Management Information System

mohamedabbas77

DSS.pptx

Home >Information Systems homework help >Management Information System

Decision support systems

Business intelligence

Business intelligence are the processes, technologies and IT applications that support decision making.

Decision Making

Strategic decision making determines the objectives, resources and policies of an organisation.

Decision for management control is concerned with how efficiently and effectively resources are controlled

Operational decision making determines how to carry out the specific tasks set out by management

Types of Decisions

Decisions can be structured, semi-structured or unstructured.

Structured: repetitive and routine. There is a definite procedure for handling them.

Unstructured: decision maker must provide insight, judgment, evaluation to solve problem.

Semi structured only part of the problem has a clear structure.

In general, operational personnel face more well structured problems.

Strategic planners handled unstructured problems.

All levels of an organization will handle both types of problems, to a greater or lesser extent

Examples:

Senior managers:

Make many unstructured decisions

E.g., Should we enter a new market?

Middle managers:

Make more structured decisions but these may include unstructured components

E.g., Why is order fulfillment report showing a decline in Minneapolis?

Operational managers, rank and file employees

Make more structured decisions

E.g., Does customer meet criteria for credit?

Can you think of any decision support systems that you have used (or use on a daily basis?)

Information systems and levels of decision making

TPS

MIS

DSS

ESS

Type of

Decision

Organizational level

Operational Management Strategic

Accounts Receivable

Production cost overruns

Production facility location

New products, New markets

Structured

Semi- structured

Unstructured

…examples

The decision making process requires:

Intelligence - Identifying and understanding the problems facing the organisation, why, where and with what effects

Design - create possible solutions to problems

Choice - Choose between alternatives

Implementation

Stages in decision making

This graphic illustrates the four stages of decision making introduced in the previous slide, emphasizing that steps can be repeated as needed, depending on the outcome at each stage.

Information systems can only assist in some of the roles played by managers

Classical model of management

Five functions of managers

Planning, organizing, coordinating, deciding, and controlling

More contemporary behavioral models

Actual behavior of managers appears to be less systematic, more informal, less reflective, more reactive, and less well organized than in classical model

Mintzberg’s behavioral model of managers defines 10 managerial roles falling into 3 categories

This slide discusses the idea that, while information systems can assist in making decisions by providing information and tools for analysis, they cannot always improve on decisions being made. Ask the students to provide examples of decisions that an information system might not be able to assist in. Is there any similarity among these example decisions, and what does this say about the types of decisions an information system can help with?

You can understand the complexity and breadth of some of the decisions being made within an organization by looking at the activities of its managers. While the classical model of management sees five functional roles of managers, real-life observation of managers sees far more complexity in managerial activities. Ask the students to recall the five attributes listed in the book as differing greatly from the classical description. Managers (1) perform a great deal of work at an unrelenting pace (2) managerial activities are fragmented; lasting for less than 9 minutes (3) managers prefer current, specific, ad hoc information; (4) manager prefer oral forms of communication and (5) managers give high priority to maintaining a complex web of contacts as an informal information system

Mintzberg’s 10 managerial roles

Interpersonal roles: Figurehead

Leader Liaison

Informational roles: Nerve center

Disseminator

Spokesperson

Decisional roles: Entrepreneur

Disturbance handler

Resource allocator

Negotiator

This slide expands on the behavioral model of managers and lists Mintzberg’s 10 managerial roles that fall into three categories. Which of these roles can be assisted by information systems and which cannot?

Three main reasons why investments in information technology do not always produce positive results

Information quality

High-quality decisions require high-quality information

Management filters

Managers have selective attention and have variety of biases that reject information that does not conform to prior conceptions

Organizational culture

Strong forces within organizations resist making decisions calling for major change

Even in decision-making situations that can be helped by information systems, the information system may fail in helping to solve the problem or lead to a better decision. This slide describes the three main reasons why investments in information systems do not always produce positive results. What is meant by information quality? The text lists seven quality dimensions: accuracy, integrity, consistency, completeness, validity, timeliness, accessibility. Ask students to identify and/or describe these dimensions?

Ask students to provide an example of what a management “filter” might be. Have they ever witnessed someone in a managerial position be unable to recognize or handle a problem because of a “filter?”

Ask students why people within an organization would resist using an information system.

Three kinds of systems for decision support

Management information systems (MIS)

Decision support systems (DSS)

Executive support systems (ESS)

MIS

MIS provide information on a firm’s performance, to help managers control the business

Provide fixed regular scheduled reports, on summarised data, normally extracted from a TPS.

Exception reports highlight exceptional conditions, e.g. sales below target, etc.

E.g. A fast food restaurant system provides information on food, labour and material costs. It may inform management if a given store is not performing to standard, is providing excessive quantities, etc, etc.

DSS – Model driven

DSS support semi-structured and unstructured problem analysis

Early DSS were driven by models. Now DSS can analyze huge amounts of data.

Model driven DSS are typically standalone, providing what if analysis. E.g. voyage estimating, cargo optimising

DSS models

Model:

Abstract representation that illustrates components or relationships of phenomenon; may be physical, mathematical, or verbal model

Statistical models

A statistic al model helps establish relationships, such as relating product sales to differences in age, income or other factors.

Optimization models

Optimization models determine optimal resource allocation to maximize or minimize specific variables such as cost and time.

Forecasting models

Forecasting models are used to forecast future conditions, such as sales, given a range of historical data.

Sensitivity analysis models

Sensitivity analysis models ask “what if” questions to determine the impact of changes in one or more factors.

A statistic al model helps establish relationships, such as relating product sales to differences in age, income or other factors.

Optimization models determine optimal resource allocation to maximize or minimize specific variables such as cost and time.

Forecasting models are used to forecast future conditions, such as sales, given a range of historical data.

Sensitivity analysis models ask “what if” questions to determine the impact of changes in one or more factors.

Continental Airlines – Cargo Prof

Cargo revenue optimisation

Off the shelf package from manugistics, then customised, called it CargoProf

Ensures that Continental sells all available freight space at most profitable price.

System forecasts cargo capacity and sets optimal price.

Continental Airlines – Cargo Prof 2

Continental booking agent transmits freight order request for a given flight

Legacy system captures order data: weight, dimensions, contract price and forwards data to CargoProf

CargoProf checks available capacity on aircraft, and checks the order against a pricing model.

Then considers anticipated passenger baggage and other requirements, (e.g. fuel).

System then accepts or rejects contract price for that flight, depending on whether it is cost effective.

If rejected, the system can check other flights to see if they can profitably carry the cargo.

CargoProf can handle other incremental price changes to rush shipments.

By making shipments more effective, CargoProf saved Continental over 9M USD in two years.

DSS – Data driven

Analyse large pools of data.

Information from TPS is collected in datawarehouses.

Online analytical processing and datamining used to analyse data.

Traditional database queries answer relatively simple questions: “How many Xs were sold between June and September in region A?”

Multidimensional analysis can support more complex queries: “Compare the sales of X and Y between June and September in region A relative to forecast?” or

How many females aged 22 to 30 accessed our website looking for X between Monday and Friday compared to the same group between Friday and Sunday during the whole of 2014?

Datamining

Datamining provides insights into corporate data by finding hidden patterns and relationships in large databases.

These patterns can then help decision making

Typically focus on associations, sequences, classifications (which recognise patterns within a group), clusters (when no groups have been set), forecasts.

Datamining has to use a representative sample

Online analytical processing (OLAP)

Essentially querying against a database

Program extracts data from the database and structures it by individual dimensions, such as region or dealer

OLAP described as human-driven, whereas data mining is technique-driven

Components of a DSS

DSS

Database

External data

TPS

DSS Software

Models

OLAP tools

Datamining tools

User Interface

DSS software system contains the tools for data analysis

(Olap, datamining, mathematical and analytical models, etc)

One of the most widely used models are sensitivity analysis, to provide what-if scenarios. (One of the simplest forms of sensitivity analysis can be performed on excel: break even analysis)

Applications of DSS

There are many ways in which DSS can support manager’s decision making.

It is an ever developing field, and solutions are more powerful year after year.

Examples: Continental Airlines – customer segmentation, target marketing, revenue management

Bank of America – customer profiling

United Airlines - Flight scheduling and passenger demand forecasting

DSS for SCM

SCM is concerned with purchasing, selling and transportation of materials.

SCM systems use data about inventory, logistics, production schedules, costs to help managers search for the best available combinations for moving goods through a supply chain.

DSS for CRM

DSS can help managers analyse pricing, customer retention and new revenue streams.

Can be used to identify profitable customers

Segment customers into smaller groups for more targeted marketing.

Predictive analysis: using historical data and assumptions about future conditions to predict outcomes of events.

For example Body Shop mailed a catalogue to it US customers. It wanted to cut costs, and improve the response rate of those catalogues that it did send out. Using datamining techniques and specific software from Sightward, it was able to identify those customers that were more likely to make a catalog purchase. Revenue per catalog subsequently increased.

Data visualisation and GIS

Data from information systems can be made easier to understand using graphs, tables, charts, images, maps etc.

If data is presented in an interactive graphic form, users can manipulate data.

Geographic Information Systems present data with digitized maps. The software assembles, stores, manipulates and displays geographically referenced information, tying data to points, lines or areas on a map.

Applications are found in research, resource management and development planning.

E.g. GIS help banks find the best locations for new store locations or ATM terminals.

Crime prevention, etc.

South Carolina used a GIS-based program called HAZUS to estimate and map the regional damage and losses resulting from an earthquake of a given location and intensity. HAZUS estimates the degree and geographic extent of earthquake damage across the state based on inputs of building use, type, and construction materials. The GIS helps the state plan for natural hazards mitigation and response.

What makes this map more effective than a printed report or a graph?

Examples

Event: Clinical application http://www.youtube.com/watch?v=mGrgYnU9BsA