network Analysis to Build a Better Business,” Society for Human Resource Manage- ment,” May 2015, www.shrm.org/publications/hrmaga zine/editorialcontent/2015/0515/pages/05115-organiza tional-network-analysis.aspx.

13. “HanseMerkur Automates Its Service Billing Processes Using inubit,” Bosch Financial Software, www.bosch -si.com/media/en/finance_7/documents_2/brochures_1 /success_stories/insurance_3/hansemerkur.pdf, accessed February 7, 2015.

14. “Adobe Fast Facts,” Adobe, www.images.adobe.com/con tent/dam/acom/en/fast-facts/pdfs/fast-facts.pdf, accessed April 6, 2016.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 451

15. “Adobe,” Progress, www.progress.com/customers/adobe, accessed February 7, 2015.

16. “About Us,” DBS, www.dbs.com/about-us/default.page, accessed January 29, 2016.

17. “DBS,” Progress, www.progress.com/customers/dbs, accessed January 29, 2016.

18. Andrews, Whit and Koehler-Kruener, Hanns, “Magic Quadrant for Enterprise Search,” Gartner, August 19, 2015, www.gartner.com/doc/reprints?id=1 -2LH0EF0&ct=150820&st=sg.

19. “Company Information,” Harris Corporation,” http://har ris.com/about, accessed April 6, 2017.

20. “Harris Corporation Increases Productivity with Coveo,” Coveo, www.coveo.com/~/media/Files/CaseStudies/Har ris_Corporation_Increases_Productivity_with_Coveo _ for_Advanced_Enterprise_Search.ashx, accessed April 6, 2016.

21. Deutchman, Leonard, “2015 in Review: What Happened, Is Happening and Will Happen,” Legal Intelligencer, January 5, 2016, www.thelegalintelligencer.com/latest -news/id=1202746177129/2015-in-Review-What-Hap pened-Is-Happening-aibnd-Will-Happen?mcode =1395262324557&curindex=918&slreturn=2016 0307160543.

22. Krause, Jason, “The Battle of Rio Tinto–Predictive Cod- ing Hits Snags in Marquee Case,” Association of Certified E-Discovery Specialists (blog), July 24, 2015, www.aceds .org/the-battle-of-rio-tinto-predictive-coding-hits-snags -in-marquee-case.

23. McCarthy, J., Minsky, M.L., Rochester, N., and Shannon, C.E., “A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence,” August 31, 1955, www-formal.stanford.edu/jmc/history/dartmouth/dart mouth.html.

24. “What is Watson,” http://www.ibm.com/smarterplanet /us/en/ibmwatson/what-is-watson.html, accessed April 26, 2016.

25. “By 2029 No Computer—or ‘Machine Intelligence’—Will Have Passed the Turing Test,” Long Bets, http://longbets .org/1/, accessed May 2, 2014.

26. “The Aspen Institute: Google’s Eric Schmidt with Walter Isaacson on ‘The New Digital Age,’ YouTube video, 1:09, July 24, 2013, www.youtube.com/watch? v=3Ox4EMFMy48.

27. Gee, Sue, “Mitsuku Wins Loebner Prize 2013,” I-Programmer, September 15, 2013, www.i-programmer .info/news/105-artificial-intelligence/6382-mitsuku-wins -loebner-prize-2013.html.

28. 20Q Web site, www.20q.net, accessed May 2, 2014. 29. “FICO® Blaze Advisor® Decision Rules Management

System,” FICO, www.fico.com/en/products/fico-blaze -advisor-decision-rules-management-system#overview, accessed April 7, 2016.

30. Capek, Karel, “Beyond the Robots,” Legacy.com, www .legacy.com/news/legends-and-legacies/karel-capek –beyond-the-robots/302/#sthash.RQyp1B1R, accessed May 3, 2014.

31. The Web site for iRobot Web page, www.irobot.com, accessed April 8, 2016.

32. “REC (Robotics Engineering Curriculum),” Intelitek, www.intelitek.com/engineering/rec_curriculum, accessed April 8, 2016.

33. “Meet Dash and Dot,” Wonder Workshop, www.make wonder.com, accessed April 8, 2016

34. Martin, Alexander, “SoftBank, Alibaba Team Up on Robot,” The Wall Street Journal, June 18, 2015, www.wsj .com/articles/pepper-softbanks-emotional-robot-goes -global-1434618111.

35. Smith, Aaron and Janna Anderson, “Predictions for the State of AI and Robotics in 2025,” PewResearchCenter, August 6, 2014, www.pewinternet.org/2014/08/06/pre dictions-for-the-state-of-ai-and-robotics-in-2025.

36. “Facebook’s New Face Recognition Knows You from the Side,” CNN Money, April 4, 2014, http://money.cnn.com /2014/04/04/technology/innovation/facebook-facial-rec ognition/.

37. Elgan, Mike, “Is Facial Recognition a Threat on Facebook and Google,” June 29, 2015, www.computerworld.com /article/2941415/data-privacy/is-facial-recognition-a -threat-on-facebook-and-google.html.

38. D’Orazio, Dante, “Google Now and Speech Recognition Get Big Updates in Android 4.4 Kitkat,” The Verge, October 31, 2013, www.theverge.com/2013/10/31 /5051458/android-kit-kat-bring-big-updates-to-google -now-and-speech-recognition.

39. “Neural Tools Used for Tumor Diagnosis,” Palisade Case Studies, www.palisade.com/cases/katherinenhospital .asp?caseNav=byIndustry, accessed May 3, 2014.

40. Taft, Darryl, “IBM Partners with American Cancer Society on Watson Cancer Advisor,” eWeek, April 12, 2016, http:// www.eweek.com/database/ibm-partners-with-american -cancer-society-on-watson-cancer-advisor.html.

41. “KEITV Customer Success Stories—Sunkist Growers,” http://ketiv.com/company/customer-success-stories/sunk ist, accessed May 6, 2014.

42. Seymour, Mike, “Pixar’s RenderMan Turns 25 (Exclu- sive),” fx Guide, July 25, 2013, www.fxguide.com/fea tured/pixars-renderman-turns-25.

43. King, Leo, “Facebook, Oculus, and Businesses’ Thirst for Virtual Reality,” Forbes, March 30, 2014, www.forbes .com/sites/leoking/2014/03/30/facebook-oculus-and-busi nesses-thirst-for-virtual-reality.

44. Elgan, Mike, “Why Virtual Reality Is the Next Social Net- work,” Computerworld, November 2, 2015, www.compu terworld.com/article/2999819/social-media/why-virtual -reality-is-the-next-social-network.html?phint=newt% 3Dcomputerworld_thisweek&phint=idg_eid% 3D144dfe65577ba4663c4cefdafdff5932#tk. CTWNLE_nlt_thisweek_2015-11-02.

45. Casti, Taylor, “6 Ways Virtual Reality Is Already Chang- ing the World (No Facebook Required),” Huffington Post, March 28, 2014, www.huffingtonpost.com/2014 /03/28/virtual-reality-uses-medicine-autism-ptsd-burn -amputee-victims_n_5045111.html.

46. Lardinois, Frederic, “Google Expands Its VR Program for Students to More U.S. Schools,” TechCrunch, November 9, 2015, http://techcrunch.com/2015/11/09/google -expands-its-vr-program-for-students-to-more-u-s-schools.

47. “Introducing the Expeditions Pioneer Program,” Google, www.google.com/edu/expeditions, accessed April 8, 2016.

48. King, Leo, “Ford, Where Virtual Reality Is Already Manufacturing Reality,” Forbes, May 3, 2014, www .forbes.com/sites/leoking/2014/05/03/ford-where-virtual -reality-is-already-manufacturing-reality.

452 PART 3 • Business Information Systems

49. Gaudiosi, John, “Here’s Why NFL Teams Are Training in Virtual Reality,” Fortune, August 10, 2015, http://fortune .com/2015/08/10/strivr-virtual-reality-nfl.

50. “The Story,” STRIVR, www.strivrlabs.com/about, accessed April 9, 2016.

51. “IKEA Launches Pilot Virtual Reality (VR) Kitchen Experience for HTC Vive on Steam,” IKEA, April 4, 2016, www.ikea.com/us/en/about_ikea/newsitem/040516 _Virtual-Reality.

52. Hart, Brian, “Microsoft Targets Enterprise Virtual Reality with Business Oriented VR Headset,” Road to VR, April 1, 2014, www.roadtovr.com/microsoft-enterprise-virtual -reality-business-vr-headset/.

53. “Microsoft Accessibility—Types of Assistive Technology Products,” www.microsoft.com/enable/at/types.aspx, accessed May 12, 2014.

54. “Assistive Listening Systems and Devices,” http://nad .org/issues/technology/assistive-listening/systems-and -devices, accessed May 12, 2014.

55. Savani, Rahul and von Stengel, Bernhard, “Game Theory Explorer—Software for the Applied Game Theorist,” March 16, 2014, www.maths.lse.ac.uk/Personal/stengel /TEXTE/largeongte.pdf.

56. Levenson, Eric, “Jeopardy’s New Game-Theory Devotee Is One to Keep an Eye on,” The Wire, Jan 31, 2014, www .thewire.com/entertainment/2014/01/jeopardys-newest -star-proves-optimal-strategy-really-unfriendly/357609.

57. “Port Resilience Operational / Tactical Enforcement to Combat Terrorism (PROTECT) Model for the United States Coast Guard,” http://teamcore.usc.edu/projects /coastguard/, accessed May 13, 2014.

58. Leventhal, Rajiv, “At Boston Medical Center, Using Ana- lytics to Improve OR Efficiencies,” Healthcare Informat- ics,” March 24, 2016, www.healthcare-informatics.com /article/boston-medical-center-using-analytics-improve -or-efficiencies.

CHAPTER 10 • Knowledge Management and Specialized Information Systems 453