Chapter 4

MODELING AND ANALYSIS

Learning Objectives
Understand the basic concepts of management

support system (MSS) modeling Describe how MSS models interact with data and the user Understand some different, well-known model classes Understand how to structure decision making with a few alternatives

Learning Objectives
Describe how spreadsheets can be used for MSS

modeling and solution Explain the basic concepts of optimization, simulation, and heuristics, and when to use them Describe how to structure a linear programming model

Learning Objectives
Understand how search methods are used to solve

MSS models Explain the differences among algorithms, blind search, and heuristics Describe how to handle multiple goals Explain what is meant by sensitivity analysis, what-if analysis, and goal seeking Describe the key issues of model management

MSS Modeling
Current modeling issues

1) Identification of the problem and environmental analysis Environmental scanning and analysis A process that involves conducting a search for and an analysis of information in external databases and flows of information

MSS Modeling
Current modeling issues

2)

Variable identification
Influence diagram: A diagram that shows the various

types of variables in a problem (e.g., decision, independent, result) and how they are related to each other

3)

Forecastings
Predicting the future

E-commerce: CRM, RMS Example forecasting software: SAS (sas.com) Predictive analytics systems attempt to predict the most profitable customers, the worst customers, and focus on identifying products and services at

MSS Modeling
Current modeling issues 4) Multiple models

A DSS can include several models, each of which represents a different part of the decision-making problem Problem: multiple goals

5)

Model categories
Optimization of problems with few alternatives Optimization via algorithm Optimization via an analytic formula Simulation Predictive models Other models

MSS Modeling
Current modeling issues

6)
7)

Model management
MBMS

Knowledge-based modeling
DSS use mostly quantitative models

8)

Current trends
Model libraries and solution technique libraries Development and use of Web tools Multidimensional analysis (modeling)

A modeling method that involves data analysis in several dimensions

The Structure of Mathematical Models for Decision Support

The Structure of Mathematical Models for Decision Support

Uncontrollable variable (parameter)

A factor that affects the result of a decision but is not under the control of the decision maker. These variables can be internal (e.g., related to technology or to policies) or external (e.g., related to legal issues or to climate) Intermediate result variable A variable that contains the values of intermediate outcomes in mathematical models

The Structure of Mathematical Models for Decision Support

Components of decision support mathematical

models
Result (outcome) variable

A variable that expresses the result of a decision (e.g., one concerning profit), usually one of the goals of a decision-making problem Decision variable A variable of a model that can be changed and manipulated by a decision maker. The decision variables correspond to the decisions to be made, such as quantity to produce and amounts of resources to allocate

Influence Diagrams
Variables: Decision
Intermediate or uncontrollable Result or outcome (intermediate or final)

Arrows indicate type of relationship and direction of influence

Certainty

Amount in CDs

Interest earned

Sales

Uncertainty

Price

4-12

Static and Dynamic Models

Static models

Models that describe a single interval of a situation Dynamic models Models whose input data are changed over time (e.g., a five-year profit or loss projection)

Static versus Dynamic models

Static Model
Single photograph of

situation Single interval Time can be rolled forward, a photo at a time Usually repeatable Steady state Optimal operating parameters Continuous Unvarying Primary tool for process design

Dynamic model Represent changing situations Time dependent Varying conditions Generate and use trends Occurrence may not repeat

Chapter 4 - 15

Certainty, Uncertainty, and Risk

Certainty, Uncertainty, and Risk

Certainty

A condition under which it is assumed that future values are known for sure and only one result is associated with an action
Uncertainty

In expert systems, a value that cannot be determined during a consultation. Many expert systems can accommodate uncertainty; that is, they allow the user to indicate whether he or she does not know the answer

Certainty, Uncertainty, and Risk

Risk

A probabilistic or stochastic decision situation Risk analysis A decision-making method that analyzes the risk (based on assumed known probabilities) associated with different alternatives. Also known as calculated risk

Decision Structures

Uncertainty

Uncertainty

Decision

Objective

Decision

Objective

Basic risky decision

Basic risky policy

Decision Structures
Decision
Uncertainty Objective

Objective

Objective

Objective Decision

Total satisfaction

Objective Multiple objective, no-risk decision

Objective

Total satisfaction

Basic risky decision with multiple objectives

Decision Structures
Uncertainty Uncertainty Uncertainty Uncertainty Decision

Decision
Decision Decision Objective Objective Objective

Objective
Total satisfaction

Multiple period sequential decision

MSS Modeling with Spreadsheets

Models can be developed and implemented in a

variety of programming languages and systems The spreadsheet is clearly the most popular end-user modeling tool because it incorporates many powerful financial, statistical, mathematical, and other functions Spreedsheet: tools for analysis, planning and modelling
Strength and flexibility
Ease-to-use implementation Extensive statistical, forecasting Database management capabilities and functions

MSS Modeling with Spreadsheets

MSS Modeling with Spreadsheets

Other important spreadsheet features include: what-if analysis, goal seeking, data management, and programmability Most spreadsheet packages provide fairly seamless

integration because they read and write common file structures and easily interface with databases and other tools Static or dynamic models can be built in a spreadsheet

MSS Modeling with Spreadsheets

Decision Analysis with Decision Tables and Decision Trees

Decision analysis

Methods for determining the solution to a problem, typically when it is inappropriate to use iterative algorithms Decision situation that involve finite and usually not too large number of alternatives
Decision table

A table used to represent knowledge and prepare it for analysis in:

Treating uncertainty

Treating risk

Decision Analysis with Decision Tables and Decision Trees

Decision tree

A graphical presentation of a sequence of interrelated decisions to be made under assumed risk

Multiple goals

Refers to a decision situation in which alternatives are evaluated with several, sometimes conflicting, goals

Win contest
Uncertainty
Enter contest

Win large return on wager

Lose wager

Decision

Lose contest

Do not enter contest

Lose/ gain nothing

Decision Tables
Investment example One goal: maximize the yield after one year Yield depends on the status of the economy

(the state of nature)

Solid growth Stagnation Inflation Possible situation 1. 2.

3.

If solid growth in the economy, bonds yield 12%; stocks 15%; time deposits 6.5% If stagnation, bonds yield 6%; stocks 3%; time deposits 6.5% If inflation, bonds yield 3%; stocks lose 2%; time deposits yield 6.5%

Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson, 6th edition 30 Copyright 2001, Prentice Hall, Upper Saddle River, NJ

Investment Problem Decision Table Model

Mathematical Programming Optimization

Mathematical programming

A family of tools designed to help solve managerial problems in which the decision maker must allocate scarce resources among competing activities to optimize a measurable goal
Optimal solution

A best possible solution to a modeled problem

Linear programming (LP)

A mathematical model for the optimal solution of resource allocation problems. All the relationships among the variables in this type of model are linear

LP Problem Characteristics
1. Limited quantity of economic resources 2. Resources are used in the production of products or services 3. Two or more ways (solutions, programs) to use the resources 4. Each activity (product or service) yields a return in terms of the goal 5. Allocation is usually restricted by constraints

Linear Programming Steps

1. Identify the Decision variables Objective function Objective function coefficients Constraints
Capacities / Demands

Line

2. Represent the model LINDO: Write mathematical formulation EXCEL: Input data into specific cells in Excel 3. Run the model and observe the results

LP Example
The Product-Mix Linear Programming Model

MBI Corporation Decision: How many computers to build next month? Two types of mainframe computers: CC7 and CC8 Constraints: Labor limits, Materials limit, Marketing lower limits
CC7 CC8 300 500 10,000 15,000 1 1 8,000 12,000 Rel <= <= >= >= Max Limit 200,000 /mo 8,000,000 /mo 100 200

Labor (days) Materials ($) Units Units Profit ($)

Objective: Maximize Total Profit / Month

LP Solution

LP Solution

Decision Variables: X1: unit of CC-7 X2: unit of CC-8

Objective Function: Maximize Z (profit)

Z=8000X1+12000X2

Subject To 300X1 + 500X2 200K 10000X1 + 15000X2 8000K X1 100 X2 200

Multiple Goals, Sensitivity Analysis, What-If Analysis, and Goal Seeking

Multiple goals

Refers to a decision situation in which alternatives are evaluated with several, sometimes conflicting, goals
Sensitivity analysis

A study of the effect of a change in one or more input variables on a proposed solution

Multiple Goals, Sensitivity Analysis, What-If Analysis, and Goal Seeking

Sensitivity analysis tests relationships such as: The impact of changes in external (uncontrollable) variables and parameters on the outcome variable(s) The impact of changes in decision variables on the outcome variable(s) The effect of uncertainty in estimating external variables The effects of different dependent interactions among variables The robustness of decisions under changing conditions

Multiple Goals, Sensitivity Analysis, What-If Analysis, and Goal Seeking

Sensitivity analyses are used for: Revising models to eliminate too-large sensitivities Adding details about sensitive variables or scenarios Obtaining better estimates of sensitive external variables Altering a real-world system to reduce actual sensitivities Accepting and using the sensitive (and hence vulnerable) real world, leading to the continuous and close monitoring of actual results The two types of sensitivity analyses are automatic and

trial-and-error

Multiple Goals, Sensitivity Analysis, What-If Analysis, and Goal Seeking

Automatic sensitivity analysis
Automatic sensitivity analysis is performed in standard

quantitative model implementations such as LP

Trial-and-error sensitivity analysis
The impact of changes in any variable, or in several

variables, can be determined through a simple trial-anderror approach

Multiple Goals, Sensitivity Analysis, What-If Analysis, and Goal Seeking

What-If Analysis

A process that involves asking a computer what the effect of changing some of the input data or parameters would be

Multiple Goals, Sensitivity Analysis, What-If Analysis, and Goal Seeking

Multiple Goals, Sensitivity Analysis, What-If Analysis, and Goal Seeking

Goal seeking

Asking a computer what values certain variables must have in order to attain desired goals

Multiple Goals, Sensitivity Analysis, What-If Analysis, and Goal Seeking

Problem-Solving Search Methods

Problem-Solving Search Methods

Analytical techniques use mathematical formulas to

derive an optimal solution directly or to predict a certain result An algorithm is a step-by-step search process for obtaining an optimal solution

Problem-Solving Search Methods

Problem-Solving Search Methods

A goal is a description of a desired solution to a

problem The search steps are a set of possible steps leading from initial conditions to the goal Problem solving is done by searching through the possible solutions

Problem-Solving Search Methods

Blind search techniques are arbitrary search

approaches that are not guided

In a complete enumeration all the alternatives are

considered and therefore an optimal solution is discovered In an incomplete enumeration (partial search) continues until a good-enough solution is found (a form of suboptimization)

Problem-Solving Search Methods

Heuristic searching
Heuristics

Informal, judgmental knowledge of an application area that constitutes the rules of good judgment in the field. Heuristics also encompasses the knowledge of how to solve problems efficiently and effectively, how to plan steps in solving a complex problem, how to improve performance, and so forth Heuristic programming The use of heuristics in problem solving

Simulation
Simulation

An imitation of reality Major characteristics of simulation

Simulation is a technique for conducting experiments Simulation is a descriptive rather than a normative

method Simulation is normally used only when a problem is too complex to be treated using numerical optimization techniques
Complexity

A measure of how difficult a problem is in terms of its formulation for optimization, its required optimization effort, or its stochastic nature

Simulation
Advantages of simulation
The theory is fairly straightforward. A great amount of time compression can be attained A manager can experiment with different alternatives The MSS builder must constantly interact with the manager The model is built from the managers perspective. The simulation model is built for one particular problem Simulation can handle an extremely wide variety of problem types Simulation can include the real complexities of problems Simulation automatically produces many important performance measures Simulation can readily handle relatively unstructured

Simulation
Disadvantages of simulation
An optimal solution cannot be guaranteed Simulation model construction can be a slow and costly

process Solutions and inferences from a simulation study are usually not transferable to other problems Simulation is sometimes so easy to explain to managers that analytic methods are often overlooked Simulation software sometimes requires special skills

Simulation

Simulation

Methodology of simulation
1. 2. 3. 4.

5.
6. 7.

Define the problem Construct the simulation model Test and validate the model Design the experiment Conduct the experiment Evaluate the results Implement the results

Simulation

Simulation types
Probabilistic simulation

Discrete distributions Continuous distributions

Time-dependent versus time-independent simulation Object-oriented simulation Visual simulation Simulation software

Continuous versus discrete- event models

Continuous system simulation models describe

physical or economic processes in which the numbers that describe the system vary continuously. Discrete event models deal with systems in which individual events occur at identifiable points in time and change the state of the system instantaneously from one value to a different one. Socio economic planners use continuous models, because their work is not in the mainstream of corporate DSS. Since discrete event models suit most business planning needs, they are the most common type of dynamic system model found in real DSS.
Chapter 4 - 58

Visual Interactive Simulation

Conventional simulation inadequacies

Simulation reports statistical results at the end of a set of experiments Decision makers are not an integral part of simulation development and experimentation Decision makers experience and judgment cannot be used directly Confidence gap occurs if the simulation results do not match the intuition or judgment of the decision maker

Visual Interactive Simulation

Visual interactive simulation or visual interactive modeling (VIM)

A simulation approach used in the decision-making process that shows graphical animation in which systems and processes are presented dynamically to the decision maker. It enables visualization of the results of different potential actions

Visual Interactive models and DSS

Waiting-line management (queuing) is a good example of VIM The VIM approach can also be used in conjunction with artificial intelligence General-purpose commercial dynamic VIS software is readily available

Quantitative Software Packages and Model Base Management

Quantitative software packages A preprogrammed (sometimes called ready-made) model or optimization system. These packages sometimes serve as building blocks for other quantitative models

Quantitative Software Packages and Model Base Management

Model base management

Model base management system (MBMS) Software for establishing, updating, combining, and so on (e.g., managing) a DSS model base Relational model base management system (RMBMS) A relational approach (as in relational databases) to the design and development of a model base management system Object-oriented model base management system (OOMBMS) An MBMS constructed in an object-oriented environment