Analytics, Data Science and A I:

Systems for Decision Support

Eleventh Edition

Chapter 8
Prescriptive Analytics: Optimization
and Simulation

Slide in this Presentation Contain Hyperlinks.

JAWS users should be able to get a list of links by
using INSERT+F77

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Learning Objectives (1 of 2)
8.1 Understand the applications of prescriptive analytics
techniques in combination with reporting and predictive
analytics
8.2 Understand the basic concepts of analytical decision
modeling
8.3 Understand the concepts of analytical models for
selected decision problems, including linear
programming and simulation models for decision
support
8.4 Describe how spreadsheets can be used for analytical
modeling and solutions

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Learning Objectives (2 of 2)
8.5 Explain the basic concepts of optimization and
when to use them
8.6 Describe how to structure a linear programming model
8.7 Explain what is meant by sensitivity analysis, what-if
analysis, and goal seeking
8.8 Understand the concepts and applications of different
types of simulation
8.9 Understand potential applications of discrete
event simulation

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Opening Vignette (1 of 2)
School District of Philadelphia Uses
Prescriptive Analytics to Find Optimal Solution
for Awarding Bus Route Contracts

• Background and the problem

• The solution…
• The results…

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Opening Vignette (2 of 2)
Questions for the Opening Vignette:
1. What decision was being made in this vignette?
2. What data (descriptive and or predictive) might one need
to make the best allocations in this scenario?
3. What other costs or constraints might you have to
consider in awarding contracts for such routes?
4. Which other situations might be appropriate for
applications of such models?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.1
Canadian Football League Optimizes Game
Schedule
Questions for Discussion:
1. List three ways in which Solver-based scheduling of
games could result in more revenue as compared to the
manual scheduling.
2. In what other ways can CF L leverage the Solver
software to expand and enhance their other business
operations?
3. What other considerations could be important in
scheduling such games?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Model Based Decision Making (2 of 2)
• Prescriptive Analytics
– Use models/representations of real-world… to make decision on
real-world situations
– To determine what to do next!
• Prescriptive analytics model examples
– Mathematical models for decision recommendation
• Example:
– deciding which customers are likely to buy from us and making
an offer or giving a price point that will maximize the likelihood
that they would buy and our profit would be optimized.
– Conversely, it might involve being able to predict which customer
is likely to go somewhere else and making a promotion offer to
retain them as a customer and optimize our value.
• Maximize response rate within budget
Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Model Based Decision Making – Major Model Issues

• Identification of the problem and Environment Analysis

– Environmental scanning and analysis
 Monitoring, scanning, and interpretation of collected
information
 Analyze the scope of the domain and the forces
and dynamics of the environment (environment
caused the problem?)
 BI can facilitate to build models in a general form 
every stakeholder can understand
– Variable identification
 Decision, result, uncontrollable variables and their
relationship
 Influence diagrams can facilitate
Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Model Based Decision Making – Major Model Issues

• Forecasting (predictive analytics)

– As part of prescriptive analytics (pre-requisite)
– What is likely to happen in the future based on what
happened in the past
– Influencing the decision made for the future

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
 4-10

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

Arrows indicate type of relationship and direction of influence

Certainty Amount Interest

in CDs earned

Sales
Uncertainty Price

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
 4-11

Influence Diagrams

~
Random (risk) Demand
Place tilde above Sales
variable’s name
Sleep all
day
Graduate Get job
Preference University
(double line arrow)
Ski all
day

Arrows can be one-way or bidirectional, based upon the

direction of influence

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
 4-12

Example
Consider the following profit model:
Profit = income - expenses
Income = units sold * unit price
Units sold = 0.5 * amount used in advertisement
Expenses = unit cost * units sold + fixed cost

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
INFLUENCE DIAGRAMS
 Once a decision-making problem is understood and defined,
it must be analyzed
 Graphical representation of a model used to assist in model
design, development, and understanding.
 An influence diagram (ID) (also called a relevance diagram,
decision diagram or a decision network) is a compact
graphical and mathematical representation of a decision
situation.
 provides visual communication to the model builder or
development team
 Serves as a framework for expressing the exact nature of
the relationships of the MSS model and to focus on the most
important ones
 Influence refers to the dependency of a variable on the level
of another variable
 Any level of detail
4-
 Shows impact of change 13
SOFTWARE FOR INFLUENCE
DIAGRAMS
 Analytica
 DecisionPro
 DATA and DataPro
 iDecide
 PrecisionTree
 …......

4-
14
Application Case 8.2
Ingram Micro Uses Business Intelligence
Applications to Make Pricing Decisions
Questions for Discussion:
1. What were the main challenges faced by Ingram Micro in
developing a BI C?
2. List all the business intelligence solutions developed by
Ingram to optimize the prices of their products and to
profile their customers.
3. What benefits did Ingram receive after using the newly
developed B I applications?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Model Categories
• Static versus dynamic models
– Time dependent / time independent
• Model management
– to maintain their integrity, and thus their applicability.
– with the aid of model-base management systems
• Knowledge-based modeling
• Current trends in modeling
– Model libraries
– Cloud-based modeling tools/platforms
– Model transparency / multi-dimensional modeling
– Influence diagrams for better modeling
Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
 STATIC AND DYNAMIC
MODELS
 Static Analysis
 Single snapshot of a situation
 Everything occurs in a single interval
 Usually repeatable, occurs in a fixed timeframe
 Eg: quarterly account statement
 Assumes stability of the relevant data
 Eg:
 process simulation begins with steady-state, which models a
static representation of a plant to find its optimal operating
parameters.
 A static representation assumes that the flow of materials into
the plant will be continuous and unvarying.
 Steady-state simulation is the main tool for process design,
when engineers must determine the best trade-off between
capital costs, operational costs, process performance, product
quality, environmental and safety factors.
4-
17
 STATIC AND DYNAMIC MODELS

 Dynamic Analysis
 Represents scenarios that change over time
 Eg: 5-year profit-and-loss projection in which the input data,
such as costs, prices, and quantities, change from year to
year.
 Time-dependent
 Eg: in determining how many checkout points should be open
in a supermarket, one must take the time of day into
consideration
 Demands must be forecasted over time.
 Dynamic simulation
 represents what happens when conditions vary from the
steady state over time
 they use, represent, or generate trends and patterns over
time
 Used to handle problems that are dynamic nature
4-
18
DSS MODELS
 Algorithm-based models
 Statistic-based models
 Linear programming models
 Graphical models
 Quantitative models
 Qualitative models
 Simulation models

4-
19
Categories of Models
Table 8.1 Categories of Models.
Category Process and Objective Representative Techniques
Optimization of problems Find the best solution from a small Decision tables, decision trees,
with few alternatives number of alternatives analytic hierarchy process
Optimization via Find the best solution from a large Linear and other mathematical
algorithm number of alternatives, using a programming models, network
step-by-step improvement process models
Optimization via an Find the best solution in one step, Some inventory models
analytic formula using a formula
Simulation Find a good enough solution or the Several types of simulation
best among the alternatives
checked, using experimentation
Heuristics Find a good enough solution, using Heuristic programming, expert
rules systems
Predictive models Predict the future for a given Forecasting models, Markov
scenario analysis
Other models Solve a what-if case, using a Financial modeling, waiting
formula lines

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Structure of Mathematical Models for Decision
Support
• Quantitative Models: Mathematically links decision
variables, uncontrollable variables, and result variables
– Non-quantitative models: qualitative models

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Structure of Mathematical Models for
Decision Support (1 of 2)
• Result (outcome variable)
– reflect the level of effectiveness of a system; that is, they indicate
how well the system performs or attains its goal(s).
• Decision variables
– describe alternative courses of action
– Under the control of decision maker
• Uncontrollable variable (or parameters)
– factors that affect the result variables
– not under the control of the decision maker
• Intermediate result variables
– reflect intermediate outcomes in mathematical models

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Examples of Components of Models
Table 8.2 Examples of Components of Models.
Uncontrollable Variables
Area Decision Variables Result Variables
and Parameters
Financial Investment alternatives Total profit, risk Rate of return Inflation rate
investment and amounts on investment (RO I) Prime rate
Earnings per share Competition
Liquidity level
Marketing Advertising budget Market share Customer’s income
Where to advertise Customer satisfaction Competitor’s actions
Manufacturing What and how much to Total cost Machine capacity
produce
Inventory levels Quality level Technology
Compensation programs Employee satisfaction Materials prices
Accounting Use of computers Data processing cost Computer technology
Audit schedule Error rate Tax rates
Legal requirements
Transportation Shipments schedule use Total transport cost Delivery distance
of smart cards Payment float time Regulations
Services Staffing levels Customer satisfaction Demand for services

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
The Structure of a Mathematical
Model
• The components of a quantitative model are linked
together by mathematical (algebraic) expressions—
equations or inequalities.
– Example: Profit
where P = profit, R = revenue, and C = cost
– Example - Simple Present-Value -
F 100, 000
P n
 5
62, 092
(1  i ) (1  0.1)
where P = present value, F = future cash-flow,
i = interest-rate, and n = number of period (years)

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Modeling and Decision Making -
Under Certainty, Uncertainty, and Risk (1 of 2)

• Decision situations are often classified on the basis of what the

decision-maker knows (or believes) about the forecasted results
• Certainty
– Assume complete knowledge
– All potential outcomes are known
• Uncertainty
– Several outcomes for each decision
– Probability of each outcome is unknown
• Risk analysis (probabilistic decision making)
– Probability of each outcomes is known
– Level of uncertainty  Risk (expected value)

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Modeling and Decision Making -
Under Certainty, Uncertainty, and Risk (2 of 2)

The Zones of Decision Making

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
CERTAINTY, UNCERTAINTY,
AND RISK
 Decision-making under Certainty

 Assumes that complete knowledge is available

 Assumes All potential outcomes known (may not be
100%)
 This assumption simplifies the model and makes it
tractable
 Easy to develop and solve
 occurs most often with structured problems and short
time horizons (up to 1 year)
 Eg: the need to meet customer, contract or regulatory
requirements. The outcomes (consequences) are
known to you, should you fail to comply
 Many financial models are constructed under assumed 4-
certainty, even though the market is anything but 27
CERTAINTY, UNCERTAINTY,
AND RISK
 Decision-making under Risk

 Probabilistic or stochastic Decision-Making

situation
 Decision under risk
 Several possible outcomes
 Probability of each of possible outcomes are
known/estimated
 Assess the risk (calculated risk)
 Risk analysis
 is a decision-making method that analyzes the risk
(based on assumed known probabilities) associated
with different alternatives.
 Calculate expected value of each alternative 4-
28
 Select best expected value
CERTAINTY, UNCERTAINTY,
AND RISK
 Decision-making under Uncertainty

 situations in which several outcomes are possible for

each course of action.
 the decision maker does not know, or cannot estimate,
the probability of occurrence of the possible outcomes.
 Difficult, because of nsufficient information.
 Modeling of such situations involves assessment of the
decision maker’s (or the organization’s) attitude toward
risk.
 Managers attempt to avoid uncertainty as much as
possible
 they attempt to obtain more information so that the
problem can be treated under certainty (because it can
be “almost” certain) or under calculated (i.e., 4-
29
assumed) risk.
 Application Case 8.3
American Airlines Uses Should-Cost Modeling to Assess the Uncertainty of Bids for Shipment Routes

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
 Application Case 8.3
American Airlines Uses Should-Cost Modeling to Assess the Uncertainty of Bids for Shipment Routes

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.3
American Airlines Uses Should-Cost Modeling
to Assess the Uncertainty of Bids for Shipment
Routes
Questions for Discussion:
1. Besides reducing the risk of overpaying or underpaying
suppliers, what are some other benefits A A would derive from
its “should-be” model?
2. Can you think of other domains besides air transportation
where such a model could be used?
3. Discuss other possible methods with which A A could have
solved its bid overpayment and underpayment problem.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Decision Modeling with
Spreadsheets
• Spreadsheet
– Most popular end-user modeling tool
– Flexible and easy to use
– Powerful functions (add-in functions)
– Programmability (via macros)
– What-if analysis and goal seeking
– Simple database management
– Seamless integration of model and data
– Incorporates both static and dynamic models
– Examples: Microsoft Excel (with Solver add-in)

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.4
Pennsylvania Adoption Exchange Uses
Spreadsheet Model to Better Match Children
with Families
Questions for Discussion:
1. What were the challenges faced by PA E while making
adoption matching decisions?
2. What features of the new spreadsheet tool helped PA E
solve their issues of matching a family with a child?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.5
Metro Meals on Wheels Treasure Valley Uses
Excel to Find Optimal Delivery Routes

Questions for Discussion:

1. What were the challenges faced by Metro Meals on
Wheels Treasure Valley related to meal delivery before
adoption of the spreadsheet-based tool?
2. Explain the design of the spreadsheet-based model.
3. What are the intangible benefits of using the Excel-
based model to Metro Meals on Wheels?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Excel Spreadsheet - Static Model
Example
(Simple loan calculation of monthly payments)

F P(1  i ) n
 i (1  i )n 
A P  n 
 (1  i )  1 

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Excel Spreadsheet - Dynamic Model
Example
(Simple loan calculation – effect of prepayment)

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
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: The best possible solution to a modeled
problem
– Linear programming (L P): A mathematical model for
the optimal solution of resource allocation problems.
All the relationships are linear.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.6
Mixed-Integer Programming Model Helps the
University of Tennessee Medical Center with
Scheduling Physicians
Questions for Discussion:
1. What was the issue faced by the Regional Neonatal
Associates group?
2. How did the HPS M model solve all of the physician’s
requirements?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
L P 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

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Linear Programming (L P) Steps
1. Identify the …
– Decision variables
– Objective function
– Objective function coefficients
– Constraints
 Capacities / Demands / …
2. Represent the model
– LI ND O: Write mathematical formulation
– EXCE L: Input data into specific cells in Excel
3. Run the model and observe the results

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
L P Modeling – Example (1 of 3)
The Product-Mix Linear Programming Model
• MB I Corporation
• Decision variable: How many computers to build next month?
• Two types of mainframe computers: C C-7 and C C-8
• Constraints: Labor limits, Materials limit, Marketing lower limits
CC-7 CC-8 Rel Limit
Labor (days) 300 500 <= 200,000 /mo
Materials ($) 10,000 15,000 <= 8,000,000 /mo
Units 1 >= 100
Units 1 >= 200
Profit ($) 8,000 12,000 Max

Objective: Maximize Total Profit / Month

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
L P Modeling – Example (2 of 3)

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
L P Modeling – Example (3 of 3)
Figure 8.6 Excel Solver Solution to the Product-Mix
Example.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Illustrating the Power of Spreadsheet
Modeling
• Election Resource Allocation Problem

Analysis of “swing states” for the 2012 election…

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Common Optimization Models
• Assignment (best matching of objects)
• Dynamic programming
• Goal programming
• Investment (maximizing rate of return)
• Linear and integer programming
• Network models for planning and scheduling
• Nonlinear programming
• Replacement (capital budgeting)
• Simple inventory models (e.g., economic order quantity)
• Transportation (minimize cost of shipments)
Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
8.7 Multiple Goals, Sensitivity Analysis, What-If
Analysis, and Goal Seeking (1 of 6)
• Multiple Goals
– Simple-goal vs. multiple goals
– Vast majority of managerial problems has multiple goals (objectives) to
achieve
– Simultaneous, often conflicting goals sought by management

• Eg: consider a profit-making firm.

– In addition to earning money, the company wants to grow, develop its
products and employees, provide job security to its workers, and serve
the community.
– Managers want to satisfy the shareholders and at the same time enjoy
high salaries and expense accounts, and employees want to increase
their take-home pay and benefits.
– When a decision is to be made, say, about an investment project, some
of these goals complement each other, whereas others conflict.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
MULTIPLE GOALS
 Many quantitative models of decision theory are based on
comparing a single measure of effectiveness, generally
some form of utility to the decision maker.
 Determining single measure of effectiveness is difficult
 transform a multiple-goal problem into a single measure-of-
effectiveness problem before comparing the effects of the
solutions.
 Handling methods:
 Utility theory
 Goal programming
 Linear programming with goals as constraints
 A point system

4-
48
 MULTIPLE GOALS
 Certain difficulties may arise when analyzing multiple goals:
 It is usually hard to obtain an explicit statement of the
organization's goals.
 The decision-maker may change the importance assigned to
specific goals over time or for different decision scenarios.
 Goals and subgoals are viewed differently at various levels of the
organization and within different departments.
 Goals change in response to changes in the organization and its
environment.
 The relationship between alternatives and their role in determining
goals may be difficult to quantify.
 Complex problems are solved by groups of decision-makers, each
of whom has a personal agenda.
 Participants assess the importance (priorities) of the various goals
differently.

4-
49
 SENSITIVITY ANALYSIS

 attempts to assess the impact of a change in the input data

or parameters on the proposed solution (the result variable).
 Allows for adaptability and flexibility to changing conditions and
requirements of different situations
 Better understanding of the model
 Allows to input data

 Tests such relationships as

 The impact of changes in external (uncontrollable) variables and
parameters on outcome variable(s)
 The impact of changes in decision variables on 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.
4-
50
 SENSITIVITY ANALYSIS

 Used for

 Revising models to eliminate too large sensitivities

 Adding details about sensitive variables or scenarios
 Obtaining better estimates of sensitive external
variables
 Altering the 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

4-
51
SENSITIVITY ANALYSIS
 Two types
 Automatic sensitivity analysis
 performed in standard quantitative model implementations such as linear
programming.
 For example, it reports the range within which a certain input variable or
parameter value (e.g., unit cost) can vary without making any significant
impact on the proposed solution.
 Automatic sensitivity analysis is usually limited to one change at a time,
and only for certain variables.
 However, it is very powerful because of its ability to establish ranges and
limits very fast (and with little or no additional computational effort).

 Trial and Error

 The impact of changes in any variable, or in several variables, can be
determined
 You change some input data and solve the problem again.
 When the changes are repeated several times, better and better solutions
may be discovered.
 Such experimentation, which is easy to conduct when using appropriate
modeling software like Excel, has two approaches:
 what-if analysis and
 goal seeking. 4-
52
WHAT-IF ANALYSIS

 What will happen to the solution if an input

variable, an assumption, or a parameter value
is changed?
 Examples:
 What will happen to the total inventory cost if
the cost of carrying inventories increases by 10
percent?
 What will be the market share if the advertising
budget increases by 5 percent?

4-
53
GOAL SEEKING ANALYSIS

 calculates the values of the inputs necessary to

achieve a desired level of an output (goal).
 represents a backward solution approach.
 Examples:
 What annual R&D budget is needed for an annual
growth rate of 15 percent by 2025?
 How many nurses are needed to reduce the
average waiting time of a patient in the
emergency room to less than 10 minutes?
 Can compute break-even points (application of goal
seeking)
4-
54
Multiple Goals, Sensitivity Analysis,
What-If Analysis, and Goal Seeking
(2 of 6)
• Certain difficulties may arise when analyzing multiple
goals
– Difficult to obtain a single organizational goal
– The importance of goals change over time
– Goals and sub-goals are viewed differently
– Goals change in response to other changes
– Dynamics of groups of decision makers
– Assessing the importance (priorities)
– …

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Multiple Goals, Sensitivity Analysis,
What-If Analysis, and Goal Seeking
(3 of 6)
• Sensitivity analysis
– It is the process of assessing the impact of change in
inputs on outputs
– Helps to …
 eliminate (or reduce) variables
 revise models to eliminate too-large sensitivities
 adding details about sensitive variables or scenarios
 obtain better estimates of sensitive variables
 alter a real-world system to reduce sensitivities
 …
– Can be automatic or “trial and error”
Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Multiple Goals, Sensitivity Analysis,
What-If Analysis, and Goal Seeking
(4 of 6)
• What-if analysis
– Assesses solutions based on changes in variables or
assumptions (scenario analysis)
– What if we change our capacity at the milling station by
40% [what would be the impact]
• Goal seeking
– Backwards approach, starts with the goal and determines
values of inputs needed
– Example is break-even point determination
 In-order to break even (profit = 0), how many products
do we have to sell each month
Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Multiple Goals, Sensitivity Analysis,
What-If Analysis, and Goal Seeking
(5 of 6)
Figure 8.10 Example of a What-If Analysis Done in an
Excel Worksheet.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Multiple Goals, Sensitivity Analysis,
What-If Analysis, and Goal Seeking
(6 of 6)
Figure 8.11 Goal-Seeking Analysis.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Decision Analysis with Decision
Tables and Decision Trees
• Decision Tables – a tabular representation of the decision
situation (alternatives)
• Investment Example
– Goal: maximize the yield after one year
– Yield depends on the status of the economy (the state
of nature)
 Solid growth
 Stagnation
 Inflation

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Decision Table: Investment Example
(1 of 2)
1. If solid growth in the economy, bonds yield 12%; stocks
15%; time deposits 6.5%
2. If stagnation, bonds yield 6%; stocks 3%; time deposits
6.5%
3. If inflation, bonds yield 3%; stocks lose 2%; time
deposits yield 6.5%

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Decision Table: Investment Example
(2 of 2)
• Payoff decision variables (alternatives)
• Uncontrollable variables (states of economy)
• Result variables (projected yield)
• Tabular representation:

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Decision Table: Investment Example
(Treating Uncertainty)
• Optimistic approach
• Pessimistic approach
• Treating Risk/Uncertainty:
– Use known probabilities
– Expected values

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Decision Table: Investment Example
(Multiple Goals)
• Multiple goals
– Yield, safety, and liquidity

Alternative Yield (%) Safety Liquidity

Bonds 8.4 High High
Stocks 8.0 Low High
CDs 6.5 Very high High

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Decision Trees
• Graphical representation of relationships
• Multiple criteria approach
• Demonstrates complex relationships
• Cumbersome, if many alternatives exists
• Tools include
– Mind Tools Ltd., mindtools.com
– TreeAge Software Inc., treeage.com
– Palisade Corp., palisade.com

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
 Introduction to
Computer Simulations

66
 Computer Simulations
Computer simulation is the process of making a
computer behave the same as ...whatever it is we
are interested in.

• Atoms
• Cooling metal alloy
• A society of voters
• Climate change
• A galaxy

67
 Computer Simulations
Simulations have applications across a range of
disciplines:

Physics – solids, gases, fluids, solar systems

Chemistry – molecular dynamics
Biology – gene networks, predator-prey populations
Sociology – socio networks, opinion propagation
Technology – internet traffic, local networks
Management – queuing, workflow models
Finance & Economics – stock markets, supply-demand

68
 Computer Simulations
Computer simulations allow us to observe the
behaviour of these systems at (relatively) low
cost.

Other methods of investigating these systems may

involve complicated theoretical research or
experimental research with potentially expensive
equipment.

69
 Computer Simulations
There are some definitions of simulations:

"The representation of the dynamic behaviour of the system by

moving it from state to state in accordance with well-defined
operating rules." – A. Alan B. Pritsker (1984)

"We can therefore define simulation as the technique of solving

problems by the observation of the performance, over the time, of a
dynamic model of the system." – Bernard P. Zeigler (1976)

"A simulation is a method for implementing a model." – Defense

Acquisition University

70
 Computer Simulations
To create a computer simulation to approximate a
system, a model of that system must first be
made. These are most often mathematical models.

"A model is a description of some system intended to predict what

happens if certain actions are taken"
– Bratley, Bennet & Schrage (1987)

71
 Models
Modelling is a large discipline in itself and
creating a system requires a lot of mathematical
ability and understanding of the system.

Models are usually composed of variables and

relationships between them. Exactly what these
variables represent and what the relationships
between them are can vary.

72
 Models
The variables of the model must represent the
state of the system. The state is split into different
components to represent the different parts of the
system. These are sometimes called model
components.

For example:
A car in a traffic simulator may have a position, a
size and a velocity.

73
 Models
The relationships between these model
components define the behaviour of the system.
Going back to our previous example some rules
may be:

• The position of the car changes based on the velocity.

• If the distance to the car in front is less than X, decelerate.

74
 Models
The types of relationships depend on the type of
model. Some categories of model include:

Linear vs Nonlinear
Static vs Dynamic
Explicit vs Implicit
Discrete vs Continuous
Deterministic vs Probabilistic

75
 Models
Models are also limited in the accuracy with
which they describe the model. The usefulness of
a model depends on a number of factors

• model validity
• level of simplification
• credibility
• tractability

76
 Models
When using computer simulations, it is important
to understand the limitations of the model you
are using. A simulation (no matter how accurate)
cannot provide useful results if the model is not
suitable for the system you are studying.

A model is considered valid if the system it

describes sufficiently near to the real system.

77
 Models
For example:

Newton's Laws of Motion are a perfectly suitable

model for describing the behaviour of object in
our natural environment.

However, they are not sufficient to describe

objects that are travelling extremely fast or
extremely large – like Mercury for instance.

78
 Simulations
Models are approximations of a real system (for
the most part). Simulations approximate the
behaviour of the system described by the model.

For example:
An object travelling according to Newton's Laws
of Motion.

79
80
81
82
 Simulations
This type of model describes a system that
continuously changes with time. Computers don't
do things continuously so we resort to using a
time-step to jumping forward in time repeatedly.

This is an approximation of the model because we

are simulating a continuous system with discrete
time-steps.

83
 Continuous Simulations
Continuous simulations (like our example) are
simulations that compute models that change
continuously (usually over time). This type of
simulation is extremely common in the physical
sciences.

Continuous simulation and discrete event

simulation are usually considered the two main
categories of simulation.

84
 Continuous Simulations
Continuous simulations are most often based on
models described by ordinary differential
equations (ODEs) or partial differential equations
(PDEs).

Because computers are limited to discrete

calculations, these simulations update the system
with discrete time-steps.

85
 Continuous Simulations
To do this, the continuous model must be
integrated over that time-step to calculate the
total change in each of the variables representing
our model.

There are a number of different numerical

methods that can be used for this purpose.
(more on this later).

86
 Discrete Event Simulations
The other main category of simulations are
discrete event simulations. This type of
simulation is no longer computing a continuously
changing system but one that considers discrete
events.

For example – a stock exchange

87
 Discrete Event Simulations
A stock exchange could be modelling by a set of
agents (traders) that perform discrete events
(buy and sell).

There is still a time component to this simulation

but there are no variables in the state of the
system that need to be continuously modelled
over time.

88
 Discrete Event Simulations
This type of model is generally easier to simulate
on a computer because they consist of a set of
discrete events and changes.

Computers are good a doing discrete calculations.

89
 Summary

Models describe systems that are approximations

of real systems.

Simulations are programs that approximate

systems described by models.

Modeling and Simulation: The Computer Science of Illusion

- Stanislaw Raczynski, Wiley (2006)

90
8.9 Simulation
• Simulation is the “appearance” of reality
• It is often used to conduct what-if analysis on the model of
the actual system
• It is a popular DS S technique for conducting experiments
with a computer on a comprehensive model of the system
to assess its dynamic behavior
• Often used when the system is too complex for other DS S
techniques

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.7
Steel Tubing Manufacturer Uses a Simulation-
Based Production Scheduling System
Questions for Discussion:
1. Explain the advantages of using Simio’s simulation model
over traditional methods.
2. In what ways has the predictive analysis approach helped
management achieve the goals of analyzing the production
schedules?
3. Besides the steel manufacturing industry, in what other
industries could such a modeling approach help improve
quality and service?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Major Characteristics of Simulation
• Imitates reality and captures its richness both in shape
and behavior
• “Represent” versus “Imitate”
• Technique for conducting experiments
• Descriptive, not normative tool
• Often to “solve” [i.e., analyze] very complex
systems/problems
• Simulation should be used only when a numerical
optimization is not possible

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Advantages of Simulation
• The theory is fairly straightforward
• Simpler solutions to complex problems
• 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
• …

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Disadvantages of Simulation
• Cannot guarantee an optimal solution
• Slow and costly construction process
• Cannot transfer solutions and inferences to solve other
problems (problem specific)
• So easy to explain/sell to managers, may lead to
overlooking analytical solutions
• Software may require special skills
• Invalid model may result in confidence in wrong results

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Simulation Methodology
Steps:
1. Define problem 5. Conduct experiments
2. Construct the model 6. Evaluate results
3. Test and validate model 7. Implement solution
4. Design experiments

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Simulation Types
• Probabilistic/Stochastic vs. Deterministic Simulation
– Uses probability distributions
• Time-dependent vs. Time-independent Simulation
– Monte Carlo technique (X = A + B)
[A, B, and X are all distributions]
• Discrete Event vs. Continuous Simulation
• Simulation Implementation
– Visual Simulation and/or Object-Oriented Simulation

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.8
Cosan Improves Its Renewable Energy Supply
Chain Using Simulation
Questions for Discussion:
1. What type of supply chain disruptions might occur in
moving the sugar cane from the field to the production
plants to develop sugar and ethanol?
2. What types of advanced planning and prediction might
be useful in mitigating such disruptions?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Visual Interactive Simulation (VI S)
• Visual interactive modeling (VI M), also called Visual
Interactive Simulation or Visual interactive problem solving
• Uses computer graphics to present the impact of different
management decisions
• Often integrated with 3G and G I S
• Users can perform sensitivity analysis
• Static or dynamic (animation) systems
• Virtual reality, immersive, …

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Simulation Software
• Stand-alone desktop simulation tools
• Web-based simulation tools
• See O R/M S Today for software reviews
• Examples:
– Simio
– Arena
– ExtendSim
– SA S Simulation Studio
– …

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.9
Improving Job-Shop Scheduling Decisions
through R F I D: A Simulation-Based Assessment
Questions for Discussion:
1. In situations such as what this case depicts, what other
approaches can one take to analyze investment
decisions?
2. How would one save time if an RFI D chip can tell the
exact location of a product in process?
3. Research to learn about the applications of RFI D sensors
in other settings. Which one do you find most interesting?

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.9 – Simio Model
(1 of 4)
Figure 8.13 Simio Interface View of the Simulation System.

Source: Used with permission from Simio LL C.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.9 – Simio Model
(2 of 4)
Figure 8.14 Process View of the Simulation Model.

Back End - Process

Source: Used with permission from Simio LL C.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.9 – Simio Model
(3 of 4)
Figure 8.15 Standard Report View.

Source: Used with permission from Simio LL C.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Application Case 8.9 – Simio Model
(4 of 4)
Figure 8.16 Pivot Grid Report from a Simio Run.

OUTPUT

Source: Used with permission from Simio LL C.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
End of Chapter 8
• Questions & Comments

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved
Copyright

This work is protected by United States copyright laws and is

provided solely for the use of instructors in teaching their
courses and assessing student learning. Dissemination or sale of
any part of this work (including on the World Wide Web) will
destroy the integrity of the work and is not permitted. The work
and materials from it should never be made available to students
except by instructors using the accompanying text in their
classes. All recipients of this work are expected to abide by these
restrictions and to honor the intended pedagogical purposes and
the needs of other instructors who rely on these materials.

Copyright © 2020, 2015, 2011 Pearson Education, Inc. All Rights Reserved