Lecture 5: Integer Programming

OSCM 471/571: Optimization and Decision Support Modeling for Business

Seokjun Youn
([email protected])

Eller College of Management

University of Arizona
Agenda
Announcement
 Homework 3, Due: Mar 14, 11:59 pm

Today’s Plan
 Lecture 5: Binary Integer Programming
• Reading: OptDSM_Ch7_Integer Programming_Reading.pdf
1. A Case Study: California Manufacturing (Sec 7.1)
2. Project Selection: Tazer Corp. (Sec 7.2)
3. Selection of Sites: Caliente City (Sec 7.3)
4. Crew Scheduling: Southwestern Airways (Sec 7.4)
5. Mixed BIP to Deal with Setup Costs: Revised Wyndor (Sec 7.5)

Next Class
 Term Project Guidelines & Team Assignment
 Lecture 5: Binary Integer Programming (cont’d)
2
Applications of Binary Variables
Since binary variables only provide two choices, they are ideally suited to be the
decision variables when dealing with yes-or-no decisions.

Examples:
• Should we undertake a particular fixed project?
• Should we make a particular fixed investment?
• Should we locate a facility in a particular site?

3
1. A Case Study: California Manufacturing

4
California Manufacturing Company
• The California Manufacturing Company is a diversified company with several
factories and warehouses throughout California, but none yet in Los Angeles
or San Francisco.
• A basic issue is whether to build a new factory in Los Angeles or San
Francisco, or perhaps even both.
• Management is also considering building at most one new warehouse, but
will restrict the choice to a city where a new factory is being built.
• “Management wants no more than one new warehouse to be built.”
• “Management will allow a warehouse to be built in a particular city only if a
factory also is being built in that city.”

Question:
Should the California Manufacturing Company expand with factories and/or
warehouses in Los Angeles and/or San Francisco? 5
Data for California Manufacturing

Net Present Capital

Decision Yes-or-No Decision Value Required
Number Question Variable (Millions) (Millions)
1 Build a factory in Los Angeles? x1 $8 $6

2 Build a factory in San Francisco? x2 5 3

3 Build a warehouse in Los Angeles? x3 6 5

4 Build a warehouse in San Francisco? x4 4 2

Capital Available: $10 million

6
Binary Decision Variables

Decision Decision Possible Interpretation Interpretation

Number Variable Value of a Value of 1 of a Value of 0

Build a factory in Do not build

1 x1 0 or 1
Los Angeles this factory

Build a factory in Do not build

2 x2 0 or 1
San Francisco this factory

Build a warehouse in Do not build

3 x3 0 or 1
Los Angeles this warehouse

Build a warehouse in Do not build

4 x4 0 or 1
San Francisco this warehouse

7
Algebraic Formulation 1

Let x1 = 1 if build a factory in L.A.; 0 otherwise

x2 = 1 if build a factory in S.F.; 0 otherwise
x3 = 1 if build a warehouse in Los Angeles; 0 otherwise
x4 = 1 if build a warehouse in San Francisco; 0 otherwise

Maximize NPV = 8x1 + 5x2 + 6x3 + 4x4 ($millions)

subject to
Capital Spent: 6x1 + 3x2 + 5x3 + 2x4 ≤ 10 ($millions)
Max 1 Warehouse: x3 + x4 ≤ 1 : “mutually exclusive alternatives”
Warehouse only if Factory: x3 ≤ x1
: “contingent decisions”
x4 ≤ x 2
x1, x2, x3, x4 are binary variables.

8
Spreadsheet Model 1

9
 Management’s Conclusion
• Management’s initial tentative decision had been to make $10 million of capital available:
• With this much capital, the best plan would be to build a factory in both Los Angeles and San
Francisco, but no warehouses.
• An advantage of this plan is that it only uses $9 million of this capital, which frees up $1 million for
other projects.
• If below $9 million of capital available:
• A heavy penalty (a reduction of $4 million in total net present value) would be paid if the capital
made available were to be reduced below $9 million.
• If $11 million of capital available:
• Increasing the capital made available by $1 million (to $11 million) would enable a substantial ($4
million) increase in the total net present value. Management decides to do this.
• With this much capital available, the best plan is to build a factory in both cities and a warehouse in
San Francisco.

10
Some Other Applications
Investment Analysis
• Should we make a certain fixed investment?
• Examples: Petroleum Refineries, National Defense Force.
Site Selection
• Should a certain site be selected for the location of a new facility?
• Example: AT&T

Designing a Production and Distribution Network

• Should a certain plant remain open? Should a certain site be selected for a new plant?
• Should a distribution center remain open? Should a certain site be selected for a new
distribution center?
• Should a certain distribution center be assigned to serve a certain market area?

11
Some Other Applications
Dispatching Shipments
• Should a certain route be selected for a truck? Should a certain size truck be used?
Should a certain time period for departure be used?

Scheduling Interrelated Activities

• Should a certain activity begin in a certain time period?

Scheduling Asset Divestitures

• Should a certain asset be sold in a certain time period?

Airline Applications:
• Should a certain type of airplane be assigned to a certain flight leg?
• Should a certain sequence of flight legs be assigned to a crew?
12
2. Project Selection: Tazer Corp.

13
Project Selection at Tazer Corp.
Tazer Corporation, a pharmaceutical manufacturing company, is searching for a new
breakthrough drug.
Five potential research and development projects:
• Project Up: Develop a more effective antidepressant that doesn’t cause mood swings.
• Project Stable: Develop a drug that addresses manic depression.
• Project Choice: Develop a less intrusive birth control method for women.
• Project Hope: Develop a vaccine to prevent HIV infection.
• Project Release: Develop a more effective drug to lower blood pressure.

$1.2 billion available for investment (enough for 2 or 3 projects).

Question:
Which projects should be selected to research and develop?

14
Data for the Tazer Project Selection Problem
Blank 1 2 3 4 5
Up Stable Choice Hope Release

R&D
($million) 400 300 600 500 200

Success Rate
50% 35% 35% 20% 45%

Revenue if
Successful 1,400 1,200 2,200 3,000 600
($million)

Expected
Profit 300 120 170 100 70
($million)

Expected Profit = Revenue if Successful * Success Rate – R&D Cost 15

Algebraic Formulation of Tazer Project Selection
Let xi = 1 if approve project i; 0 otherwise (for i = 1, 2, 3, 4, and 5)

Maximize P = 300x1 + 120x2 + 170x3 + 100x4 + 70x5 ($million)

subject to
R&D Budget: 400x1 + 300x2 + 600x3 + 500x4 +200x5 ≤ 1,200 ($million)

and xi are binary (for i = 1, 2, 3, 4, and 5).

16
 Spreadsheet for Tazer Project Selection Problem

A B C D E F G H I J
1 Tazer Corp. Project Selection Problem
2
3
4 Up Stable Choice Hope Release Total Budget
5 R&D Investment ($million) 400 300 600 500 200 1200 <= 1200
6 Success Rate 50% 35% 35% 20% 45%
7 Revenue if Successful ($million) 1400 1200 2200 3000 600
8 Expected Profit ($million) 300 120 170 100 70 540
9
10 Do Project? 1 0 1 0 1

17
Agenda
• Announcement
• Homework 3, Due: Mar 14, 11:59 pm

• Today’s Plan
• Lecture 5: Binary Integer Programming
• Reading: OptDSM_Ch7_Integer Programming_Reading.pdf
1. A Case Study: California Manufacturing (Sec 7.1)
2. Project Selection: Tazer Corp. (Sec 7.2)
3. Selection of Sites: Caliente City (Sec 7.3)
4. Crew Scheduling: Southwestern Airways (Sec 7.4)
5. Mixed BIP to Deal with Setup Costs: Revised Wyndor (Sec 7.5)

• Next Class
• Lecture 6: Nonlinear Programming

18
3. Selection of Sites for Emergency Services:
Caliente City

19
20
Selection of Sites for Emergency Services:
The Caliente City Problem
• Caliente City is growing rapidly and spreading well beyond its original borders.
• They still have only one fire station, located in the congested center of town.
• The result has been long delays in fire trucks reaching the outer part of the city.

Goal: Develop a plan for locating multiple fire stations throughout the city

New Policy: Ensure “Response Time ≤ 10 minutes” anywhere in the city.

21
 Response Time and Cost Data for Caliente City
“Response Time ≤ 10 minutes” Consider Tract 1: The only
tracts close enough that a fire
station would provide a
response time not exceeding 10
minutes are tract 1, 2, and 4.

Thus, at least one of these

three tracts needs to have a fire
station. That is,
𝑥𝑥1 + 𝑥𝑥2 + 𝑥𝑥3 ≥ 1

The city is divided into eight tracts and then data is gathered on the estimated response time 22
for a fire in each tract from a potential fire station in each of the eight tracts.
Algebraic Formulation of Caliente City Problem
Set Covering Problem

Let xj = 1 if tract j is selected to receive a fire station; 0 otherwise

( j = 1,2,...,8 )
Minimize C = 350x1 + 250x2 + 450x3 + 300x4 + 50x5 + 400x6 + 300x7 + 200x8
subject to Tract 1 : x1 + x2 + x4 ≥ 1
“set covering constraints”
Tract 2 : x1 + x2 + x3 ≥ 1
Tract 3 : x2 + x3 + x6 ≥ 1 because it covers the requirement of having a fire
Tract 4 : x1 + x4 + x7 ≥ 1 station located in at least one member of the set of
tracts.
Tract 5 : x5 + x7 ≥ 1
Tract 6 : x3 + x6 + x8 ≥ 1
Tract 7 : x4 + x7 + x8 ≥ 1
Tract 8 : x6 + x7 + x8 ≥ 1
and xj are binary ( for j = 1,2,...,8 ) .
23
Spreadsheet Model for Caliente City Problem
A B C D E F G H I J K L M N
1 Caliente City Fire Station Location Problem
2
3 Fire Station in Tract
4 1 2 3 4 5 6 7 8
5 1 2 8 18 9 23 22 16 28
6 Response 2 9 3 10 12 16 14 21 25
7 Times 3 17 8 4 20 21 8 22 17
8 (Minutes) 4 10 13 19 2 18 21 6 12
9 for a Fire 5 21 12 16 13 5 11 9 12
10 in Tract 6 25 15 7 21 15 3 14 8
11 7 14 22 18 7 13 15 2 9
12 8 30 24 15 14 17 9 8 3
13
14 Cost of Station 350 250 450 300 50 400 300 200
15 ($thousands) Number
16 Covering
17 1 1 1 0 1 0 0 0 0 1 >= 1
18 Response 2 1 1 1 0 0 0 0 0 1 >= 1
19 Time 3 0 1 1 0 0 1 0 0 1 >= 1
20 <= 4 1 0 0 1 0 0 1 0 1 >= 1
21 10 5 0 0 0 0 1 0 1 0 1 >= 1
22 Minutes? 6 0 0 1 0 0 1 0 1 1 >= 1
23 7 0 0 0 1 0 0 1 1 2 >= 1
24 8 0 0 0 0 0 1 1 1 2 >= 1
25
26 Total
27 Fire Station in Tract Cost
28 1 2 3 4 5 6 7 8 ($thousands)
29 Station in Tract? 0 1 0 0 0 0 1 1 750
24
4. Crew Scheduling:
Southwestern Airways
Another application of “Set Covering Problem”

25
Southwestern Airways Crew Scheduling
• Throughout the travel industry (airlines, rail travel, cruise ships, tour companies,
etc.), one of the most challenging problems in maintaining an efficient operation
is the scheduling of its crews who serve customers during their travels.

• Southwestern Airways needs to assign crews to cover all its upcoming flights.
• We will focus on assigning 3 crews based in San Francisco (SFO) to 11 flights.

Question:
How should the 3 crews be assigned 3 sequences of flights so that every one of
the 11 flights is covered?

26
Southwestern Airways Flights

Flight Sequence Key

1 SFO-LAX
2 SFO-DEN-SFO
3 SFO-SEA-SFO
4 SFO-LAX-ORD-DEN-SFO
5 SFO-DEN-ORD-DEN-SFO
6 SFO-SEA-LAX-SFO
7 SFO-LAX-ORD-SEA-SFO
8 SFO-DEN-ORD-SEA-SFO
9 SFO-SEA-LAX-ORD-DEN-SFO
10 SFO-LAX-ORD-SEA-LAX-SFO
11 SFO-DEN-ORD-SEA-LAX-SFO
12 SFO-SEA-LAX-ORD-SEA-SFO

27
 Data for the Southwestern Airways Problem
“Given many feasible overlapping sequences of trips for a crew, to which ones
should a crew be assigned so as to cover all the trips at a minimum cost?”
Feasible Sequence of Flights
Flights 1 2 3 4 5 6 7 8 9 10 11 12
1. SFO–LAX 1 blank
1 1 1
blank blank blank blank blank blank blank

2. SFO–DEN blank
1 blank blank
1 1 1 blank
blank blank blank blank

Flight Sequence Key

3. SFO–SEA 1 blank
1 blank
1 1
blank blank blank blank blank blank

1 SFO-LAX
2 SFO-DEN-SFO
4. LAX–ORD 2 blank
2 3 2 blank
3
blank blank blank blank blank

3 SFO-SEA-SFO
5. LAX–SFO 2 blank blank
3 blank
5 5 blank
blank blank blank blank

4 SFO-LAX-ORD-DEN-SFO
5 SFO-DEN-ORD-DEN-SFO
6. ORD–DEN 3 3 4
blank blank blank blank blank blank blank blank blank

6 SFO-SEA-LAX-SFO
7 SFO-LAX-ORD-SEA-SFO
7. ORD–SEA blank blank
3 3 blank
3 3 4
blank blank blank blank

8 SFO-DEN-ORD-SEA-SFO
9 SFO-SEA-LAX-ORD-DEN-SFO
8. DEN–SFO blank
2 4 4 blank
5
blank blank blank blank blank blank

10 SFO-LAX-ORD-SEA-LAX-SFO
9. DEN–ORD 2 2 2 blank
blank blank blank blank blank blank blank blank

11 SFO-DEN-ORD-SEA-LAX-SFO
12 SFO-SEA-LAX-ORD-SEA-SFO
10. SEA–SFO 2 4 4 blank
5
blank blank blank blank blank blank blank

11. SEA–LAX blank

2 blank blank
2 4 4 2
blank blank blank blank

Cost, $1,000s 2 3 4 6 7 5 7 8 9 9 8 9
28
The numbers in each column indicate the order of the flights.
Algebraic Formulation
Let xj = 1 if flight sequence j is assigned to a crew; 0 otherwise. ( j = 1,2, ...,12).

Minimize Cost = 2x1 + 3x2 + 4x3 + 6x4 + 7x5 + 5x6 + 7x7 + 8x8 + 9x9 + 9x10 + 8x11 + 9x12 (in $thousands)

subject to
Flight 1 covered: x1 + x4 + x7 + x10 ≥ 1
Flight 2 covered: x2 + x5 + x8 + x11 ≥ 1
⁞ ⁞
Flight 11 covered: x6 + x9 + x10 + x11 + x12 ≥ 1
Three Crews: x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + x12 ≤ 3

and
xj are binary ( j = 1,2, ...,12).

29
Spreadsheet Model

Flight Sequence Key

1 SFO-LAX
2 SFO-DEN-SFO
3 SFO-SEA-SFO
4 SFO-LAX-ORD-DEN-SFO
5 SFO-DEN-ORD-DEN-SFO
6 SFO-SEA-LAX-SFO
7 SFO-LAX-ORD-SEA-SFO
8 SFO-DEN-ORD-SEA-SFO
9 SFO-SEA-LAX-ORD-DEN-SFO
10 SFO-LAX-ORD-SEA-LAX-SFO
11 SFO-DEN-ORD-SEA-LAX-SFO
12 SFO-SEA-LAX-ORD-SEA-SFO

30
5. Mixed BIP to Deal with Setup Costs:
Revised Wyndor

31
Wyndor with Setup Costs

Suppose that two changes are made to the original Wyndor problem:
1. For each product, producing any units requires a substantial one-time setup cost
for setting up the production facilities.
2. The production runs for these products will be ended after one week, so D and W
in the original model now represent the total number of doors and windows
produced, respectively, rather than production rates. Therefore, these two
variables need to be restricted to integer values.

32
Graphical Solution to Original Wyndor Problem
W
Production rate
for windows
8

Optimal solution

6 (2, 6)

Feasible P = 3,600 = 300 D+ 500 W

4 Region

0 2 4 6 8 10 D
Production rate for doors
33
Net Profit for Wyndor Problem with Setup Costs

Number of Net Profit of Doors in Net Profit of Windows

Units Produced Dollars in Dollars
0 0(300) − 0 =0 0(500) − 0 =0
1 1(300) − 700 =
−400 1(500) − 1,300 =
−800
2 2(300) − 700 =
−100 2(500) − 1,300 =
−300
3 3(300) − 700 =
200 3(500) − 1,300 =
200
4 4(300) − 700 =
500 4(500) − 1,300 =
700
5 5(500) − 1,300 =
1,200
6 6(500) − 1,300 =
1,700

34
 Feasible Solutions for Wyndor with Setup Costs
W

8
Production
quantity for
windows (0, 6) gives P = 1700
6 (2, 6) gives P = -100 + 1700
= 1600 The dots are the feasible solutions for
the revised Wyndor problem. Also
shown is the calculation of the total
4 net profit P (in dollars) for each corner
point from the net profits given in
(4, 3) gives P = 500 + 200
= 700 previous Table.
2

(0, 0) (4, 0) gives P = 500

gives P = 0
0 2 4 6 8 D
Production quantity for doors
35
“The Big M Transportation Problem”
Let D = Number of doors to produce,
W = Number of windows to produce,
y1 = 1 if perform setup to produce doors; 0 otherwise,
y2 = 1 if perform setup to produce windows; 0 otherwise.

Maximize P = 300D + 500W − 700y1 − 1,300y2

subject to
Original Constraints: Plant 1: D ≤ 4
Plant 2: 2W ≤ 12
Plant 3: 3D + 2W ≤ 18
Produce only if Setup: Doors: D ≤ My1 where M is a big enough
Windows: W ≤ My2 constant number, e.g., 99
in this problem.
and
D ≥ 0, W ≥ 0, y1 and y2 are binary. 36
Spreadsheet Model

37
Practice Problem Ch7-Q7:
Distribution Problem

38
Practice Problem Ch7-Q7:
Distribution Problem

39
Practice Problem Ch7-Q7:
Distribution Problem

40
Practice Problem Ch7-Q7:
Distribution Problem

41
Practice Problem Ch7-Q7:
Distribution Problem

42