Managerial Analysis & Decision Making Slides - Shinawatra MBA - Semester 2-2010 CC

MF1002 Managerial Analysis & Decision Making
Introduction to Managerial Decision Modeling

(Chapter 1 of Balakrishnan et al., 2007)
Assoc. Prof. Dr. Chuvej Chansa-ngavej

Program Director - Ph.D. in Management Science Shinawatra University (SIU International)
AnInternationalUniversitywithanEmphasisonResearch
Session Instructor
Associate Prof. Dr. Chuvej Chansa-ngavej Program Director, PhD in Management Science Graduate Building, SIU Room 317, 3rd Floor, BBD-Viphavadi Building Tel. 02-650-6035; 081-912-1535 Email: chuvej@siu.ac.th
PhD (Management Science in Capital Investment) Ohio State University, USA M.Eng. (Management Science in Marketing and Operations) University of New South Wales, Australia B.Eng. (1st Class Honors) in Industrial Engineering University of New South Wales, Australia
Learning Objectives
Know the historical development and origin of managerial decision modeling Recognize how managerial decision modeling can be applied Able to explain significant development directions of managerial decision modeling
3
What is Managerial Decision Modeling?

A scientific approach to managerial decision making
The development of a (mathematical) model of a real-world problem scenario The model provides insight into the solution of the managerial problem Also known as Quantitative Analysis, Management Science, Operations Research
Actual Applications in Business

Frequently used in such organizations as:

American Airlines IBM Merrill Lynch AT&T
Used in All Kinds of Enterprises

Business and Industry Government Health Care Education Agriculture Military Etc.
Origin of Managerial Decision Modeling

Eve of World War II in British military operations
Recommending the optimum location of radar signal masts Optimum size of merchant ship convoys to avoid enemy detection Optimum detonation of depth charges to destroy uboat submarines
Decision Modeling played a significant role in helping allied forces won the war
Development of Managerial Decision Modeling After World War II, applications spread to business and industry, especially in USA Courses were quickly established in prestigious universities (Massachusetts Institute of Technology, Case Western Reserve University, Ohio State University)
Areas of Modern-day Applications

Designing optimum factory layout Minimum cost construction of telecommunication network Road traffic management and one-way street allocation Determining optimal school bus routes Minimum-time design of computer-chip layout Efficient customer response tactics
9
Areas of Modern-day Applications

Flow management of raw materials and products in a supply chain
10
An Example of Modern-day Applications
Blending of raw materials in oil refineries

X2
250 objective function 350X1 + 300X2 = 35000
200
150 optimal solution 100 objective function 50 350X1 + 300X2 = 52500
0 100 0 AnInternationalUniversitywithanEmphasisonResearch 150 200 250 50
X1
11
Example of Managerial Decision Modeling in Queueing Problems

Waiting line situations are a part of everyday life People often react in an adverse manner to excessive waits in queues
Lining up for movie tickets in 1920
12
13
Characteristics of Queueing Problems

Customers arrive at some types of systems according to some type of probability distribution
Arrival rate (number of customers per time period)
Servers provide the service

Service rate (number of customers served per time period)
Customer departs the system

14
Fields Invented by MS
Decision to Launch as Planned or Delay
Market Reaction to Delay
Marketing Science
Delay Launch 1 Year
0.4
Mild
$3.6 M
Decision Analysis
EV = $1.2
0.4
Moderate
$0.3 M
0.2
Extreme
($2.0 M)
Feature Delivery EV = $6.2 M Launch as Planned Increase R&D $

0.3
Fully Featured
$17.0 M
Mild
0.2 0.7
De-featured
0.4
Moderate
0.4
Extreme
15
Fields Invented by MS
Search Theory Financial Engineering Transportation Science
16
Types of Decision Models

Deterministic Models Assume that all the input data value are known with
certainty
Probabilistic Models
Assume that some input data values are not known with certainty. Hence probability is used to represent the uncertainty Examples of probabilistic modeling techniques: queueing theory, decision analysis, simulation
17
Quantitative vs. Qualitative Data

The modeling process begins with data Quantitative Data
Numerical factors such as costs and revenues
Qualitative Data
Factors that effect the environment which are difficult to quantify into numerical measures, and subjective opinion must be used instead
18
Spreadsheets in Decision Making

Computers are used to create and solve models Spreadsheets are a convenient alternative to specialized software Microsoft Excel has extensive modeling capability via the use of add-ins (Every student must buy a new copy of the textbook to get these essential add-ins)
Premium Solver, Tree Plan, Crystal Ball, ExcelModules
19
Steps in Managerial Decision Modeling

1.
Formulation Translating a problem scenario from words to a mathematical model
2.
Solution Solving the model to obtain the optimal solution
3.
Interpretation and Sensitivity Analysis Analyzing results and implementing a solution

20
Steps in Managerial Decision Modeling
21
The Apprentice Question

Suppose you were one of the contestants, how would you use managerial decision modeling to help the team in the decision making. Answer this based on what you have learned about managerial decision modeling.
Video: The Apprentice UK, Series 6, Episode 3, October 2010
22
Example Spreadsheet Model: Tax Computation

Self employed couple must estimate and pay quarterly income tax (joint return) Income amount is uncertain 5% of income to retirement account, up to $4000 max Personal exemption = 2 x $3200 = $6400 Standard deduction = $10,000 No other deductions
File 1-1.xls
23
Tax Brackets
Taxable Income
up to $14,600 $14,601 to $59,400 $59,401 to $119,950
Percent of Taxable Income

10% 15% 25%
24
Example Spreadsheet Model: Break-Even Analysis

Profit = Revenue Costs Revenue = (Selling price) x (Num. units) Costs = (Fixed cost) + (Cost per unit) x (Num. units)
File 1-2.xls
25
The Break Even Point (BEP) is the number of units where:
Profit = 0, so Revenue = Costs
BEP
Fixed cost (Selling price) (Cost per unit)

26
Possible Problems in Developing Decision Models

Defining the Problem four typical roadblocks:
Conflicting viewpoints Impact on other departments Beginning assumptions Solution outdated

27

Developing a Model
A managers perception of a problem does not always match the textbook approach Managers do not use the results of a model they do not understand
Acquiring Input Data

Problem in using accounting data Available data must often be distilled and manipulated
28

Developing a Solution
Hard-to-understand mathematics Most managers would like to have a range of options instead of only one answer Assumptions should be reviewed
Testing the Solution
Analyzing the Results Implementation
Management support and user involvement are important

29
Session 2 - Linear Programming Models: Graphical and Computer Methods


Steps in Developing a Linear Programming (LP) Model

1)
Formulation
2)
Solution
3)
Interpretation and Sensitivity Analysis

Properties of LP Models
1) 2) 3) 4)
Seek to minimize or maximize Include constraints or limitations There must be alternatives available All equations are linear
Example LP Model Formulation: The Product Mix Problem

Decision: How much to make of > 2 products?
Objective: Maximize profit
Constraints: Limited resources
Example: Flair Furniture Co.

Two products: Chairs and Tables
Decision: How many of each to make this month?
Flair Furniture Co. Data

Tables
(per table)
Chairs
(per chair)
Profit Contribution Carpentry Painting
$7 3 hrs 2 hrs
$5 4 hrs 1 hr
Hours Available 2400 1000
Other Limitations: Make no more than 450 chairs Make at least 100 tables
6
Decision Variables: T = Num. of tables to make C = Num. of chairs to make
Objective Function: Maximize Profit Maximize $7 T + $5 C
Constraints: Have 2400 hours of carpentry time available 3 T + 4 C < 2400 (hours) Have 1000 hours of painting time available 2 T + 1 C < 1000 (hours)
More Constraints: Make no more than 450 chairs C < 450 T > 100 Nonnegativity:
Cannot make a negative number of chairs or tables (number of chairs)
Make at least 100 tables

(number of tables)
T>0 C>0
9
Model Summary
Max 7T + 5C
Subject to the constraints:
(profit)
3T + 4C < 2400 2T + 1C < 1000 C < 450 T > 100 T, C > 0
(carpentry hours) (painting hours) (max # chairs) (min # tables) (nonnegativity)

10
Graphical Solution
Graphing an LP model helps provide insight into LP models and their solutions.
While this can only be done in two dimensions, the same properties apply to all LP models and solutions.
11
Carpentry Constraint Line 3T + 4C = 2400
600
3T
Infeasible > 2400 hrs

+ 4C
Intercepts (T = 0, C = 600) (T = 800, C = 0)

0 0
Feasible < 2400 hrs
24 00
800 T
12
C 1000
Painting Constraint Line 2T + 1C = 1000

600
+ 2T = 1C 0 100
Intercepts (T = 0, C = 1000) (T = 500, C = 0)

0 0 500 800 T
13
Max Chair Line C = 450
C 1000
Min Table Line T = 100
600 450
Feasible
0
Region
0 100 500 800 T
14
7T C +5
Objective Function Line 7T + 5C = Profit

500
4 4,0 =$ 0
400
Optimal Point (T = 320, C = 360)
7T C +5 00 2,8 =$ 7T
300
C +5
00 2,1 =$
200
100
0
100
200
300
400
500 T15
Additional Constraint Need at least 75 more chairs than tables C > T + 75 Or C T > 75
200 300 500
New optimal point T = 300, C = 375
400
T = 320 C = 360 No longer feasible
100
0
100
200
300
400
500 T16
LP Characteristics
Feasible Region: The set of points that satisfies all constraints Corner Point Property: An optimal solution must lie at one or more corner points Optimal Solution: The corner point with the best objective function value is optimal
17
Special Situation in LP
1.
Redundant Constraints - do not affect the feasible region
Example:
x < 10 x < 12
The second constraint is redundant because it is less restrictive.

18
2.
Infeasibility when no feasible solution exists (there is no feasible region)
Example:
x < 10 x > 15
19
3.
Alternate Optimal Solutions when there is more than one optimal solution
C
Max 2T + 2C
Subject to:
10
2T + 2C
T + C < 10 T < 5 C< 6 T, C > 0
All points on Red segment are optimal
= 20
0 0 5 10 AnInternationalUniversitywithanEmphasisonResearch T
20
4.
Unbounded Solutions when nothing prevents the solution from becoming infinitely large
C
Max 2T + 2C
Subject to:
2T + 3C > 6 T, C > 0
ion n ct tio ire olu D s of
0 0 1 2 3 AnInternationalUniversitywithanEmphasisonResearch T
21
Using Excels Solver for LP

Recall the Flair Furniture Example: Max 7T + 5C
(profit)
3T + 4C < 2400 (carpentry hrs) 2T + 1C < 1000 (painting hrs) C < 450 (max # chairs) T > 100 (min # tables) T, C > 0
(nonnegativity) Go to file 2-1.xls
22
Session 4 - Linear Programming Modeling Applications


Linear Programming (LP) Can Be Used for Many Managerial Decisions:

Product mix Make-buy Media selection Marketing research Portfolio selection Shipping & transportation Multiperiod scheduling
2
For a particular application we begin with the problem scenario and data, then:
1) 2)
Define the decision variables Formulate the LP model using the decision variables

Write the objective function equation Write each of the constraint equations
3) 4)
Implement the model in Excel Solve with Excels Solver

3
Product Mix Problem: Fifth Avenue Industries

Produce 4 types of men's ties Use 3 materials (limited resources) Decision: How many of each type of ties to make per month? Objective: Maximize profit
4
Resource Data
Yards available Cost per yard per month $20 1,000 $6 $9 2,000 1,250
Material Silk Polyester Cotton
Labor cost is $0.75 per tie
Product Data
Type of Tie Silk Selling Price
(per tie)
Polyester Blend 1 Blend 2 $3.55 10,000 14,000 0.08 $4.31 13,000 16,000 0.10 $4.81 6,000 8,500 0.10
6
$6.70 6,000 7,000 0.125
Monthly Minimum Monthly Maximum Total material

(yards per tie)
Material Requirements (yards per tie)

Type of Tie Silk Material Silk Polyester Cotton Total yards 0.125 0 0 0.125 Blend 1 Polyester (50/50) 0 0.08 0 0.08 0 0.05 0.05 0.10 Blend 2 (30/70) 0 0.03 0.07 0.10
7
Decision Variables
S = number of silk ties to make per month P = number of polyester ties to make per month B1 = number of poly-cotton blend 1 ties to make per month B2 = number of poly-cotton blend 2 ties to make per month
8
Profit Per Tie Calculation

Profit per tie = (Selling price) (material cost) (labor cost)
Silk Tie Profit = $6.70 (0.125 yds)($20/yd) - $0.75 = $3.45 per tie
9
Objective Function (in $ of profit) Max 3.45S + 2.32P + 2.81B1 + 3.25B2

Material Limitations (in yards) 0.125S 0.05B1 + 0.07B2 < 1,000 (silk) < 1,250 (cotton)
10
0.08P + 0.05B1 + 0.03B2 < 2,000 (poly)
Min and Max Number of Ties to Make 6,000 < S < 7,000 10,000 < P < 14,000 13,000 < B1 < 16,000 6,000 < B2 < 8,500
Finally nonnegativity S, P, B1, B2 > 0

Go to file 3-1.xls
11
Media Selection Problem: Win Big Gambling Club

Promote gambling trips to the Bahamas Budget: $8,000 per week for advertising Use 4 types of advertising
Decision: How many ads of each type? Objective: Maximize audience reached
12
Data
Advertising Options Radio Newspaper (prime time) Radio
(afternoon)
TV Spot
Audience Reached (per ad) Cost (per ad) Max Ads Per week
5,000
8,500
2,400
2,800
$800 12
$925 5
$290 25
$380 20
13
Other Restrictions Have at least 5 radio spots per week Spend no more than $1800 on radio Decision Variables T = number of TV spots per week N = number of newspaper ads per week P = number of prime time radio spots per week A = number of afternoon radio spots per week
14
Objective Function
(in number of audience reached)
Max 5000T + 8500N + 2400P + 2800A

Budget is $8000 800T + 925N + 290P + 380A < 8000 At Least 5 Radio Spots per Week P+A>5
15
No More Than $1800 per Week for Radio 290P + 380A < 1800
Max Number of Ads per Week
T < 12 N< 5
Finally nonnegativity
P < 25 A < 20
T, N, P, A > 0
Go to file 3-3.xls
16
Portfolio Selection: International City Trust

Has $5 million to invest among 6 investments
Decision: How much to invest in each of 6 investment options?
Objective: Maximize interest earned

17
Data
Investment Trade credits Corp. bonds Gold stocks Platinum stocks Mortgage securities Construction loans Interest Rate 7% 10% 19% 12% 8% 14% Risk Score 1.7 1.2 3.7 2.4 2.0 2.9
18
Constraints
Invest up to $ 5 million No more than 25% into any one investment At least 30% into precious metals At least 45% into trade credits and corporate bonds Limit overall risk to no more than 2.0
19
Decision Variables
T = $ invested in trade credit B = $ invested in corporate bonds G = $ invested gold stocks P = $ invested in platinum stocks M = $ invested in mortgage securities C = $ invested in construction loans
20
Objective Function (in $ of interest earned) Max 0.07T + 0.10B + 0.19G + 0.12P + 0.08M + 0.14C
Invest Up To $5 Million T + B + G + P + M + C < 5,000,000

21
No More than 25% into Any One Investment T < 0.25 (T + B + G + P + M + C) B < 0.25 (T + B + G + P + M + C) G < 0.25 (T + B + G + P + M + C) P < 0.25 (T + B + G + P + M + C) M < 0.25 (T + B + G + P + M + C) C < 0.25 (T + B + G + P + M + C)
22
At Least 30% into Precious Metals G + P > 0.30 (T + B + G + P + M + C)
At Least 45% into Trade Credits and Corporate Bonds T + B > 0.45 (T + B + G + P + M + C)
23
Limit Overall Risk to No More Than 2.0

Use a weighted average to calculate portfolio risk 1.7T + 1.2B + 3.7G + 2.4P + 2.0M + 2.9C < 2.0 T+B+G+P+M+C or 1.7T + 1.2B + 3.7G + 2.4P + 2.0M + 2.9C < 2.0 (T + B + G + P + M + C) Finally, nonnegativity: T, B, G, P, M, C > 0
Go to file 3-5.xls
24
Labor Planning: Hong Kong Bank

Number of tellers needed varies by time of day
Decision: How many tellers should begin work at various times of the day?
Objective: Minimize personnel cost

25
Time Period 9 10 10 11 11 12 12 1 12 2-3 34 45
Min Num. Tellers 10 12 14 16 18 17 15 10
Total minimum daily requirement is 112 hours

26
Full Time Tellers Work from 9 AM 5 PM Take a 1 hour lunch break, half at 11, the other half at noon Cost $90 per day (salary & benefits) Currently only 12 are available
27
Part Time Tellers Work 4 consecutive hours (no lunch break) Can begin work at 9, 10, 11, noon, or 1 Are paid $7 per hour ($28 per day) Part time teller hours cannot exceed 50% of the days minimum requirement (50% of 112 hours = 56 hours)
28
Decision Variables F = num. of full time tellers (all work 95) P1 = num. of part time tellers who work 91 P2 = num. of part time tellers who work 102 P3 = num. of part time tellers who work 113 P4 = num. of part time tellers who work 124 P5 = num. of part time tellers who work 15
29
Objective Function (in $ of personnel cost) Min 90 F + 28 (P1 + P2 + P3 + P4 + P5)

Part Time Hours Cannot Exceed 56 Hours 4 (P1 + P2 + P3 + P4 + P5) < 56

30
Minimum Num. Tellers Needed By Hour

Time of Day
F + P1 F + P1 + P2 0.5 F + P1 + P2 + P3 0.5 F + P1 + P2 + P3+ P4 F + P2 + P3+ P4 + P5 F + P3+ P4 + P5 F + P4 + P5 F + P5
> 10 > 12 > 14 > 16 > 18 > 17 > 15 > 10
(9-10) (10-11) (11-12) (12-1) (1-2) (2-3) (3-4) (4-5)

31
Only 12 Full Time Tellers Available F < 12
Finally, nonnegativity: F, P1, P2, P3, P4, P5 > 0

Go to file 3-6.xls
32
Vehicle Loading: Goodman Shipping

How to load a truck subject to weight and volume limitations Decision: How much of each of 6 items to load onto a truck? Objective: Maximize the value shipped
33
Data
Item 1
Pounds
2
4500 $3.20 0.064
3
3000 $3.45 0.144
4
3500 $4.15 0.448
5
4000 $3.25 0.048
6
3500 $2.75 0.018
Value $15,500 $14,400 $10,350 $14,525 $13,000 $9,625 5000 $3.10 0.125 $ / lb Cu. ft. per lb
34
Decision Variables
Wi = number of pounds of item i to load onto truck, (where i = 1,,6)
Truck Capacity 15,000 pounds 1,300 cubic feet

35
Objective Function (in $ of load value) Max 3.10W1 + 3.20W2 + 3.45W3 + 4.15W4 + 3.25W5 + 2.75W6
Weight Limit of 15,000 Pounds W1 + W2 + W3 + W4 + W5 + W6 < 15,000

36
Volume Limit of 1300 Cubic Feet 0.125W1 + 0.064W2 + 0.144W3 + 0.448W4 + 0.048W5 + 0.018W6 < 1300 Pounds of Each Item Available W1 < 5000 W2 < 4500 W3 < 3000 W4 < 3500 W5 < 4000 W6 < 3500
Finally, nonnegativity: Wi > 0, i=1,,6

Go to file 3-7.xls
37
Blending Problem: Whole Food Nutrition Center

Making a natural cereal that satisfies minimum daily nutritional requirements Decision: How much of each of 3 grains to include in the cereal? Objective: Minimize cost of a 2 ounce serving of cereal
38
$ per pound Protein per pound Riboflavin per pound Phosphorus per pound Magnesium per pound
Minimum Daily $0.33 $0.47 $0.38 Requirement 22 16 8 5 28 14 7 0 21 25 9 6 3 2 1 0.425

39
Grain B
Decision Variables
A = pounds of grain A to use B = pounds of grain B to use C = pounds of grain C to use
Note: grains will be blended to form a 2- ounce serving of cereal
40
Objective Function (in $ of cost) Min 0.33A + 0.47B + 0.38C

Total Blend is 2 Ounces, or 0.125 Pounds A + B + C = 0.125 (lbs)
41
Minimum Nutritional Requirements 22A + 28B + 21C > 3 16A + 14B + 25C > 2 8A + 7B + 9C > 1 5A (protein) (riboflavin) (phosphorus)
+ 6C > 0.425 (magnesium)
Finally, nonnegativity: A, B, C > 0

Go to file 3-9.xls
42
Multiperiod Scheduling: Greenberg Motors

Need to schedule production of 2 electrical motors for each of the next 4 months Decision: How many of each type of motor to make each month? Objective: Minimize total production and inventory cost
43
Decision Variables
PAt = number of motor A to produce in month t (t=1,,4) PBt = number of motor B to produce in month t (t=1,,4) IAt = inventory of motor A at end of month t (t=1,,4) IBt = inventory of motor B at end of month t (t=1,,4)
44
Sales Demand Data

Motor A B 800 700 1000 1200
Month 1 (January) 2 (February) 3 (March) 4 (April)
1000 1400 1100 1400
45
Production Data
Motor (values are per motor) A B Production cost Labor hours $10 1.3 $6 0.9
Production costs will be 10% higher in months 3 and 4 Monthly labor hours must be between 2240 and 2560
46
Inventory Data
Motor A B Inventory cost (per motor per month) Beginning inventory (beginning of month 1) Ending Inventory (end of month 4) $0.18 $0.13 0 450 0 300
Max inventory is 3300 motors

47
Production and Inventory Balance

(inventory at end of previous period) + (production the period) (sales this period)
= (inventory at end of this period)
48
Objective Function (in $ of cost) Min 10PA1 + 10PA2 + 11PA3 + 11PA4 + 6PB1 + 6 PB2 + 6.6PB3 + 6.6PB4 + 0.18(IA1 + IA2 + IA3 + IA4) + 0.13(IB1 + IB2 + IB3 + IB4)
(see next slide)

49
Production & Inventory Balance

0 + PA1 800 = IA1 0 + PB1 1000 = IB1 IA1 + PA2 700 = IA2 IB1 + PB2 1200 = IB2 IA2 + PA3 1000 = IA3 IB2 + PB3 1400 = IB3 IA3 + PA4 1100 = IA4 IB3 + PB4 1400 = IB4
50
(month 1)
(month 2)
(month 3)
(month 4)
Ending Inventory IA4 = 450 IB4 = 300 Maximum Inventory level IA1 + IB1 < 3300 IA2 + IB2 < 3300 IA3 + IB3 < 3300 IA4 + IB4 < 3300 (month 1) (month 2) (month 3) (month 4)
51
Range of Labor Hours 2240 < 1.3PA1 + 0.9PB1 < 2560 (month 1) 2240 < 1.3PA2 + 0.9PB2 < 2560 (month 2) 2240 < 1.3PA3 + 0.9PB3 < 2560 (month 3) 2240 < 1.3PA4 + 0.9PB4 < 2560 (month 4) Finally, nonnegativity: PAi, PBi, IAi, IBi > 0
Go to file 3-11.xls
52
Session 6 Integer, Goal, and Nonlinear Programming Models


Variations of Basic Linear Programming

Integer Programming
Goal Programming
Nonlinear Programming
2
Integer Programming (IP)

Where some or all decision variables are required to be whole numbers. General Integer Variables (0,1,2,3,etc.) Values that count how many Binary Integer Variables (0 or 1) Usually represent a Yes/No decision
3
General Integer Example: Harrison Electric Co.

Produce 2 products (lamps and ceiling fans) using 2 limited resources Decision: How many of each product to make? (must be integers) Objective: Maximize profit
Decision Variables
L = number of lamps to make F = number of ceiling fans to make Lamps
(per lamp)
Fans
(per fan)
Profit Contribution Wiring Hours Assembly Hours
$600 2 hrs 6 hrs
$700 3 hrs 5 hr
Hours Available 12 30
5
LP Model Summary
Max 600 L + 700 F
($ of profit)
2L + 3F < 12 6L + 5F < 30
L, F > 0
(wiring hours) (assembly hours)
Graphical Solution
Properties of Integer Solutions

Rounding off the LP solution might not yield the optimal IP solution The IP objective function value is usually worse than the LP value IP solutions are usually not at corner points
Using Solver for IP

IP models are formulated in Excel in the same way as LP models The additional integer restriction is entered like an additional constraint int - Means general integer variables bin - Means binary variables
Go to file 6-1.xls
9
Binary Integer Example: Portfolio Selection

Choosing stocks to include in portfolio
Decision: Which of 7 stocks to include?
Objective: Maximize expected annual return (in $1000s)

10
Stock Data Oil Investment Opportunities (Table 6.2)
11
Decision Variables
Use the first letter of each stocks name
Example for Trans-Texas Oil: T = 1 if Trans-Texas Oil is included T = 0 if not included
12
Restrictions
Invest up to $3 million Include at least 2 Texas companies Include no more than 1 foreign company Include exactly 1 California company If British Petro is included, then Trans-Texas Oil must also be included
13
Objective Function (in $1000s return) Max 50T + 80B + 90D + 120H + 110L + 40S + 75C
Invest up to $3 Million 480T + 540B + 680D + 1000H + 700L + 510S + 900C < 3000
14
Include At Least 2 Texas Companies T+H+L > 2
Include No More than 1 Foreign Company B+D < 1
Include Exactly 1 California Company S+C = 1

15
If British Petro is included (B=1), then Trans-Texas Oil must also be included (T=1)
T=0 T=1 ok ok
Combinations of B and T
B=0
ok
B=1 not ok
B<T allows the 3 acceptable combinations and prevents the unacceptable one
Go to file 6-3.xls
16
Mixed Integer Models: Fixed Charge Problem

Involves both fixed and variable costs
Use a binary variable to determine if a fixed cost is incurred or not
Either linear or general integer variables deal with variable cost

17
Fixed Charge Example: Hardgrave Machine Co.

Has 3 plants and 4 warehouses and is considering 2 locations for a 4th plant Decisions: Which location to choose for 4th plant? How much to ship from each plant to each warehouse? Objective: Minimize total production and shipping cost
18
Supply and Demand Data

Production Monthly Warehouse Detroit Houston New York Los Angeles Total Demand 10,000 12,000 15,000 9,000 46,000 35,000 Plant Cincinnati Kansas City Pittsburgh Monthly Supply 15,000 6,000 14,000 Cost (per unit) $48 $50 $52
Note: New plant must supply 11,000 units per month

19
Possible Locations for New Plant Production Cost (per unit) Seattle Birmingham $53 $49 Fixed Cost (per month) $400,000 $325,000
20
Shipping Cost Data Hardgrave Machines (Table 6.6)
21
Decision Variables
Binary Variables Ys = 1 if Seattle is chosen = 0 if not YB = 1 if Birmingham is chosen = 0 if not Regular Variables Xij = number of units shipped from plant i to warehouse j
22
Objective Function (in $ of cost) Min 73XCD + 103XCH + 88XCN + 108XCL + 85XKD + 80XKH + 100XKN + 90XKL + 88XPD + 97XPH + 78XPN + 118XPL + 113XSD + 91XSH + 118XSN + 80XSL + 84XBD + 79XBH + 90XBN + 99XBL + 400,000YS + 325,000YB
Subject to the constraints: (see next slide)
23
Supply Constraints -(XCD + XCH + XCN + XCL) = -15,000 -(XKD + XKH + XKN + XKL) = - 6,000 -(XPD + XPH + XPN + XPL) = -15,000
(Cincinnati) (Kansas City) (Pittsburgh)
Possible Locations for New Plant
-(XSD + XSH + XSN + XSL) = -11,000YS
(Seattle)
-(XBD + XBH + XBN + XBL) = -11,000YB (Bham)

24
Demand Constraints XCD + XKD + XPD +XSD + XBD = 10,000 XCH + XKH + XPH +XSH + XBH = 12,000 XCN + XKN + XPN +XSN + XBN = 15,000 XCL + XKL + XPL +XSL + XBL = 9,000 Choose 1 New Plant Location YS + YB =1
(Detroit) (Houston) (New York) (L. A.)
Go to File 6-5.xls
25
Goal Programming Models

Permit multiple objectives
Try to satisfy goals rather than optimize
Objective is to minimize underachievement of goals
26
Goal Programming Example: Wilson Doors Co.

Makes 3 types of doors from 3 limited resources Decision: How many of each of 3 types of doors to make? Objective: Minimize total underachievement of goals
27
Data Wilson Doors (Table 6.7)
28
Goals
1. 2. 3. 4.
Total sales at least $180,000 Exterior door sales at least $70,000 Interior door sales at lest $60,000 Commercial door sales at least $35,000
29
Regular Decision Variables E = number of exterior doors made I = number of interior doors made C = number of commercial doors made Deviation Variables di+ = amount by which goal i is overachieved di- = amount by which goal i is underachieved
30
Goal Constraints
Goal 1: Total sales at least $180,000 70E + 110I + 110C + dT- - dT+ = 180,000 Goal 2: Exterior door sales at least $70,000 70E + dE- - dE+ = 70,000 Note: Each highlighted deviation variable measures goal underachievement
31
Goal 3: Interior door sales at least $60,000 110 I + dI- - dI+ = 60,000
Goal 4: Commercial door sales at least $35,000 110C + dC- - dC+ = 35,000
32
Objective Function
Minimize total goal underachievement Min dT- + dE- + dI- + dCSubject to the constraints: The 4 goal constraints The regular constraints (3 limited resources) nonnegativity
33
Weighted Goals
When goals have different priorities, weights can be used Suppose that Goal 1 is 5 times more important than each of the others Objective Function Min 5dT- + dE- + dI- + dCAnInternationalUniversitywithanEmphasisonResearch
34
Properties of Weighted Goals

Solution may differ depending on the weights used
Appropriate only if goals are measured in the same units
35
Ranked Goals
Lower ranked goals are considered only if all higher ranked goals are achieved Suppose they added a 5th goal Goal 5: Steel usage as close to 9000 lb as possible 4E + 3I + 7C + dS= 9000 (lbs steel)
(no dS+ is needed because we cannot exceed 9000 pounds)

36
Rank R1: Goal 1 Rank R2: Goal 5 Rank R3: Goals 2, 3, and 4 A series of LP models must be solved
1)
Solve for the R1 goal while ignoring the other goals Objective Function: Min dTGo to file 6-7.xls
37
2) If the R1 goal can be achieved (dT- = 0), then this is added as a constraint and we attempt to satisfy the R2 goal (Goal 5) Objective Function: Min dS3) If the R2 goal can be achieved (dS- = 0), then this is added as a constraint and we solve for the R3 goals (Goals 2, 3, and 4) Objective Function: Min dE- + dI- + dCAnInternationalUniversitywithanEmphasisonResearch
38
Nonlinear Programming Models

Linear models (LP, IP, and GP) have linear objective function and constraints If a model has one or more nonlinear equations (objective or constraint) then the model is nonlinear Example nonlinear terms: X2, 1/X, XY
39
Characteristics of Nonlinear Programming (NLP) Models

Difficult to solve Optimal solutions are not necessarily at corner points There are both local and global optimal solutions Solution may depend on starting point Starting point is usually arbitrary
40
Nonlinear Programming Example: Pickens Memorial Hospital

Patient demand exceeds hospitals capacity
Decision: How many of each of 3 types of patients to admit per week?

41
Decision Variables
M = number of Medical patients to admit S = number of Surgical patients to admit P = number of Pediatric patients to admit
Profit Function
Profit per patient increases as the number of patients increases (i.e. nonlinear profit function)
42
Constraints
Hospital capacity: 200 total patients X-ray capacity: 560 x-rays per week Marketing budget: $1000 per week Lab capacity: 140 hours per week
43
Objective Function 60P + 3P2

(in $ of profit)
Max 45M + 2M2 + 70S + 3S2 + 2MS +
M+S+P M + 3S + P 3M + 5S + 3.5P M, S, P > 0
< 200 < 560 < 1000
(patient cap.) (x-ray cap.) (marketing $) (lab hrs)
(0.2+0.001M)x(3M+3S+3P) < 140
44
Using Solver for NLP Models

Solver uses the Generalized Reduced Gradient (GRG) method GRG uses the path of steepest ascent (or descent) Moves from one feasible solution to another until the objective function value stops improving (converges)
Go to file 6-8.xls
45
Session 12 Decision Analysis

Decision Analysis
For evaluating and choosing among alternatives
Considers all the possible alternatives and possible outcomes
Five Steps in Decision Making

1. 2. 3.
Clearly define the problem List all possible alternatives Identify all possible outcomes for each alternative Identify the payoff for each alternative & outcome combination Use a decision modeling technique to choose an alternative
3
4.
5.
Thompson Lumber Co. Example

1.
Decision: Whether or not to make and sell storage sheds Alternatives:

2.
Build a large plant Build a small plant Do nothing Outcomes: Demand for sheds will be high, moderate, or low
4
3.
4.
Payoffs
Outcomes (Demand)
Alternatives Large plant Small plant No plant
High 200,000 90,000 0
Moderate 50,000 0
Low -20,000 0
100,000 -120,000
5.
Apply a decision modeling method

Decision Making Under Risk

Where probabilities of outcomes are available Expected Monetary Value (EMV) uses the probabilities to calculate the average payoff for each alternative EMV (for alternative i) = (probability of outcome) x (payoff of outcome)
6
Expected Monetary Value (EMV) Method Outcomes (Demand) Low Alternatives High Moderate Large plant 200,000 100,000 -120,000 Small plant No plant Probability of outcome 0.3 90,000 0 0.5 50,000 0 0.2 -20,000 0 EMV 86,000 48,000 0
Chose the large plant

7
Perfect Information
Perfect Information would tell us with certainty which outcome is going to occur Having perfect information before making a decision would allow choosing the best payoff for the outcome
The expected payoff of having perfect information before making a decision EVwPI = (probability of outcome)
Expected Value With Perfect Information (EVwPI)
x ( best payoff of outcome)
Expected Value of Perfect Information (EVPI)

The amount by which perfect information would increase our expected payoff Provides an upper bound on what to pay for additional information EVPI = EVwPI EMV
EVwPI = Expected value with perfect information EMV = the best EMV without perfect information
10
Payoffs in blue would be chosen based on perfect information (knowing demand level) Demand Alternatives Large plant Small plant No plant Probability 0.3 High 200,000 90,000 0 0.5 Moderate 50,000 0 0.2 Low -20,000 0 100,000 -120,000
EVwPI = $110,000
11
Expected Value of Perfect Information

EVPI = EVwPI EMV = $110,000 - $86,000 = $24,000
The perfect information increases the expected value by $24,000 Would it be worth $30,000 to obtain this perfect information for demand?
12
Decision Trees
Can be used instead of a table to show alternatives, outcomes, and payofffs Consists of nodes and arcs Shows the order of decisions and outcomes
13
Decision Tree for Thompson Lumber
14
Folding Back a Decision Tree

For identifying the best decision in the tree Work from right to left Calculate the expected payoff at each outcome node Choose the best alternative at each decision node (based on expected payoff)
15
Thompson Lumber Tree with EMVs
16
Using TreePlan With Excel

An add-in for Excel to create and solve decision trees
Load the file Treeplan.xla into Excel (from the CD-ROM)
17
Decision Trees for Multistage Decision-Making Problems

Multistage problems involve a sequence of several decisions and outcomes It is possible for a decision to be immediately followed by another decision Decision trees are best for showing the sequential arrangement
18
Expanded Thompson Lumber Example

Suppose they will first decide whether to pay $4000 to conduct a market survey Survey results will be imperfect Then they will decide whether to build a large plant, small plant, or no plant Then they will find out what the outcome and payoff are
19
20
21
Thompson Lumber Optimal Strategy

1.
Conduct the survey
2.
If the survey results are positive, then build the large plant (EMV = $141,840) If the survey results are negative, then build the small plant (EMV = $16,540)
22
Expected Value of Sample Information (EVSI)

The Thompson Lumber survey provides sample information (not perfect information) What is the value of this sample information? EVSI = (EMV with free sample information) - (EMV w/o any information)
23
EVSI for Thompson Lumber

If sample information had been free EMV (with free SI) = 87,961 + 4000 = $91,961
EVSI = 91,961 86,000 = $5,961
24
EVSI vs. EVPI

How close does the sample information come to perfect information? Efficiency of sample information = EVSI EVPI Thompson Lumber: 5961 / 24,000 = 0.248
25
Estimating Probability Using Bayesian Analysis

Allows probability values to be revised based on new information (from a survey or test market) Prior probabilities are the probability values before new information Revised probabilities are obtained by combining the prior probabilities with the new information
26
Known Prior Probabilities P(HD) = 0.30 P(MD) = 0.50 P(LD) = 0.30
How do we find the revised probabilities where the survey result is given? For example: P(HD|PS) = ?
27
It is necessary to understand the Conditional probability formula: P(A|B) = P(A and B) P(B) P(A|B) is the probability of event A occurring, given that event B has occurred When P(A|B) P(A), this means the probability of event A has been revised based on the fact that event B has occurred
28
The marketing research firm provided the following probabilities based on its track record of survey accuracy: P(PS|HD) = 0.967 P(PS|MD) = 0.533 P(PS|LD) = 0.067 P(NS|HD) = 0.033 P(NS|MD) = 0.467 P(NS|LD) = 0.933
Here the demand is given, but we need to reverse the events so the survey result is given
29
Finding probability of the demand outcome given the survey result:

P(HD|PS) = P(HD and PS) = P(PS|HD) x P(HD) P(PS) P(PS)
Known probability values are in blue, so need to find P(PS) P(PS|HD) x P(HD) + P(PS|MD) x P(MD) 0.967 x 0.30 + 0.533 x 0.50
30
AnInternationalUniversitywithanEmphasisonResearch + P(PS|LD) x P(LD) + 0.067 x 0.20
Now we can calculate P(HD|PS):

P(HD|PS) = P(PS|HD) x P(HD) = 0.967 x 0.30 P(PS) = 0.509 0.57
The other five conditional probabilities are found in the same manner Notice that the probability of HD increased from 0.30 to 0.509 given the positive AnInternationalUniversitywithanEmphasisonResearch survey result
31
Utility Theory
An alternative to EMV People view risk and money differently, so EMV is not always the best criterion Utility theory incorporates a persons attitude toward risk A utility function converts a persons attitude toward money and risk into a number between 0 and 1
32
Janes Utility Assessment
Jane is asked: What is the minimum amount that AnInternationalUniversitywithanEmphasisonResearch would cause you to choose alternative 2?
33
Suppose Jane says $15,000 Jane would rather have the certainty of getting $15,000 rather the possibility of getting $50,000 Utility calculation:
U($15,000) = U($0) x 0.5 + U($50,000) x 0.5 Where, U($0) = U(worst payoff) = 0 U($50,000) = U(best payoff) = 1
AnInternationalUniversitywithanEmphasisonResearch U($15,000) = 0 x 0.5 + 1 x 0.5 = 0.5 (for Jane)
34
The same gamble is presented to Jane multiple times with various values for the two payoffs Each time Jane chooses her minimum certainty equivalent and her utility value is calculated A utility curve plots these values
35
Janes Utility Curve
36
Different people will have different curves Janes curve is typical of a risk avoider Risk premium is the EMV a person is willing to willing to give up to avoid the risk
Risk premium = (EMV of gamble) (Certainty equivalent)
Janes risk premium = $25,000 - $15,000

= $10,000
37
Types of Decision Makers

Risk Premium Risk avoiders: Risk neutral people: Risk seekers: >0 =0 <0
38
Utility Curves for Different Risk Preferences
39
Utility as a Decision Making Criterion

Construct the decision tree as usual with the same alternative, outcomes, and probabilities Utility values replace monetary values Fold back as usual calculating expected utility values
40
Decision Tree Example for Mark
41
Utility Curve for Mark the Risk Seeker
42
Marks Decision Tree With Utility Values
43

Managerial Analysis &amp; Decision Making Slides - Shinawatra MBA - Semester 2-2010 CC

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Managerial Analysis &amp; Decision Making Slides - Shinawatra MBA - Semester 2-2010 CC

Hochgeladen von

Copyright:

Verfügbare Formate

MF1002 Managerial Analysis & Decision Making

Introduction to Managerial Decision Modeling

Assoc. Prof. Dr. Chuvej Chansa-ngavej

What is Managerial Decision Modeling?

Actual Applications in Business

American Airlines IBM Merrill Lynch AT&T

Used in All Kinds of Enterprises

Origin of Managerial Decision Modeling

Areas of Modern-day Applications

Areas of Modern-day Applications

An Example of Modern-day Applications

Blending of raw materials in oil refineries

150 optimal solution 100 objective function 50 350X1 + 300X2 = 52500

0 100 0 AnInternationalUniversitywithanEmphasisonResearch 150 200 250 50

Example of Managerial Decision Modeling in Queueing Problems

Lining up for movie tickets in 1920

Characteristics of Queueing Problems

Servers provide the service

Customer departs the system

Feature Delivery EV = $6.2 M Launch as Planned Increase R&D $

Types of Decision Models

Quantitative vs. Qualitative Data

Spreadsheets in Decision Making

Steps in Managerial Decision Modeling

Formulation Translating a problem scenario from words to a mathematical model

Solution Solving the model to obtain the optimal solution

Interpretation and Sensitivity Analysis Analyzing results and implementing a solution

Steps in Managerial Decision Modeling

The Apprentice Question

Video: The Apprentice UK, Series 6, Episode 3, October 2010

Example Spreadsheet Model: Tax Computation

Percent of Taxable Income

Example Spreadsheet Model: Break-Even Analysis

The Break Even Point (BEP) is the number of units where:

Profit = 0, so Revenue = Costs

Fixed cost (Selling price) (Cost per unit)

Possible Problems in Developing Decision Models

Conflicting viewpoints Impact on other departments Beginning assumptions Solution outdated

Possible Problems in Developing Decision Models

Acquiring Input Data

Possible Problems in Developing Decision Models

Testing the Solution

Analyzing the Results Implementation

Management support and user involvement are important

MF1002 Managerial Analysis & Decision Making

Session 2 - Linear Programming Models: Graphical and Computer Methods

Assoc. Prof. Dr. Chuvej Chansa-ngavej

Steps in Developing a Linear Programming (LP) Model

Interpretation and Sensitivity Analysis

Example LP Model Formulation: The Product Mix Problem

Objective: Maximize profit

Constraints: Limited resources

Example: Flair Furniture Co.

Decision: How many of each to make this month?

Objective: Maximize profit

Flair Furniture Co. Data

Profit Contribution Carpentry Painting

Hours Available 2400 1000

Decision Variables: T = Num. of tables to make C = Num. of chairs to make

Objective Function: Maximize Profit Maximize $7 T + $5 C

Make at least 100 tables

3T + 4C < 2400 2T + 1C < 1000 C < 450 T > 100 T, C > 0

(carpentry hours) (painting hours) (max # chairs) (min # tables) (nonnegativity)

Managerial Analysis & Decision Making Slides - Shinawatra MBA - Semester 2-2010 CC

Managerial Analysis & Decision Making Slides - Shinawatra MBA - Semester 2-2010 CC