Decision Support and Business Intelligence Systems: (8 Ed., Prentice Hall)

Decision Support and
Business Intelligence
Systems
(8th Ed., Prentice Hall)
Chapter 4:
Modeling and Analysis
Major Modeling Issues
 problem identification and environmental analysis:
scanning the environment to figure out what problems exist and
can be solved via a model
 variable identification: identifying the critical factors in a
model and their relationships
 ex: Influence diagram : Graphical representations of a model
Rectangle = a decision variable
Circle = uncontrollable or intermediate variable
Oval = result (outcome) variable: intermediate or final
Variables are connected with arrows  indicates the direction

of influence (relationship)
-2 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 forecasting: predicting the future
 It is essential for construction models because when a decision
implemented, the results occur in the future.
 E-Commerce ( Information about purchases should be analyzed to
predict demand)
 5 Rights (How to get the right product(s) to the right customer at
the right price at the right time in the right format
 CRM and RMS rely heavily on forecasting techniques
 Predict the most profitable customers
 use of multiple models: combining them to solve many parts

of a complex problem
 Each of which represents a different part of the decision – making problem
 E.g., the Procter and Gamble supply chain DSS include:
 Location model to locate distribution centre , a product strategy model,
a demand- forecasting model, cost generation model ,….

 use of multiple models
 Types of models:
1. Standard : built in to DSS or freestanding soft ware that can
interface with a DSS
2. Nonstandard : constructed from scratch.
 model categories: selecting the right type of model for the
problem or sub-problem (table 4.1)
 model management: coordinating a firm’s models and their
use
 Models like data, must be managed to maintain their integrity and
their applicability
 Management is done by MBMS
 knowledge-based modeling: how to take advantage of
human knowledge in modeling
 DSS use mostly quantitive models, wheres Expert systems use qualitiative
Categories of Models Table 4.1
Category Objective Techniques

Optimization of Find the best solution from a Decision tables,
problems with few small number of alternatives decision trees
alternatives
Optimization via Find the best solution from a Linear and other
algorithm large number of alternatives mathematical
using a step-by-step process programming models
Optimization via an Find the best solution in one Some inventory models
analytic formula step using a formula
Simulation Find a good enough solution Several types of

by experimenting with a simulation
dynamic model of the system
Heuristics Find a good enough solution Heuristic programming
using “common-sense” rules and expert systems
Predictive and Predict future occurrences, Forecasting, Markov

other models what-if analysis, … chains, financial, …
Static and Dynamic Models
 Static Analysis
 Single snapshot of the situation, every thing occurs in a single
interval
 describes relationships among parts of a system at a point in
time.
 Ex: A decision about buy a product , Annual income
statement.
 Dynamic Analysis
 Evaluate scenarios that change over time
 Time dependent
 Ex: In determining how many checkout points should be open
in a supermarket.
 A 5 year Profit and Loss projection in which input data (costs,
prices, and quantities ) change from year to year

MSS Modeling with Spreadsheets
 Spreadsheet: most popular end-user modeling tool
 Flexible and easy to use
 Powerful functions
 Add-in functions and solvers (small programs designed to
extend the capabilities of a spreadsheet package)
 Programmability (via macros)
 What-if analysis
 Goal seeking
 Simple database management
 Incorporates both static and dynamic models
 Examples: Microsoft Excel, Lotus 1-2-3

Types of Decision Making Environments
 Decision Making under Certainty
 Decision Making under Risk (Decision
 making with probability)
 Decision Making Under Uncertainty
(Decision making without probability)

The Six Steps in Decision Theory
1. Clearly define the problem at hand
2. List all the possible alternatives (decisions to be
made)
3. Identify the possible outcomes (state of nature) of
each alternative
4. List the payoff or the profit of each combination of
alternatives and outcomes
5. Select one of the mathematical decision theory
models (e.g. Decision Making under Risk)
6. Apply the model and make your decision

Certainty, Uncertainty and Risk
The Zones of Decision Making

Decision Making:Treating Certainty
 Certainty Models
 Assume complete knowledge
 All potential outcomes are known
 May yield optimal solution
 The decision maker knows exactly what the outcome of each
course of action will be.

 decision maker is to compute the optimal alternative or
outcome with some optimization criterion in mind.

 Ex: if the optimization criterion is least cost and you are
considering two different brands of a product, which appear to

be equal in value to you, one costing 20% less than the other,
then, all other things being equal, you will choose the less
expensive brand.
 decision making under certainty is rare because all other things are rarely equal.
 Linear programming is one of the techniques for finding an optimal solution under certainty

Decision Making: Uncertainty and Risk
 Uncertainty
 Several outcomes for each decision
 Probability of each outcome is unknown
 Knowledge would lead to less uncertainty
 Decision under uncertainty is very difficult
 Managers attempt to avoid uncertainty.
 Instead they attempt to obtain more
information so it can be treated under certainty

Or
 Some estimated probabilities are assigned to
the outcomes and the decision making is done

as if it is decision making under risk.
Decision Making Under Uncertainty
 Four Criteria
 MAXIMAX - find the alternative that maximizes the maximum
outcome for every alternative (Optimistic approach ) Ex: stocks
 MAXIMIN - find the alternative that maximizes the minimum
outcomes for every alternative (Pessimistic approach ) Ex : CDs
 EQUALLY LIKELY- find the alternative with the highest
average outcome
 MINIMAX REGRET- minimizes the maximum regret (regret is
the difference between the payoff from the best decision and all
the other decision payoffs)

Decision Analysis: A Few Alternatives
Single Goal Situations
1. Decision tables :organize information and knowledge in a
systmatic ,tabular manner to prepare it for analysis
 Multiple criteria decision analysis
 Features include :
 Decision variables: describe alternatives course of
variable),
 Uncontrollable variables, Parameters : factors that
effect the result variables nut not under control of decision

maker
 Result variables: reflect intermediate outcomes in
mathematical models.

Decision Making: Risk
 Risk analysis (probabilistic decision making)
 Several outcomes for each decision
 Probability of each outcome is known
 Instead of optimizing the outcomes, the general rule is to
optimize the expected outcome.

 As an example: if you are faced with a choice between two
actions one offering a 1% probability of a gain of $10000

and the other a 50% probability of a gain of $400, you as a
rational decision maker will choose the second alternative.

Investment Example
Decision Making Under Risk
 Let us suppose that based on several economic forecasts, the
investor is able to estimate
 0.50% Solid Growth
 0.30% Stagnation
 0.20% Inflation
State of nature
Alternative Solid Growth Stagnation Inflation
50%. 30%. 20%.
Bonds 12 6 3
Stocks 15 3 2-
CDs 6.5 6.5 6.5

Investment Example
 Risk analysis
1. Compute expected values or Expected payoff (EP)
 (outcome of first state of nature)*(its prob.) + (outcome of
second state of nature)*(its prob.)+…+ (outcome of last
state of nature) * (its prob.)
 E.g. , In bonds yield = 12(.5)+6(.3)+3(.2) = 8.4 percent
 The Best decision is the one with the greatest EP
 If the payoffs were in terms of costs, the best decision would
be the one with the lowest EP

Investment Example
2. Alternative approach in decision making under risk is to
minimize expected opportunity loss (EOL).
 Opportunity loss, also called regret
 EOL for an alternative is sum of all possible regrets of
alternative, each weighted by probability of state of nature for
that regret occurring.
 EOL (alternative i ) = (regret of first state of nature) x

(probability of first state of nature) + (regret of second state of
nature) x (probability of second state of nature) + . . . +
(regret of last state of nature) x (probability of last state of
nature)

Investment Example
 EOL: Opportunity loss table (=regret table)
State of nature
Alternative Solid Growth Stagnation Inflation EOL
50%. 30%. 20%. $
Bonds 3 5. 3.5 2.35 Min
Stocks 0 3.5 8.5 2.75

CDs 8.5 0 0 4.25

Investment Example
3. The Maximum Likelihood Criterion
 Identify the state of nature with the largest Probability.
 2. Choose the decisions alternative that has the largest
Payoff
State of nature
50%. 30%. 20%.
Bonds 12 6 3
Stocks 15 3 2-
CDs 6.5 6.5 6.5

Investment Example
 Expected Value of Perfect Information (EVPI)
 Is used to place an upper limit on what you should pay for
information that will aid in making a better decision.
 Is the increase in the EP that could be obtained if it were possible
to learn the true state of nature before making the decision
 Is the difference between the expected value under certainty and
the expected value under risk

Investment Example
 EVPI = A – B
 A = expected value with perfect information
 B = expected value without perfect information
 For A: The optimal values for each value are:
 Max Value (A)= 15*.5 +6.5*.3 +6.5*.2 =10.75
State of nature
50%. 30%. 20%.
Bonds 12 6 3
Stocks 15 3 2-
CDs 6.5 6.5 6.5

Investment Example
 B = expected value without perfect information
 For B: we compute the expected values for each column
first, and then select the max as below:
EVPI = 10.75-8.4 = 2.55 $

Decision Analysis: A Few Alternatives
Single Goal Situations
 Decision trees
 Graphical representation of relationships
 Multiple criteria approach
 Demonstrates complex relationships
 Cumbersome, if many alternatives exists
How can a decision tree be used

in decision making?
By showing the decision maker
the possible outcomes that
could result from a given choice,
the tree gives the decision
maker information by which to
compare choices

Decision trees
 The Five Steps
1. Define the problem
2. Structure or draw the decision tree
3. Assign probabilities to the states of nature
4. Estimate the payoffs for each possible combination of
alternative and state of nature  Solve the problem by
computing expected payoff (EP) for each state of nature
node
5. Make your decision

Decision Tree Example (Self Study)
 This is just a beginning of ADM2302 course and Andrew does not know if he
should attend all classes. He consulted some other students and came to the
following conclusions:
• chances of passing a course while attending all classes are 80%
• chances of passing a course while attending randomly are 50%.
It is well known that professor who is teaching that course is giving second
chance to the students who failed. They have to solve a pretty nasty case
study.
Again, Andrew estimates that chances of solving this case if he would go to all
the classes are 60%, while they drop to just 10% if he would attend classes
randomly.
Andrew would be very happy if he passes the course (5 on a happiness scale of
0 - 5). Clearly, he would be very disappointed if he fails (0 on a happiness
scale).
Going to a classroom requires an effort and diminished happiness associated
with passing the course.
It goes down by 3 points (happiness scale) for attending all classes and 1 point
for 39 random attendance.
Optimization
via Mathematical Programming
 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: The best possible solution to a
modeled problem
 Linear programming (LP): A mathematical model
for the optimal solution of resource allocation
problems. All the relationships are linear
 Limited quantity of economic resources

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

Sensitivity, What-if, and
Goal Seeking Analysis
 Sensitivity
 Assesses impact of change in inputs on outputs
 Eliminates or reduces variables
 Can be automatic or trial and error
1. Automatic Sensitivity analysis is performed in standard

quantitative model implementation such as LP
2. Trial and Error
 Change in any variable or in several
 Two approaches: What-if and Goal Seeking

 What-if
 Assesses solutions based on changes in variables or
assumptions (scenario analysis)

 Ex: What will happen to the total inventory cost if the cost
of carrying inventories increases by 10 percent?

 Goal seeking
 Backwards approach, starts with goal
 Determines values of inputs needed to achieve goal
 Ex: what annual R&D budget is needed for an annual
growth rate of 15 percent by 2014?

 Computing a Break-Even Point Using Goal Seeking
 Values that generate zero profit
 Break –Even= Fixed cost / (selling cost – variable cost)

 Where : fixed cost =cost that not change such as tax,
insurance ,..
 Selling price: the price that a unit sold for
 Variable cost : related to production unit.

Problem –Solving Search Methods
 Search methods used in the choice phase of problem
solving includes:
 Analytical techniques , algorithms , blind searching and heuristic
searching
 For normative models (Comparing all the outcomes of alternative) :

analytical approach is used
 For descriptive models(a comparison of a limited number of
alternatives is used) :
 blindly or heuristic s are used.

 Analytical Techniques
 Use mathematical formulas to derive optimal solution
 Solving structured problems (tactical or operational)
 Ex: inventory management, resource allocation.
 Analytical Techniques may use Algorithms
 Algorithms
 Step by step search process
 Obtaining an optimal solution
Web search engines use algorithms To
speed searches and produces accurate
results

 Blind Searching
 Problem solving is done by searching through the possible solutions
 The first search methods of problem solving
 Arbitrary search approaches that are not guided
 Two types:
 A complete enumeration : all alternatives are considered to find an optimal
solution.
 Incomplete (Partial) : continues until a good –enough solution is found
 Heuristic Searching
 Informal judgmental knowledge of an application area that
constitute the rules of the good judgment in the field.

Simulation
 Simulation is a process of designing a model
of real system a model of real system
 purpose of understanding the behavior for
the operation of the behavior for the
operation of the system.
 Frequently used in DSS tools

Simulation
 System:
 Collection of entities, ex: people machines that act and interact towards the
accomplishment.
 State:
 Collection of variables necessary to describe a system at a particular time
relative to the objective of study
 Bank model: Could include number of busy tellers, time of arrival of each
customer, etc
 System can be
1. Discrete
 State variables change instantaneously at separated points in
time
 Bank model: State changes occur only when a customer
arrives or departs

2. Continuous
 State variables change that continuously tracks system
response over time
 Airplane flight: State variables like position, velocity change
continuously

Major Characteristics of Simulation
 Imitates reality and capture its richness
 Technique for conducting experiments
 Descriptive, not normative tool
 Often to “solve” very complex problems
! Simulation is normally used only when a

problem is too complex to be treated using
numerical optimization techniques

Advantages of Simulation
 The theory is fairly straightforward
 Great deal of time compression
 Experiment with different alternatives
 The model reflects manager’s perspective
 Can handle wide variety of problem types
 Can include the real complexities of problems
 Produces important performance measures
 Often it is the only DSS modeling tool for
non-structured problems
Limitations of Simulation
 Cannot guarantee an optimal solution
 Slow and costly construction process
 Cannot transfer solutions and inferences to
solve other problems (problem specific)
 Software may require special skills

Simulation Methodology
 Model real system and conduct repetitive experiments.
 Steps:
1. Define problem 5. Conduct experiments
2. Construct simulation model 6. Evaluate results
3. Test and validate model 7. Implement solution
4. Design experiments

Visual Interactive Modeling (VIM) /
Visual Interactive Simulation (VIS)
 Visual interactive modeling (VIM)
Also called
 Visual interactive problem solving
 Visual interactive modeling
 Visual interactive simulation
 Uses computer graphics to present the impact
of different management decisions
 Often integrated with GIS
 Users perform sensitivity analysis
 Static or a dynamic (animation) systems

Decision Support and Business Intelligence Systems: (8 Ed., Prentice Hall)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Decision Support and Business Intelligence Systems: (8 Ed., Prentice Hall)

Uploaded by

Copyright:

Available Formats

Decision Support and

Variables are connected with arrows  indicates the direction

-2 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 use of multiple models: combining them to solve many parts

-3 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

Category Objective Techniques

Simulation Find a good enough solution Several types of

Predictive and Predict future occurrences, Forecasting, Markov

 Ex: In determining how many checkout points should be open

prices, and quantities ) change from year to year

-7 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-8 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-9 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

The Zones of Decision Making

-10 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 All potential outcomes are known

 May yield optimal solution

 The decision maker knows exactly what the outcome of each

course of action will be.

outcome with some optimization criterion in mind.

considering two different brands of a product, which appear to

-11 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 Probability of each outcome is unknown

 Knowledge would lead to less uncertainty

 Decision under uncertainty is very difficult

 Managers attempt to avoid uncertainty.

 Instead they attempt to obtain more

information so it can be treated under certainty

the outcomes and the decision making is done

-13 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 Decision variables: describe alternatives course of

effect the result variables nut not under control of decision

-14 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 Probability of each outcome is known

 Instead of optimizing the outcomes, the general rule is to

optimize the expected outcome.

actions one offering a 1% probability of a gain of $10000

-15 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-16 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-17 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 EOL (alternative i ) = (regret of first state of nature) x

-18 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

Stocks 0 3.5 8.5 2.75

-19 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-20 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-21 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-22 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

EVPI = 10.75-8.4 = 2.55 $

-23 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

How can a decision tree be used

-24 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-25 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 2. Represent the model

 3. Run the model and observe the results

 Eliminates or reduces variables

 Can be automatic or trial and error

1. Automatic Sensitivity analysis is performed in standard

-29 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

assumptions (scenario analysis)

of carrying inventories increases by 10 percent?

-30 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

-31 Copyright © 2011 Pearson Education, Inc. Publishing as Prentice Hall

 Break –Even= Fixed cost / (selling cost – variable cost)