Management Science

1
Management Science
QM 6433 -- Spring 2007

• Introduction to the Science of Better

Instructor John Seydel, Ph.D.
2
Why the Science of Better

• What do managers do?
• They make decisions of all kinds
• Planning
• Organizing
• Controlling
• . . .
• How well do they do it?
• Why not better?
• Why not best?
• Our goal better decision-making
• That is, Management Science is the science of
  better decision-making

3
What Takes Place in Good Decision Making?

• Decision making is based upon problem-solving
• The scientific problem-solving framework
  (SPSF)
• Define the problem
• Define decision variables
• Determine criteria of importance
• Specify whether criteria are associated with
  goals or with objectives
• Identify constraints
• Consider alternatives
• Identify them
• Evaluate them
• Select best one
• Implement solution
• Monitor and revise solution re-solve if
  appropriate

4
What Decisions Have You Made Lately?

• Possibly
• Whether or not to take this class
• How fast to drive on the way to work today
• Which clothes to wear today
• How much to spend on your best friends Christmas
  gift
• Others . . . ?
• Consider how the scientific problem-solving
  framework applies
• Realize that humans tend to employ this
  framework, even if they arent aware of it

5
And Now, This Weeks Student Objectives

• Upon completion of this weeks course
  activities, you should be able to
• Locate course materials
• Discuss and start to apply the scientific problem
  solving process
• Summarize the modeling process
• Represents simple problems as spreadsheet models
• Use Excel for basic analysis
• Perform some simple what-if analysis using Excel

6
Course Materials

• Start with the course website
• www.clt.astate.edu/jseydel/qm6433
• Under construction (always evolving)
• Major items
• Syllabus
• Grading
• Textbook (primarily used as a reference)
• Expectations
• Handouts (i.e., links thereto)
• Homework
• Later well likely be using some of
  Blackboards features
• Chat sessions
• Discussion board
• Gradebook

7
An Abbreviated Course Outline

• Spreadsheet modeling techniques
• Basic decision modeling concepts
• Using Excel to support specific modeling/solution
  techniques
• Decision analysis
• Linear programming
• Simulation modeling analysis
• Multicriteria decision making
• Project management concepts tools
• Youll understand what these specific techniques
  mean later
• Whats important is how theyre applied to
  commonly encountered business problems

8
Office 2003 Availability

• In case you havent realized it yet, Excel is
  essential for your work in this class!
• Note that, per our Microsoft Campus Agreement,
  Microsoft Office 2003 (which contains Excel) is
  available without charge for all ASU students
• Each student may either checkout an installation
  CD or download this software
• To checkout a CD
• See me
• Or go to the Library circulation desk
• Alternatively, refer to the ITS web page
  concerning the Microsoft Campus Agreement

9
More on the SPSF

• Its simply a summary of whats involved in
  organized decision-making
• It provides us a guideline to keep us on track as
  we address various problems
• Some terms you should soon be able to define
  these in non-technical terms
• Decision variables
• Criteria, objectives, goals
• Constraints
• Note that much of the framework is aimed
  developing an understanding of the problem
• In general, Management Science is concerned with
  the first two stages problem identification and
  addressing alternatives
• Consider two discipline-specific applications of
  the SPSF
• To get a better understanding about the SPSF,
  well be looking at a case . . .

10
The Gaming Company

• Read it (available from the Handouts page at the
  course website)
• Consider
• The decision(s) involved
• Whats important (i.e., how will we know if one
  decision is better than another?)
• Is there a goal, or are objectives implied
  (whats the difference)?
• What limitations are placed on the decision(s)?
• What information given is relevant to the
  decision making process here?
• Consider modeling this using Excel
• Refer to Exhibit 2 from the case
• Follow our spreadsheet design guidelines
• Start with the suggest basic worksheet layout
• For now, dont worry about a solution
• But first, lets address the concept of decision
  models . . .

11
We Will Be Using Various Decision Models

• A model involves
• Developing the model
• Inputs
• Uncontrollable (e.g., costs) also known as data
• Controllable also known as decision variables
• Outputs i.e., the results
• And typically a solution (not just an answer)
• The course of action to be taken
• That is, the values chosen for the decision
  variables
• To a great extent, developing a model is actually
  defining the problem at hand (part of the SPSF)
• Note
• Models are simplified versions of the things they
  represent
• A valid model accurately represents the relevant
  characteristics of the object or decision being
  studied consider the implications

12
Back to the Gaming Company

• Lets start working through the SPSF
• Problem identification
• Consider the alternatives
• First, we need to analyze demand, since that
  directly affects the order quantities needed
• Now, lets build a decision support worksheet and
  then do some what-if analysis
• Improving the model
• That involves working with only two decision
  variables
• Add a cell (C21) for the first decision variable,
  order quantity
• Add a cell (C22) for the other decision
  variable, reorder point
• Replace the order quantity values in Row 8 with
  an IF() function
• C8 IF(C5ltC22,C21,0)
• We also need to do some record keeping (tracking
  results)
• This should be a guided search process
• Try several different order quantities
• For each order quantity, try several different
  reorder points
• Note there are better ways to approach this,
  and well be looking at them throughout the
  semester

13
Approaches to Decision-Making

• Management Science involves four general
  categories of analysis
• What-if analysis (essentially trial and error)
• Making changes in the controllable inputs to see
  how a problems outcome is affected
• Sensitivity analysis
• Determining how changes in the uncontrollable
  inputs affect a problems solution
• Goal-seeking analysis
• Making changes in the controllable inputs until
  a desired outcome results
• Optimization analysis
• Using an mathematical approach to determine the
  values of the controllable inputs that result in
  the best (i.e., optimum) problem outcome
• Weve just gone through an example of what-if
  analysis, and well use more sophisticated
  approaches later
• We will also be exploring the other analysis
  methodologies as we discuss better ways to
  support business decision-making

14
Summary of Objectives

• Locate course materials
• Discuss and start to apply the scientific problem
  solving process
• Summarize the modeling process
• Represents simple problems as spreadsheet models
• Use Excel for basic analysis
• Perform some simple what-if analysis using Excel

15
Appendix
16
Systems Development Life Cycle

• Not a motorcycle or an exercise bike!
• One version of the SDLC
• Identifying problems, opportunities, and
  objectives
• Determining information requirements
• Analyzing system needs
• Designing the recommended system
• Developing and documenting softward
• Testing and maintaining the system
• Implementing and evaluating the system

17
Customer Decision Process
The second step in the construction of the online
offering is the articulation of the customer
decision process for the various product
categories
Flowers Example
Problem Recognition

• Need recognition, potentially triggered by a
  holiday, anniversary or everyday events

• Search for ideas and offerings, including
• Available online and offline stores
• Gift ideas and recommendations
• Advice on selection style and match

Prepurchase
Information Search
Evaluation of Alternatives

• Evaluation of alternatives along a number of
  dimensions, such as price, appeal, availability,
  etc.

Purchase
Purchase Decision

• Purchase decision
• Message selection (medium and content)

Satisfaction

• Post-sales support
• Order tracking
• Customer service

Postpurchase
Loyalty

• Education on flowers and decoration
• Post-sale perks

Disposal
18
Spreadsheet Design Guidelines

• Organize the data, then build the model around
  the data.
• Do not embed numeric constants in formulas.
• Things which are logically related should be
  physically related.
• Use formulas that can be copied.
• Column/rows totals should be close to the
  columns/rows being totalled.
• The English-reading eye scans left to right, top
  to bottom
• Use color, shading, borders and protection to
  distinguish changeable parameters from other
  model elements.
• Use text boxes and cell notes to document various
  elements of the model.

19
The Gaming CompanyA Basic Worksheet
20
The Gaming Company Problem Identification

• Note the problem identification is mostly about
  determining what the important questions are
• The decision variables (i.e., what are the
  controllable inputs?)
• How many (to order)
• When (to order)
• Criteria (i.e., whats important?)
• Total inventory-related costs minimize
• Service level maximize
• The uncontrollable inputs
• Cost rates (carrying, setup, shortage)
• Beginning inventory
• Demand . . .

21
The Gaming Company The Uncontrollable Inputs

• Cost rates
• Carrying _at_ 0.10 per case per week
• Setup _at_ 9.00 per order
• Shortage _at_ 1.00 per case per week
• Beginning inventory 43 cases
• Demand
• No trend or seasonality
• Random need an analysis
• Average
• Measure of variation
• Distribution

22
The Gaming CompanyA Univariate Analysis

• We need to understand the variation in a single
  random variable, Demand
• Recall univariate analysis tools
• Histogram (informal, visual analysis)
• Descriptive statistics
• Measures of location
• Measures of variation
• Based on the statistical analysis, the decision
  maker will have a reasonable basis for
  determining alternative order quantities
• Excel reference material see Handouts page
• First, however, we need to get the data onto an
  Excel worksheet . . .

23
The Gaming CompanyGetting The Data From The Web

• On the web page (Exhibit 1)
• Select only the data (no titles, etc.) if
  possible
• Right-click and then click on Copy
• Switch to a blank Excel worksheet
• Select top left cell
• Right-click and then on Paste Special
• Select Text and then click OK
• Remove unwanted artifacts (columns, labels, etc.)
• Reorganize as appropriate (e.g., move all the
  demand values to a single column)

24
The Gaming CompanyExcel Univariate Functions

• Functions for basic descriptive statistics
• Number of observations COUNT(range)
• Minimum value MIN(range)
• Maximum value MAX(range)
• Sample average AVERAGE(range)
• Sample standard deviation STDEV(range)
• Histogram
• No good Excel function exists
• Need to start with a frequency table
• Create a column of data values (16 up to 28)
• Use FREQUENCY(range,value) in the next column
• Example FREQUENCY(A1A1000,C11) will
  calculate how many values in the data range are
  no greater than the value in C11
• Plot a column chart from the table

25
The Gaming CompanyA Decision-Support Worksheet

• Start with the suggested basic worksheet
• Green cells indicate decision variables
  (controllable inputs)
• Yellow cells indicate uncontrollable inputs
• Sample demand (from Exhibit 2) already entered
• You need to enter the cost rates and beginning
  inventory
• Salmon cells indicate formulae
• Note there are 10 decision variables, the
  weekly order quantities
• Enter formulae . . .

26
The Gaming CompanyWorksheet Formulae

• Enter formulae for Week 1 cells, Week 2 beginning
  inventory, and the total overall cost
• C5 C15
• D5 C7C8
• C7 MAX(0,C5-C6)
• C9 IF(C8gt0,C16,0)
• C10 C7C17
• C11 IF(C6gtC5,(C6-C5)C18,0)
• C12 SUM(C9C11)
• C25 SUM(C12L12)
• Copy these across the worksheet
• Now, try various order quantities to see which
  combination results in the best bottom line total
  cost