Chapter 7—Goal Programming and Multiple Objective Optimization

MULTIPLE CHOICE

1. A constraint which cannot be violated is called a

a. binding constraint.
b. hard constraint.
c. definite constraint.
d. required constraint.
ANS: B PTS: 1

2. A constraint which represents a target value for a problem is called a

a. fuzzy constraint.
b. vague constraint.
c. preference constraint
d. soft constraint
ANS: D PTS: 1

3. Goal programming differs from linear programming or integer linear programming is that
a. goal programming provides for multiple objectives.
b. goal programming excludes hard constraints.
c. with goal programming we iterate until an acceptable solution is obtained.
d. goal programming requires fewer variables.
ANS: C PTS: 1

4. Decision-making problems which can be stated as a collection of desired objectives are known as what
type of problem?
a. A non-linear programming problem.
b. An unconstrained programming problem.
c. A goal programming problem.
d. An integer programming problem.
ANS: C PTS: 1

5. Which of the following is true regarding goal programming?

a. The objective function is not useful when comparing goal programming solutions.
b. We can place upper bounds on any of the deviation variables.
c. A preemptive goal program involves deviations with arbitrarily large weights.
d. All of these are true.
ANS: D PTS: 1

6. The RHS value of a goal constraint is referred to as the

a. target value.
b. constraint value.
c. objective value.
d. desired value.
ANS: A PTS: 1

7. The di+, di variables are referred to as

 a. objective variables.
b. goal variables.
c. target variables.
d. deviational variables.
ANS: D PTS: 1

8. Which of the following are true regarding weights assigned to deviational variables?
a. The weights assigned can be negative.
b. The weights assigned must sum to one.
c. The weight assigned to the deviation under a particular goal must be the same as the
weight assigned to the deviation above that particular goal.
d. All of these are false.
ANS: A PTS: 1

9. The di+ variable indicates the amount by which each goal's target value is
a. missed.
b. underachieved.
c. overachieved.
d. overstated.
ANS: C PTS: 1

10. Suppose that X1 equals 4. What are the values for d1+ and d1 in the following constraint?

X1 + d1 d1+ = 8

a. d1 = 4, d1+ =0
b. d1 = 0, d1+ =4
c. d1 = 4, d1+ =4
d. d1 = 8, d1+ =0
ANS: A PTS: 1

11. Suppose that the first goal in a GP problem is to make 3 X1 + 4 X2 approximately equal to 36. Using
the deviational variables d1 and d1+, what constraint can be used to express this goal?
a. 3 X1 + 4 X2 + d1  d1+  36
b. 3 X1 + 4 X2  d1  d1+ = 36
c. 3 X1 + 4 X2 + d1 + d1+ = 36
d. 3 X1 + 4 X2 + d1  d1+ = 36
ANS: D PTS: 1

12. Suppose that the first goal in a GP problem is to make 3 X1 + 4 X2 approximately equal to 36. Using
the deviational variables d1 and d1+, the following constraint can be used to express this goal.

3 X1 + 4 X2 + d1  d1+ = 36

If we obtain a solution where X1 = 6 and X2 = 2, what values do the deviational variables assume?
a. d1 = 0, d1+ = 10
b. d1 = 10, d1+ = 0
c. d1 = 5, d1+ = 5
d. d1 = 6, d1+ = 0
 ANS: B PTS: 1

13. What is the soft constraint form of the following hard constraint?

3X1 + 2 X2  10

a. 3X1 + 2 X2 + d1  d1+ = 10

b. 3X1 + 2 X2 + d1 + d1+ = 10
c. 3X1 + 2 X2  d1  d1+  10
d. 3X1 + 2 X2 + d1  d1+  10
ANS: A PTS: 1

14. Which of the following formulas is a deviation-minimizing objective function for a goal programming
problem?
a.

b.

c.

d.

ANS: D PTS: 1

15. What is the meaning of the ti term in this objective function for a goal programming problem?

a. The time required for each decision variable.

b. The percent of goal i met.
c. The coefficient for the ith decision variable
d. The target value for goal i.
ANS: D PTS: 1

16. What weight would be assigned to a neutral deviational variable?

a. 0
b. 1
c. 10
d. 100
ANS: A PTS: 1

17. Which of the following is false regarding a goal constraint?

a. A goal constraint allows us to determine how close a given solution comes to achieving a
goal.
b. A goal constraint will always contain two deviational variables.
c. Deviation variables are non-negative.
d. If two deviation variables are used in a constraint at least one will have a value of zero.
ANS: B PTS: 1
 Exhibit 7.1

The following questions are based on the problem below.

A company wants to advertise on TV and radio. The company wants to produce about 6 TV ads and
12 radio ads. Each TV ad costs $20,000 and is viewed by 10 million people. Radio ads cost $10,000
and are heard by 7 million people. The company wants to reach about 140 million people, and spend
about $200,000 for all the ads. The problem has been set up in the following Excel spreadsheet.

A B C D E
1 Problem Data TV Radio
2 Cost 20 10
3 Coverage 10 7
4
5 Goal Constraints TV Radio Cost Coverage
6 Actual Amount 0 0
7 +Under 0 0 0 0
8  Over 0 0 0 0
9 = Goal 0 0 0 0
10 Target Value 6 12 200 140
11
12 Percentage Deviation:
13 Under 1 1 1 1
14 Over 0 0 0 0
15
16 Weights
17 Under
18 Over
19
20 Objective 0

18. Refer to Exhibit 7.1. What formula goes in cell D6?

a. =SUMPRODUCT(B2:B3,B6:B7)
b. =B2*C2+B6*C6
c. =SUMPRODUCT(B2:C2,B10:C10)
d. =SUMPRODUCT(B2:C2,B6:C6)
ANS: D PTS: 1

19. Refer to Exhibit 7.1. What formula goes in cell B9?

a. =SUM(B6:B8)
b. =B6+B7-B8
c. =B6-B7+B8
d. =B10-B8
ANS: B PTS: 1

20. Refer to Exhibit 7.1. Which of the following is a constraint specified to Risk Solver Platform (RSP) for
this model?
a. $B$9:$E$9=$B$6:$E$6
b. $B$9:$E$9<$B$10:$E$10
c. $B$9:$E$9=$B$10:$E$10
d. $B$9:$E$9>$B$10:$E$10
ANS: C PTS: 1
21. Refer to Exhibit 7.1. Which cells are the variable cells in this model?
a. $B$6:$C$6, $B$7:$E$8
b. $B$6:$C$6
c. $B$9:$E$9
d. $B$6:$E$8
ANS: A PTS: 1

22. Refer to Exhibit 7.1. Which cell(s) is(are) the objective cell(s) in this model?
a. $B$20
b. $D$6
c. $E$6
d. $B$13:$E$14, $B$9:$E$9
ANS: A PTS: 1

23. Refer to Exhibit 7.1. If the company is very concerned about going over the $200,000 budget, which
cell value should change and how should it change?
a. D13, increase
b. D13, decrease
c. D14, increase
d. D14, decrease
ANS: C PTS: 1

24. A manager wants to ensure that he does not exceed his budget by more than $1000 in a goal
programming problem. If the budget constraint is the third constraint in the goal programming problem
which of the following formulas will best ensure that the manager's objective is met?
a. MIN d3+
b. d3  1000
c. d3+ = 1000
d. d3+  1000
ANS: D PTS: 1

25. An optimization technique useful for solving problems with more than one objective function is
a. dual programming.
b. sensitivity analysis.
c. multi-objective linear programming.
d. goal programming.
ANS: C PTS: 1

26. The MINIMAX objective

a. yields the smallest possible deviations.
b. minimizes the maximum deviation from any goal.
c. chooses the deviation which has the largest value.
d. maximizes the minimum value of goal attainment.
ANS: B PTS: 1

27. The primary benefit of a MINIMAX objective function is

a. it yields any feasible solution by changing the weights.
b. it is limited to all corner points.
c. it yields a larger variety of solutions than generally available using an LP method.
 d. it makes many of the deviational variables equal to zero.
ANS: C PTS: 1

28. The decision maker has expressed concern with Goal 1, budget, achievement. He indicated that future
candidate solutions should stay under budget. How can you modify your goal programming model to
accommodate this change?
a. Make budget a hard constraint in the model.
b. Give d1+ an extremely large weight in the objective function.
c. Remove d1+ from the goal constraint.
d. All of these.
ANS: D PTS: 1

29. MINIMAX solutions to multi-objective linear programming (MOLP) problems are

a. dually optimal.
b. Pareto optimal.
c. suboptimal.
d. maximally optimal.
ANS: B PTS: 1

30. If no other feasible solution to a multi-objective linear programming (MOLP) problem allows an
increase in any objective without decreasing at least one other objective, the solution is said to be
a. dually optimal.
b. Pareto optimal.
c. suboptimal.
d. maximally optimal.
ANS: B PTS: 1

Exhibit 7.2

The following questions are based on the problem below.

An investor has $150,000 to invest in investments A and B. Investment A requires a $10,000 minimum
investment, pays a return of 12% and has a risk factor of .50. Investment B requires a $15,000
minimum investment, pays a return of 10% and has a risk factor of .20. The investor wants to
maximize the return while minimizing the risk of the portfolio. The following multi-objective linear
programming (MOLP) has been solved in Excel.

A B C D
1 Problem data A B
2 Expected return 12% 10%
3 Risk rating 0.50 0.20
4
5 Variables A B Total
6 Amount invested 0 0 0
7 Minimum required $10,000 $15,000 $150,000
8
9 Objectives:
10 Average return 0
11 Average risk 0

31. Refer to Exhibit 7.2. What formula goes in cell B10?

 a. =SUMPRODUCT(B2:C2,$B$6:$C$6)/$D$7
b. =B2*C2+B3*C3
c. =SUMPRODUCT(B3:C3,$B$6:$C$6)/$D$7
d. =SUMPRODUCT(B2:C2,$B$6:$C$6)
ANS: A PTS: 1

32. Refer to Exhibit 7.2. What formula goes in cell B11?

a. =SUMPRODUCT(B2:C2,$B$6:$C$6)/$D$7
b. =B2*C2+B3*C3
c. =SUMPRODUCT(B3:C3,$B$6:$C$6)/$D$7
d. =SUMPRODUCT(B3:C3,$B$6:$C$6)
ANS: C PTS: 1

33. Refer to Exhibit 7.2. What Risk Solver Platform (RSP) constraint involves cells $B$6:$C$6?
a. $B$6:$C$6=$B$7:$C$7
b. $B$6:$C$6$B$7:$C$7
c. $B$6:$C$6$B$7:$C$7
d. $B$6:$C$6=$D$7
ANS: B PTS: 1

34. Refer to Exhibit 7.2. Which cells are the changing cells in this model?
a. $B$6:$C$6, $B$10:$B$11
b. $B$6:$C$6
c. $B$6:$D$6
d. $B$10:$B$11
ANS: B PTS: 1

35. Refer to Exhibit 7.2. Which cell(s) is(are) the target cells in this model?
a. $B$6:$C$6, $B$10:$B$11
b. $B$6:$C$6
c. $B$6:$D$6
d. $B$10:$B$11
ANS: D PTS: 1

Exhibit 7.3

The following questions are based on the problem below.

An investor has $150,000 to invest in investments A and B. Investment A requires a $10,000 minimum
investment, pays a return of 12% and has a risk factor of .50. Investment B requires a $15,000
minimum investment, pays a return of 10% and has a risk factor of .20. The investor wants to
maximize the return while minimizing the risk of the portfolio. The following minimax formulation of
the problem has been solved in Excel.

A B C D E
1 Problem data A B
2 Expected return 12% 10%
3 Risk rating 0.50 0.20
4
5 Variables A B Total
 6 Amount invested 0 0 0
7 Minimum required $10,000 $15,000 $150,000
8
9 Weighted
10 Goals Actual Target Weights % Deviation
11 Average return 0 11.8% 1 0
12 Average risk 0 0.22 1 0
13
14 Objective: 0

36. Refer to Exhibit 7.3. What formula goes in cell E11?

a. =D11*(C11B11)/C11
b. =(C11B11)/C11
c. =D11*C11
d. =D11*(C11B11)
ANS: A PTS: 1

37. Refer to Exhibit 7.3. Which value should the investor change, and in what direction, if he wants to
reduce the risk of the portfolio?
a. D11, increase
b. D12, increase
c. C12, increase
d. D12, decrease
ANS: B PTS: 1

38. Goal programming solution feedback indicates that the d4+ level of 50 should not be exceeded in future
solution iterations. How should you modify your goal constraint

40 X1 + 20 X2 + d4 + d4+ = 300

to accommodate this requirement?

a. Increase the RHS value from 300 to 350.
b. Replace the constraint with 40 X1 + 20 X2  350.
c. Do not modify the constraint, add a constraint d4+  50.
d. Do not modify the constraint, add a constraint d4+ = 50.
ANS: C PTS: 1

PROBLEM

39. Given the following goal constraints

5 X1 + 6 X2 + 7 X3 + d1  d1+ = 87
3 X1 + X2 + 4 X3 + d2  d2+ = 37
7 X1 + 3 X2 + 2 X3 + d3  d3+ = 72

and solution (X1, X2, X3) = (7, 2, 5), what values do the deviational variables assume?

ANS:
d1 = 5, d2+ = 6, d3 = 7, all others equal to zero.
 PTS: 1

40. Consider the following MOLP:

MAX: 3 X1 + 4 X2
MAX: 2 X1 + X2
Subject to: 6 X1 + 13 X2  78
12 X1 + 9 X2  108
8 X1 + 10 X2  80
X1, X2  0

Graph the feasible region for this problem and compute the value of each objective at each extreme
point. What are the solutions to each of the component LPs?

ANS:
 X1 X2 OBJ 1 OBJ 2
0 0 0 0
0 6 24 6
9 0 27 18
5.9 3.27 30.78 15.07
7.5 2 30.5 17

The solution to MAX 3 X1 + 4 X2 is (X1, X2) = (5.9, 3.27) and objective function value of 30.78.
The solution to MAX 2 X1 + X2 is (X1, X2) = (9, 0) and objective function value of 18.

PTS: 1

41. A company makes 2 products A and B from 2 resources, labor and material. The products have the
following resource requirements and produce the accompanying profits. The available quantity of
resources is also shown in the table.

Product A B Available resources

Labor (hr/unit) 3 2 150
Material (ounces/unit) 1 2 200
Profit($/unit) 7 6

Management has developed the following set of goals

Goal 1: Produce approximately 40 units of product 1.

Goal 2: Produce approximately 70 units of product 2.
Goal 3: Achieve a profit over $400.
Goal 4: Consume less than 150 hours of labor
Goal 5: Consume less than 200 ounces of material

Formulate a goal programming model of this problem.

ANS:
Let X1 = number of product 1
X2 = number of product 2

MIN: d1 + d1+ + d2  + d2+ + d3 + d4+ + d5+

Subject to: X1 + d1  d1+ = 40 product 1
X2 + d2  d2+ = 70 product 2
7 X1 + 6 X2 + d3  d3+ = 400 profit
3 X1 + 2 X2 + d4  d4+ = 150 labor
1 X1 + 2 X2 + d5  d5+ = 200 material
Xi, di, di+  0 for all i

PTS: 1

42. A company makes 2 products A and B from 2 resources. The products have the following resource
requirements and produce the accompanying profits. The available quantity of resources is also shown
in the table.

Product 1 2 Available resources

Labor (hr/unit) 3 2 150
Material (ounces/unit) 1 2 200
Profit($/unit) 7 6

Management has developed the following set of goals

Goal 1: Produce approximately 40 units of product 1.

Goal 2: Produce approximately 70 units of product 2.
Goal 3: Achieve a profit over $400.
Goal 4: Consume less than 150 hours of labor
Goal 5: Consume less than 200 ounces of material

Based on this GP formulation of the problem and the associated optimal integer solution what values
should go in cells B2:F16 of the following Excel spreadsheet?

Let X1 = number of product 1

X2 = number of product 2

MIN: d1 + d1+ + d2 + d2+ + d3 + d4+ + d5+

Subject to: X1 + d 1   d 1 + = 40 product 1
X2 + d 2   d 2 + = 70 product 2
7 X1 + 6 X2 + d3  d3+ = 400 profit
3 X1 + 2 X2 + d4  d4+ = 150 labor
1 X1 + 2 X2 + d5  d5+ = 200 material
Xi, di, di+  0 for all i

(X1, X2) = (4, 69)

A B C D E F
1 Problem Data A B
2 Labor
3 Material
4 Profit
5 Goal Constraints A B Labor Material Profit
6 Actual Amount
7 + Under
8  Over
9 = Goal
10 Target Value
11
12 Weights
13 Under
14 Over
15
16 Objective

ANS:

A B C D E F
1 Problem Data A B
2 Labor 3 2
3 Material 1 2
4 Profit 7 6
5 Goal Constraints A B Labor Material Profit
6 Actual Amount 4 69 150 142 442
7 + Under 36 1 0 58 0
 8  Over 0 0 0 0 42
9 = Goal 40 70 150 200 400
10 Target Value 40 70 150 200 400
11
12 Weights
13 Under 1 1 0 0 1
14 Over 1 1 1 1 0
15
16 Objective 37

PTS: 1

43. A company makes 2 products A and B from 2 resources. The products have the following resource
requirements and produce the accompanying profits. The available quantity of resources is also shown
in the table.

Product 1 2 Available resources

Labor (hr/unit) 3 2 150
Material (ounces/unit) 1 2 200
Profit($/unit) 7 6

Management has developed the following set of goals

Goal 1: Produce approximately 40 units of product 1.

Goal 2: Produce approximately 70 units of product 2.
Goal 3: Achieve a profit over $400.
Goal 4: Consume less than 150 hours of labor
Goal 5: Consume less than 200 ounces of material

Based on the following GP formulation of the problem, and the associated optimal solution, what
formulas should go in cells D6:F6, B9:F9, and B16 of the following Excel spreadsheet? NOTE:
Formulas are not required in all of these cells.

Let X1 = number of product 1

X2 = number of product 2

MIN: d1 + d1+ + d2 + d2+ + d3 + d4+ + d5+

Subject to: X1 + d1  d1+ = 40 product 1
X2 + d2  d2+ = 70 product 2
7 X1 + 6 X2 + d3  d3+ = 400 profit
3 X1 + 2 X2 + d4  d4+ = 150 labor
1 X1 + 2 X2 + d5  d5+ = 200 material
Xi, di, di+  0 for all i

(X1, X2) = (4, 69)

A B C D E F
1 Problem Data A B
2 Labor 3 2
3 Material 1 2
4 Profit 7 6
5 Goal Constraints A B Labor Material Profit
6 Actual Amount 4 69 150 142 442
 7 + Under 36 1 0 58 0
8  Over 0 0 0 0 42
9 = Goal 40 70 150 200 400
10 Target Value 40 70 150 200 400
11
12 Weights
13 Under 1 1 0 0 1
14 Over 1 1 1 1 0
15
16 Objective 37

ANS:
Cell Formula Copied to
D6 =SUMPRODUCT(B2:C2,$B$6:$C$6)
E6 =SUMPRODUCT(B3:C3, $B$6:$C$6)
F6 =SUMPRODUCT(B4:C4, $B$6:$C$6)
B9 =B6+B7-B8 C9:F9
B16 =SUMPRODUCT(B7:F8,B13:F14)

PTS: 1

44. A company wants to purchase large and small delivery trucks. The company wants to purchase about
10 large and 15 small trucks. Each large truck costs $30,000 and has a 10 ton capacity. Each small
truck costs $20,000 and has a 7 ton capacity. The company wants to have about 200 tons of capacity
and spend about $600,000.

Formulate a goal programming model of this problem.

ANS:
Let X1 = number of large trucks
X2 = number of small trucks

MIN: d1 + d1+ + d2 + d2+ + d3 + d3+ + d4  d4+

Subject to: X1 + d1  d1+ = 10 large trucks
X2 + d2  d2+ = 15 small trucks
10 X1 + 7 X2 + d3  d3+ = 200 capacity
30 X1 + 20 X2 + d4  d4+ = 600 cost
Xi, di, di+  0 for all i

PTS: 1

45. A company wants to purchase large and small delivery trucks. The company wants to purchase about
10 large and 15 small trucks. Each large truck costs $30,000 and has a 10 ton capacity. Each small
truck costs $20,000 and has a 7 ton capacity. The company wants to have about 200 tons of capacity
and spend about $600,000.

Based on the following formulation and associated integer solution, what values should go in cells
B2:E16 of the spreadsheet?

Let X1 = number of large trucks

X2 = number of small trucks

MIN: d1 + d1+ + d2 + d2+ + d3 + d3+ + d4  d4+

 Subject to: X1 + d1  d1+ = 10 large trucks
X2 + d2  d2+ = 15 small trucks
10 X1 + 7 X2 + d3  d3+ = 200 capacity
30 X1 + 20 X2 + d4  d4+ = 600 cost
Xi, di, di+  0 for all I

(X1, X2) = (10, 15)

A B C D E
1 Problem Data Large Small
2 Cost
3 Capacity
4
5 Goal Constraints Large Small Cost Capacity
6 Actual Amount
7 + Under
8  Over
9 = Goal
10 Target Value
11
12 Weights
13 Under
14 Over
15
16 Objective

ANS:

A B C D E
1 Problem Data Large Small
2 Cost 10 7
3 Capacity 30 20
4
5 Goal Constraints Large Small Cost Capacity
6 Actual Amount 10 15 205 600
7 + Under 0 0 0 0
8  Over 0 0 5 0
9 = Goal 10 15 200 600
10 Target Value 10 15 200 600
11
12 Weights
13 Under 1 1 1 1
14 Over 1 1 1 1
15
16 Objective 5

PTS: 1

46. A company wants to purchase large and small delivery trucks. The company wants to purchase about
10 large and 15 small trucks. Each large truck costs $30,000 and has a 10 ton capacity. Each small
truck costs $20,000 and has a 7 ton capacity. The company wants to have about 200 tons of capacity
and spend about $600,000.
 Based on the following goal programming formulation, associated solution, and spreadsheet model,
what formulas should go in cells D6:E6, B9:E9, and B16 of the spreadsheet?

Let X1 = number of large trucks

X2 = number of small trucks

MIN: d1 + d1+ + d2 + d2+ + d3 + d3+ + d4  d4+

Subject to: X1 + d1  d1+ = 10 large trucks
X2 + d2  d2+ = 15 small trucks
10 X1 + 7 X2 + d3  d3+ = 200 capacity
30 X1 + 20 X2 + d4  d4+ = 600 cost
Xi, di, di+  0 for all I

(X1, X2) = (10, 15)

A B C D E
1 Problem Data Large Small
2 Cost 10 7
3 Capacity 30 20
4
5 Goal Constraints Large Small Cost Capacity
6 Actual Amount 10 15 205 600
7 + Under 0 0 0 0
8  Over 0 0 5 0
9 = Goal 10 15 200 600
10 Target Value 10 15 200 600
11
12 Weights
13 Under 1 1 1 1
14 Over 1 1 1 1
15
16 Objective 5

ANS:
Cell Formula Copied to
D6 =SUMPRODUCT(B2:C2,$B$6:$C$6)
E6 =SUMPRODUCT(B3:C3, $B$6:$C$6)
B9 =B6+B7-B8 C9:E9
B16 =SUMPRODUCT(B13:E14,B7:E8)

PTS: 1

47. A dietician wants to formulate a low cost, high calorie food product for a customer. The following
information is available about the 2 ingredients which can be combined to make the food. The
customer wants 1000 pounds of the food product and it should contain 250 pounds of Food 1 and 300
pounds of Food 2. The final cost of the blend should be about $1.15 and contain about 2500 calories
per pound. The percent of fat, protein, carbohydrate in each food is summarized below with the target
values for the goals. The dietician would prefer the food product be low in fat while also high in
protein and carbohydrates.

Food 1 Food 2 TARGET

Cost ($/pound) $1.00 $1.25 $1.15
Fat 15% 25% 300 pounds
 Protein 35% 40% 370 pounds
Carbohydrate 50% 35% 400 pounds
Calories/pound 3000 2000 2500
Pounds of food 1 250
Pounds of food 2 300

Formulate the GP for this problem

ANS:
Let X1 = pounds of food 1
X2 = pounds of food 2

MIN: d1 + d1+ + d2+ + d3 + d4 + d5 + d5+ + d6 + d6+ + d7 + d7+
Subject to: 1 X1 + 1.25 X2 + d1  d1+ = 1.15 cost

.15 X1 + .25 X2 + d2  d2 = 300
+
fat
.35 X1 + .40 X2 + d3  d3+ = 370 protein

.50 X1 + .35 X2 + d4  d4 = 400
+
carbohydrate

3000 X1 + 2000 X2 + d5  d5 = 2500
+
calories
X1 + d6  d6+ = 250 food 1

X2 + d7  d7 = 300
+
food 2
X1 + X2 = 1000 food product required
Xi, di, di+  0 for all i

PTS: 1

48. A dietician wants to formulate a low cost, high calorie food product for a customer. The following
information is available about the 2 ingredients which can be combined to make the food. The
customer wants 1000 pounds of the food product and it must contain at least 250 pounds of Food 1 and
300 pounds of Food 2.

Food 1 Food 2 Requirements

Cost ($/pound) 1.00 1.25 Minimize
% Fat 15 25 Less than 30%
% Protein 35 40 At least 37%
% Carbohydrate 50 35 At least 40%
Calories/pound 3000 2000 Maximize
Minimum pounds of food 250 300

Formulate the MOLP for this problem.

ANS:
Let X1 = pounds of food 1
X2 = pounds of food 2

MIN: 1 X1 + 1.25 X2 minimize cost or food product

MAX: 3000 X1 + 2000 X2 maximize the calories
Subject to: .15 X1 + .25 X2  300 fat
.35 X1 + .40 X2  370 protein
.50 X1 + .35 X2  400 carbohydrate
X1  250 food 1
X2  300 food 2
X1 + X2 = 1000 total food product required
 Xi  0 for all i

PTS: 1

49. A company needs to supply customers in 3 cities from its 3 warehouses. The supplies, demands and
shipping costs are shown below.

Destination
Warehouse 1 2 3 Supply
1 34 60 36 400
2 70 40 50 300
3 56 40 32 200
Demand 500 300 200

The company has identified the following goals:

Goal 1: The company would like to come as close as possible to satisfying its customers
demand.
Goal 2: It would also like to ensure that the cost is approximately $290,000.

Formulate a goal programming model of this problem.

ANS:
Let Xij = number of units shipped from warehouse i to destination j

MIN:
Subject to:
X11 + X12 + X13 + d 1   d 1 + = 400
X21 + X22 + X23 + d 2   d 2 + = 300
X31 + X32 + X33 + d 3   d 3 + = 200
X11 + X21 + X31 + d 4   d 4 + = 500
X12 + X22 + X32 + d 5   d 5 + = 300
X13 + X23 + X33 + d 6   d 6 + = 200
34 X11 + 60 X12 + 36 X13 + 70 X21 + 40 X22 + 50 X23 + 56 X31 + 40 X32 + 32 X33 +
d7 + d7+ = 290000
Xij, di, di+  0 for all i,j

PTS: 1

50. An investor wants to invest $50,000 in two mutual funds, A and B. The rates of return, risks and
minimum investment requirements for each fund are:

Fund Rate of return Risk Minimum investment

A 12% 0.5 $20,000
B 9% 0.3 $10,000

Note that a low Risk rating means a less risky investment. The investor can invest to maximize the
expected rate of return or minimize risk. Any money beyond the minimum investment requirements
can be invested in either fund.

Formulate the MOLP for this investor.

 ANS:
Let X1 = dollars in investment A
X2 = dollars in investment B

MAX: 0.12 X1/50000 + 0.09 X2/50000

MIN: 0.5 X1/50000 + 0.3 X2/50000
Subject to: X1 + X2 = 50000
X1  20000
X2  10000
Xi  0 for all i

PTS: 1

51. An investor wants to invest $50,000 in two mutual funds, A and B. The rates of return, risks and
minimum investment requirements for each fund are:

Fund Rate of return Risk Minimum investment

A 12% .5 $20,000
B 9% .3 $10,000

Note that a low Risk rating means a less risky investment. The investor can invest to maximize the
expected rate of return or minimize risk. Any money beyond the minimum investment requirements
can be invested in either fund.

The following is the MOLP formulation for this problem:

Let X1 = dollars in investment A

X2 = dollars in investment B

MAX: 0.12 X1/50000 + 0.09 X2/50000

MIN: 0.5 X1/50000 + 0.3 X2/50000
Subject to: X1 + X2 = 50000
X1  20000
X2  10000
Xi  0 for all i

The solution for the second LP is (X1, X2) = (20,000, 30,000).

Based on this solution, what values should go in cells B2:D11 of the spreadsheet?

A B C D
1 Problem data A B
2 Expected return
3 Risk rating
4
5 Variables A B Total
6 Amount invested
7 Minimum required
8
9 Objectives:
10 Average return
11 Average risk
 ANS:

A B C D
1 Problem data A B
2 Expected return 12% 9%
3 Risk rating 0.50 0.20
4
5 Variables A B Total
6 Amount invested $20,000 $30,000 $50,000
7 Minimum required $20,000 $10,000 $50,000
8
9 Objectives:
10 Average return 10.2%
11 Average risk 0.32

PTS: 1

52. An investor wants to invest $50,000 in two mutual funds, A and B. The rates of return, risks and
minimum investment requirements for each fund are:

Fund Rate of return Risk Minimum investment

A 12% 0.5 $20,000
B 9% 0.3 $10,000

Note that a low Risk rating means a less risky investment. The investor can invest to maximize the
expected rate of return or minimize risk. Any money beyond the minimum investment requirements
can be invested in either fund.

The following is the MOLP formulation for this problem:

Let X1 = dollars in investment A

X2 = dollars in investment B

MAX: 0.12 X1/50000 + 0.09 X2/50000

MIN: 0.5 X1/50000 + 0.3 X2/50000
Subject to: X1 + X2 = 50000
X1  20000
X2  10000
Xi  0 for all i

The solution for the second LP is (X1, X2) = (20,000, 30,000).

What formulas should go in cells B2:D11 of the spreadsheet? NOTE: Formulas are not required in all
of these cells.

A B C D
1 Problem data A B
2 Expected return 12% 9%
3 Risk rating 0.50 0.20
4
5 Variables A B Total
6 Amount invested $20,000 $30,000 $50,000
7 Minimum required $20,000 $10,000 $50,000
8
 9 Objectives:
10 Average return 10.2%
11 Average risk 0.32

ANS:
Cell Formula
D6 =B6+C6
B10 =SUMPRODUCT(B2:C2,B6:C6)/D7
B11 =SUMPRODUCT(B3:C3,B6:C6)/D7

PTS: 1

53. An investor wants to invest $50,000 in two mutual funds, A and B. The rates of return, risks and
minimum investment requirements for each fund are:

Fund Rate of return Risk Minimum investment

A 12% 0.5 $20,000
B 9% 0.3 $10,000

Note that a low Risk rating means a less risky investment. The investor wants to maximize the
expected rate of return while minimizing his risk. Any money beyond the minimum investment
requirements can be invested in either fund. The investor has found that the maximum possible
expected rate of return is 11.4% and the minimum possible risk is 0.32.

Formulate a goal programming model with a MINIMAX objective function.

ANS:
Let X1 = dollars in investment A
X2 = dollars in investment B

MINIMIZE Q
Subject to: X1 + X2 = 50000
X1  20000
X2  10000

Xi  0 for all i, Q  0

PTS: 1
54. An investor wants to invest $50,000 in two mutual funds, A and B. The rates of return, risks and
minimum investment requirements for each fund are:

Fund Rate of return Risk Minimum investment

A 12% 0.5 $20,000
B 9% 0.3 $10,000

Note that a low Risk rating means a less risky investment. The investor wants to maximize the
expected rate of return while minimizing his risk. Any money beyond the minimum investment
requirements can be invested in either fund. The investor has found that the maximum possible
expected rate of return is 11.4% and the minimum possible risk is 0.32.

The following Excel spreadsheet has been created to solve a goal programming problem with a
MINIMAX objective based on the following goal programming formulation with MINIMAX
objective and corresponding solution.

MINIMIZE Q
Subject to: X1 + X2 = 50000
X1  20000
X2  10000

Xi  0 for all i, Q  0

with solution (X1, X2) = (15,370, 34,630).

What values should go in cells B2:D14 of the spreadsheet?

A B C D E
1 Problem data A B
2 Expected return
3 Risk rating
4
5 Variables A B Total
6 Amount invested
7 Minimum required
8
9 Weighted
10 Goals Actual Target Weights % Deviation
11 Average return 1
12 Average risk 1
 13
14 Objective:

ANS:

A B C D E
1 Problem data A B
2 Expected return 0.12 0.09
3 Risk rating 0.5 0.3
4
5 Variables A B Total
6 Amount invested 15370 34630 50000
7 Minimum required 10000 20000 50000
8
9 Weighted
10 Goals Actual Target Weights % Deviation
11 Average return 0.09922 0.114 1 0.12963
12 Average risk 0.36148 0.320 1 0.12963
13
14 Objective: 0.12963

PTS: 1

55. Robert Gardner runs a small, local-only delivery service. His fleet consists of three smaller panel
trucks. He recently accepted a contract to deliver 12 shipping boxes of goods for delivery to 12
different customers. The box weights are: 210, 160, 320, 90, 110, 70, 410, 260, 170, 240, 80 and 180
for boxes 1 through 12, respectively. Since each truck differs each truck has different load capacities as
given below:

Truck Weight Capacity Box Capacity Cost per pound

1 800 pounds 5 $0.34
2 900 pounds 6 $0.42
3 700 pounds 4 $0.25

Robert would like each truck equally loaded, both in terms of number of boxes and in terms of total
weight, while minimizing his shipping costs. Assume a cost of $50 per item for trucks carrying extra
boxes and $0.10 per pound cost for trucks carrying less weight.

Formulate the integer goal programming problem for Robert. (Hint: objective function involves
decision and deviation variables.)

ANS:
Y1 = weight loaded in truck 1; Y2 = weight loaded in truck 2; Y3 = weight loaded in truck 3;
Xi,j = 0 if truck i not loaded with box j; 1 if truck i loaded with box j.

MIN
0.34Y1 + 0.42Y2 + 0.25Y3 + 50*(d1+ + d2+ + d3+) + 0.10*(d4 + d5 + d6)*100
Subject to:
Y1 = 210X11 + 160X12 + 320X13 + 90X14 + 110X15 + 70X16 + 410X17 + 260X18
+ 170X19 + 240X1,10 + 80X1,11 + 180X1,12
Y2 = 210X21 + 160X22 + 320X23 + 90X24 + 110X25 + 70X26 + 410X27 + 260X28
+ 170X29 + 240X2,10 + 80X2,11 + 180X2,12
Y3 = 210X31 + 160X32 + 320X33 + 90X34 + 110X35 + 70X36 + 410X37 + 260X38
+ 170X39 + 240X3,10 + 80X3,11 + 180X3,12
Y1  800
Y2  900
Y3  700
Y1 + d1  d1+ = 767
Y2 + d2  d2+ = 767
Y3 + d3  d3+ = 767
X11 + X21 + X31 = 1
X12 + X22 + X32 = 1
X13 + X23 + X33 = 1
X14 + X24 + X34 = 1
X15 + X25 + X35 = 1
X16 + X26 + X36 = 1
X17 + X27 + X37 = 1
X18 + X28 + X38 = 1
X19 + X29 + X39 = 1
X1,10 + X2,10 + X3,10 = 1
X1,11 + X2,11 + X3,11 = 1
X1,12 + X2,12 + X3,12 = 1
X11 + X12 + X13 + X14 + X15 + X16 + X17 + X18 + X19
+ X1,10 + X1,11 + X1,12 + d4  d4+ = 4
X21 + X22 + X23 + X24 + X25 + X26 + X27 + X28 + X29
+ X2,10 + X2,11 + X2,12 + d5  d5+ = 4
X31 + X32 + X33 + X34 + X35 + X36 + X37 + X38 + X39
+ X3,10 + X3,11 + X3,12 + d6  d6+ = 4
Xij  0, dk, dk+  0 for k = 1,2,3,4,5,6

PTS: 1

Exhibit 7.4

The following questions are based on the problem below.

Robert Gardner runs a small, local-only delivery service. His fleet consists of three smaller panel
trucks. He recently accepted a contract to deliver 12 shipping boxes of goods for delivery to 12
different customers. The box weights are: 210, 160, 320, 90, 110, 70, 410, 260, 170, 240, 80 and 180
for boxes 1 through 12, respectively. Since each truck differs each truck has different load capacities as
given below:

Truck Weight Capacity Box Capacity Cost per pound

1 800 pounds 5 $0.34
2 900 pounds 6 $0.42
3 700 pounds 4 $0.25

Robert would like each truck equally loaded, both in terms of number of boxes and in terms of total
weight, while minimizing his shipping costs. Assume a cost of $50 per item for trucks carrying extra
boxes and $0.10 per pound cost for trucks carrying less weight.

The following integer goal programming formulation applies to his problem.

Y1 = weight loaded in truck 1; Y2 = weight loaded in truck 2; Y3 = weight loaded in truck 3;

Xi,j = 0 if truck i not loaded with box j; 1 if truck i loaded with box j.

MIN
0.34Y1 + 0.42Y2 + 0.25Y3 + 50*(d1+ + d2+ + d3+) + 0.10*(d4 + d5 + d6)*100
Subject to:
Y1 = 210X11 + 160X12 + 320X13 + 90X14 + 110X15 + 70X16 + 410X17 + 260X18
+ 170X19 + 240X1,10 + 80X1,11 + 180X1,12
Y2 = 210X21 + 160X22 + 320X23 + 90X24 + 110X25 + 70X26 + 410X27 + 260X28
+ 170X29 + 240X2,10 + 80X2,11 + 180X2,12
Y3 = 210X31 + 160X32 + 320X33 + 90X34 + 110X35 + 70X36 + 410X37 + 260X38
+ 170X39 + 240X3,10 + 80X3,11 + 180X3,12
Y1  800
Y2  900
Y3  700
Y1 + d1  d1+ = 767
Y2 + d2  d2+ = 767
Y3 + d3  d3+ = 767
X11 + X21 + X31 = 1
X12 + X22 + X32 = 1
X13 + X23 + X33 = 1
X14 + X24 + X34 = 1
X15 + X25 + X35 = 1
X16 + X26 + X36 = 1
X17 + X27 + X37 = 1
X18 + X28 + X38 = 1
X19 + X29 + X39 = 1
X1,10 + X2,10 + X3,10 = 1
X1,11 + X2,11 + X3,11 = 1
X1,12 + X2,12 + X3,12 = 1
X11 + X12 + X13 + X14 + X15 + X16 + X17 + X18 + X19 + X1,10 + X1,11 + X1,12 + d4  d4+ = 4
X21 + X22 + X23 + X24 + X25 + X26 + X27 + X28 + X29 + X2,10 + X2,11 + X2,12 + d5  d5+ = 4
X31 + X32 + X33 + X34 + X35 + X36 + X37 + X38 + X39 + X3,10 + X3,11 + X3,12 + d6  d6+ = 4
Xij  0, dk, dk+  0 for k = 1,2,3,4,5,6

Given the following spreadsheet solution of this integer goal programming formulation, answer the
following questions.

A B C D E F G
1 Cost $0.34 $0.42 $0.25
2 Capacity 8.0 9.0 7.0
3 1 2 3 Weight Assigned Required
4 1 0 1 0 2.1 1 1
5 2 1 0 0 1.6 1 1
6 3 1 0 0 3.2 1 1
7 4 0 0 1 0.9 1 1
8 5 0 1 0 1.1 1 1
9 6 0 1 0 0.7 1 1
10 7 0 1 0 4.1 1 1
11 8 0 0 1 2.6 1 1
12 9 0 0 1 1.7 1 1
13 10 1 0 0 2.4 1 1
14 11 1 0 0 0.8 1 1
15 12 0 0 1 1.8 1 1
16 Total Items 4 4 4
17 Under 0 0 0 Item Cost $50
18 Over 0 0 0 Wt. Cost $0.25
19 Actual 4 4 4
20 Target Items 4 4 4
 21 Total Wt 8 8 7
22 Under 0.00 0.00 0.67
23 Over 0.33 0.33 0.00
24 Actual Wt 7.67 7.67 7.67
25 Target Wt 7.67 7.67 7.67
26 Cost $7.8967

56. Refer to Exhibit 7.4. Given the solution indicated in the spreadsheet, which trucks, if any, are under an
equal weight amount, and which trucks are over an equal weight amount?

ANS:
Trucks 1 and 2 are over the target weight by 33 pounds. Truck 3 is under the target weight by 67
pounds.

PTS: 1

57. Refer to Exhibit 7.4. The solution indicates Truck 3 is under the target weight by 67 pounds. What if
anything can be done to this model to provide a solution in which Truck 3 is closer to the target
weight?

ANS:
Nothing. Truck 3 is at its capacity of 700 pounds.

PTS: 1

58. Refer to Exhibit 7.4. The spreadsheet model has scaled all the weights from pounds into 100s pounds.
How does this scaling effect the solution obtained using the Risk Solver Platform (RSP)?

ANS:
The solution obtained is the same regardless of scaling. In terms of pounds, the total cost would be
multiplied by 100, to obtain $78,967.

PTS: 1

59. Refer to Exhibit 7.4. What formulas should go in cell E26 of the spreadsheet?

ANS:
=SUMPRODUCT(B21:D21, B1:D1) + F17*SUM(B18:D18) + F18*SUM(B22:D22)

PTS: 1

60. Refer to Exhibit 7.4. Based on the integer goal programming formulation, the associated solution, and
spreadsheet model, what formulas should go in cells B19:E19 and B24:E24 of the spreadsheet?

ANS:
Cell Formula Copied to
B19 =B16 + B17  B18 C19:E19
B24 =B21 + B22  B23 C24:E24

PTS: 1

PROJECT
61. Project 7.1  Selecting Day Care Facilities for Twin Rivers

You have been hired by the city of Twin Rivers to conduct analyses in support of their "Care for our
Children" program.

The city council of Twin Rivers recently passed a resolution to fund much needed child care facilities
for the city. A total of $850,000 has been made available to buy or lease child care sites throughout the
city. An initial survey of the requirements for city-supported facilities has already been conducted and
that data will form the basis of the siting and facility selection analysis process. That initial survey
indicated the need for facilities providing four categories of child care: infant care, pre-school care,
after school care, and drop-in care. The survey indicated the following minimum requirements for each
category of child care. Naturally, a goal of this program is not only to try and cover these minimum
requirements, but also to try and maximize the total number of child care slots created.

Category Requirements*
Infant care 200
Pre-school care 250
After-school care 325
Drop-in care 150
*all requirements are in terms of available slots

Various sites have been identified and these sites are either for lease or for sale. The leased sites
require no additional modification at city expense. All modification expenses are borne by the owner.
The sites for sale will require some level of renovation and these renovation costs have been rolled into
the purchase cost for the site. Additionally, some types of maintenance on leased sites will be borne by
the owner reducing the expected maintenance costs the city must cover. The sites and the pertinent
information about those sites is summarized in the table below. Note that all costs are in $1000s. Each
site was also assessed regarding the relative "safety" of the site. Safety issues included such things as
parking, condition of the grounds, and condition of the structure. A safety index was used with each
site rated as 1 for acceptably safe, 2 for marginally safe, and 3 for marginally safe given some
modifications. Any sites surveyed found unacceptable in the safety department were dropped from
further consideration.

Child Slots Supported

Safety
Infant Pre-school After Drop-in Type of Cost of Maint. Index of
Facility Care Care school care Care Facility Facility Cost Facility
1 50 -- 75 50 RENT $145.00 $11.98 1
2 40 -- 75 60 RENT $146.00 $12.03 2
3 -- 70 75 30 RENT $139.50 $11.71 2
4 75 40 -- 30 RENT $136.50 $11.56 1
5 50 -- -- 75 BUY $125.00 $26.98 1
6 -- 90 -- 75 BUY $156.50 $60.56 3
7 25 -- 80 40 BUY $115.50 $42.51 2
8 -- 85 100 -- RENT $137.25 $11.59 2
9 85 50 75 -- BUY $169.00 $61.18 3
10 100 75 75 -- BUY $203.75 $30.92 1
11 60 -- -- 75 RENT $134.00 $11.43 3
12 40 40 40 40 RENT $139.00 $11.68 2
13 -- 85 -- 65 BUY $142.25 $43.84 2
14 75 85 75 -- BUY $189.75 $30.22 1
15 65 -- 75 30 BUY $138.50 $43.66 2
These new facilities will require staffing. The city council would like insight into the number of new
staff required and the subsequent annual personnel budget to accommodate these new positions.
Medical, administration, and of course providers are required for the facilities. The desired level of
staffing is based on a projected child-to-staff ratio. For planning purposes, a Created Slot-to-Staff ratio
can be employed.

Ratios Refer to Child-to-Staff Ratio

Type of Annual Cost Overall Infant Pre-School After-School Drop-In
Personnel of Position Ratio Ratio Ratio Ratio Ratio
Medical $16,000 40:1
Administration $25,000 80:1
Providers $13,000 3:1 6:1 15:1 7:1
Maintainers** $21,000
** One maintainer per facility.

Although the city council has agreed to fund this project for approximately $850,000, reasonable
targets for maintenance and personnel costs, and safety index levels must be established. The city
council members would not only like recommendations on site selection that account for all the goals
established for this project, they would like a summary breakout of the personnel requirements and
costs associated with the recommended options.

Requirements:

a. Formulate the MOLP appropriate for this problem.

b. Implement and solve the MOLP using the Risk Solver Platform (RSP).
c. Formulate the GP appropriate for this problem.
d. Implement and solve the GP using the Risk Solver Platform (RSP).
e. Report your initial recommendation and a summary of the requirements for that
recommendation.
f. Consider other issues that might arise in the current situation, adjust your model and
report the results accordingly.

ANS:
a. Formulate the MOLP appropriate for this problem.

Define: Xi = 1 if site selected, 0 otherwise, for i = 1,...,15

Yj = Number of slots of type j created, for j = 1,2,3,4

Determine minimum safety index

Minimize
X1 + 2X2 + 2X3 + X4 + X5 + 3X6 + 2X7 + 2X8 + 3X9 + X10 + 3X11 + 2X12 + 2X13
+ X14 + 2X15

Determine minimum maintenance costs

Minimize
11.98X1 + 12.03X2 + 11.71X3 + 11.56X4 + 26.98X5 + 60.56X6 + 42.51X7 + 11.59X8
+ 61.18X9 + 30.92X10 + 11.43X11 + 11.68X12 + 43.84X13 + 30.22X14 + 43.66X15

Determine the maximum number of slots created

Maximize Y1 + Y2 + Y3 + Y4

Subject to:
Infant child care slots required:
50X1 + 40X2 + 75X4 + 50X5 + 25X7 + 85X9 + 100X10 + 60X11 + 40X12 + 75X14
 + 65X15  200
Y1  (50X1 + 40X2 + 75X4 + 50X5 + 25X7 + 85X9 + 100X10 + 60X11 + 40X12
+ 75X14 + 65X15) = 0

Pre-School child care slots required:

70X3 + 40X4 + 90X6 + 85X8 + 50X9 + 75X10 + 40X12 + 85X13 + 85X4  250
Y2  (70X3 + 40X4 + 90X6 + 85X8 + 50X9 + 75X10 + 40X12 + 85X13 + 85X4) = 0

After-school child care slots required:

70X3 + 75X2 + 75X3 + 80X7 + 100X8 + 75X9 + 75X10 + 40X12 + 75X14
+ 75X15  325
Y3  (70X3 + 75X2 + 75X3 + 80X7 + 100X8 + 75X9 + 75X10 + 40X12 + 75X14
+ 75X15) = 0

Drop-in child care slots required:

50X1 + 60X2 + 30X3 + 30X4 + 75X5 + 75X6 + 40X7 + 75X11 + 40X12 + 65X13
+ 30X15  150
Y4  (50X1 + 60X2 + 30X3 + 30X4 + 75X5 + 75X6 + 40X7 + 75X11 + 40X12 + 65X13
+ 30X15) = 0

Budget constraint:
145X1 + 146X2 + 139.5X3 + 136.5X4 + 125X5 + 156.5X6 + 115.5X7 + 137.25X8
+ 169X9 + 203.75X10 + 134X11 + 139X12 + 142.25X13 + 189.75X14
+ 138.5X5  850

Xi binary

b. Implement and solve the MOLP using the Risk Solver Platform.

Minimum safety level is 9.

Minimum maintenance cost is 134.56.
Maximum child care slots created is 1025.
Minimum cost to meet minimum child care slot requirements is $781,000.

c. Formulate the GP appropriate for this problem.

Define:
Xi = 1 if site selected, 0 otherwise, for i = 1,...,15
Yj = Number of slots of type j created, for j = 1,2,3,4
Zk = Number of personnel slots created for k = 1,..., 6

Minimize:

Subject to:
Infant child care slots goal:
50X1 + 40X2 + 75X4 + 50X5 + 25X7 + 85X9 + 100X10 + 60X11 + 40X12 + 75X14
+ 65X15 + d1  d1+ = 200

Pre-School child care slots goal:

70X3 + 40X4 + 90X6 + 85X8 + 50X9 + 75X10 + 40X12 + 85X13 + 85X4
+ d2  d2+ = 250
After-school child care slots goal:
70X3 + 75X2 + 75X3 + 80X7 + 100X8 + 75X9 + 75X10 + 40X12 + 75X14 + 75X15
+ d3  d3+ = 325

Drop-in child care slots goal:

50X1 + 60X2 + 30X3 + 30X4 + 75X5 + 75X6 + 40X7 + 75X11 + 40X12 + 65X13
+ 30X15 + d4  d4+ = 150

Budget goal:
145X1 + 146X2 + 139.5X3 + 136.5X4 + 125X5 + 156.5X6 + 115.5X7 + 137.25X8
+ 169X9 + 203.75X10 + 134X11 + 139X12 + 142.25X13 + 189.75X14
+ 138.5X5 + d5  d5+ = 850

Compute total child-care slots created for each category.

Y1  (50X1 + 40X2 + 75X4 + 50X5 + 25X7 + 85X9 + 100X10 + 60X11 + 40X12
+ 75X14 + 65X15) = 0
Y2  (70X3 + 40X4 + 90X6 + 85X8 + 50X9 + 75X10 + 40X12 + 85X13 + 85X4) = 0
Y3  (70X3 + 75X2 + 75X3 + 80X7 + 100X8 + 75X9 + 75X10 + 40X12 + 75X14
+ 75X15) = 0
Y4  (50X1 + 60X2 + 30X3 + 30X4 + 75X5 + 75X6 + 40X7 + 75X11 + 40X12
+ 65X13 + 30X15) = 0

Safety index goal

X1 + 2X2 + 2X3 + X4 + X5 + 3X6 + 2X7 + 2X8 + 3X9 + X10 + 3X11 + 2X12
+ 2X13 + X14 + 2X15 + d 6   d 6 + = 9

Maintenance cost goal

11.98X1 + 12.03X2 + 11.71X3 + 11.56X4 + 26.98X5 + 60.56X6 + 42.51X7
+ 11.59X8 + 61.18X9 + 30.92X10 + 11.43X11 + 11.68X12 + 43.84X13
+ 30.22X14 + 43.66X15 + d7  d7+ = 134.56

Number of slots created goal

Y1 + Y2 + Y3 + Y4 + d8  d8+ = 1025

Child-care slots to new medical slots ratio goal:

40Z1  (Y1 + Y2 + Y3 + Y4) + d9  d9+ = 0

Child-care slots to new administration slots ratio goal:

80Z1  (Y1 + Y2 + Y3 + Y4) + d10  d10+ = 0

Infant child-care slots to new infant child-care provider ratio goal:

3Z3  Y1 + d11  d11+ = 0

Pre-school child-care slots to new pre-school child-care provider ratio goal:

6Z4  Y2 + d12  d12+ = 0

After-school child-care slots to new after-school child-care provider ratio goal:

15Z5  Y3 + d13  d13+ = 0

Drop-in child-care slots to new drop-in child-care provider ratio goal:

7Z6  Y4 + d14  d14+ = 0
 Xi binary, Yj non-negative integer, Zk non-negative integer

d. Implement and solve the GP using the Risk Solver Platform (RSP).

e. Report your initial recommendation and a summary of the requirements for that
recommendation.

The recommended site selections are sites: 1, 3, 5, 7, 8, and 14. The resulting data is:

Summary of Information for City Council

Program Cost Totals: ($1000s) Personnel Requirements:

Leasing Costs $ 421.75 Medical 26
Purchase Costs $ 430.25 Admin 13
Maintenance Costs $ 134.56 Infant 67
Personnel Costs $2,973.00 Pre-School 40
Medical $ 416.00 After School 27
Admin $ 325.00 Drop In 28
Providers $2,106.00 Maintainers 6
Maintainers $ 126.00 Total Slots Created 1040

PTS: 1

62. Project 7.2  Maxwell Junior College Staffing

The Dean of Maxwell College has a problem. Recent growth in the college's enrollment has been
disproportional among her five primary departments. As a result, she is concerned about the mix of
faculty and staff within each of the departments. Her concerns have been echoed in some recent staff
meetings involving her department heads. This concern prompted a call to the consulting company you
work for which in turn led to your assignment to investigate the Dean's problem.

Since the Dean feels the faculty mix and staff allocation is out-of-kilter across the departments, she
would like an independent staffing analysis completed. She further wants this analysis completed "in
the blind" with respect to her current staffing. Part of her rationale for this blind study concept is to
avoid a rampant rumor mill suggesting a consulting was going to be cutting faculty and staff.

Since Maxwell is primarily a teaching college, the teaching load among all the faculty is quite high.
Naturally, the more senior in academic rank to receive some perks. The faculty teach courses both for
students enrolled in their department's programs and to students enrolled in programs within other
departments. These courses are referred to as In-Department and Out-Department courses,
respectively. The number of In-Department and Out-Department courses per faculty member varies by
academic rank and by the department. Within a department, the numbers are consistent among
comparable faculty members.

The following data was provided by the Dean.

Professor Ranks
Department Full Associate Assistant Instructor
Mathematics Salary $85,000 $68,000 $54,000 $42,000
In-Dept. load 3 3 2 2
Out-Dept. load 0 1 2 2
Computer Sci. Salary $87,000 $75,000 $58,000 $35,000
In-Dept. load 3 3 2 2
 Out-Dept. load 0 1 2 2
Management Salary $91,000 $77,000 $58,000 $37,000
In-Dept. load 3 3 2 2
Out-Dept. load 0 0 2 2
Systems Eng. Salary $87,000 $75,000 $63,000 $35,000
In-Dept. load 3 3 2 2
Out-Dept. load 0 0 2 2
Humanities Salary $80,000 $70,000 $57,000 $37,000
In-Dept. load 3 3 2 2
Out-Dept. load 0 1 2 2

For the purposes of the study, the Dean suggested using an average of 3 classes per semester per
student for In-Department course requirements and 2 classes per semester per student for
Out-Department course requirements. The average class size is 20 students per class. For the purposes
of the student, you may assume every department offers an equal share of the Out-Department course
requirements across all five departments. Naturally, the Dean wants to ensure the staffing study
suggests a faculty sufficient to handle to the expected number of courses offered.

Through discussions with the Dean, various study goals became apparent and are listed below:

 Maintain department-specific established student-faculty ratios (or improve on them);

 Maintain at least the minimum level of full professors on staff (determined by
department);
 Provide each department a staff member for every 15 faculty members authorized, and at
least one in every department;
 Ensure there are adequate faculty to offer the number of courses that department must
offer;
 Ensure the faculty mix is appropriate for the courses that must be offered;
 Keep the number of instructors in each department greater than the total number of
associate and assistant professors in the department;
 Do not exceed a total faculty size of 150 faculty (includes all types) across the five
departments;
 Minimize the total faculty salary across the five departments; and
 Maximize the overall quality of the faculty across the five departments.

Additional data provided by the Dean pertinent to these goals is provided in the following table:

Desired Student/ Number of

Department Number Enrolled Faculty Ratio Full Professors
Mathematics 100 15:1 1
Computer Science 250 20:1 2
Management 400 22:1 2
Systems Engineering 125 17:1 2
Humanities 375 25:1 3

Full Professor Quality 10

Associate Professor Quality 7
Assistant Professor Quality 5
Instructor Quality 2
You have quickly realized that this is a multi-objective linear programming problem to find a solution
that will minimize faculty salary costs and to find a solution to maximize the quality of a faculty mix.
Since these are competing objectives, once determined, you will need to roll these values into a goal
programming model to then interact with the Dean and arrive at a suitable faculty mix solution.

a. Formulate a MOLP for this problem.

b. Implement the MOLP in Excel and use the Risk Solver Platform (RSP) to find solutions.
Report the suggested faculty and staff mix to obtain a minimum cost faculty, a maximum
quality faculty, and the minimum size faculty required.
c. Formulate the GP for this problem.
d. Implement the GP in Excel and solve the model using the Risk Solver Platform (RSP).
Report back the suggested faculty and staff mix.
e. Examine the solution obtained in Part d. If any of the goals seem to far out of line, adjust
your goal programming weights, resolve and re-examine the suggested solution.

ANS:
a. Formulate a MOLP for this problem.

Define the following decision and definitional variables:

Xij is the number of faculty type j in department i
Yi is the total faculty in department i
Zi is the total number of staff allocated to department i

Minimize the total faculty costs:

Minimize
85X11 + 68X12 + 54X13 + 42X14 + 87X21 + 75X22 + 58X23 + 35X24 + 91X31
+ 77X32 + 58X33 + 37X34 + 87X41 + 75X42 + 63X43 + 35X44 + 80X51
+ 70X52 + 57X53 + 37X54

Maximize the total faculty quality points:

Maximize
10X11 + 7X12 + 5X13 + 2X14 + 10X21 + 7X22 + 5X23 + 2X24 + 10X31 + 7X32 + 5X33
+ 2X34 + 10X41 + 7X42 + 5X43 + 2X44 + 10X51 + 7X52 + 5X53 + 2X54

Subject to:
Equal instructor to (associate + assistant) professor numbers:
X12 + X13  X14  0
X22 + X23  X24  0
X32 + X33  X34  0
X42 + X43  X44  0
X52 + X53  X54  0

Compute the total faculty in each of the departments

Y1  (X11 + X12 + X13 + X14) = 0
Y2  (X21 + X22 + X23 + X24) = 0
Y3  (X31 + X32 + X33 + X34) = 0
Y4  (X41 + X42 + X43 + X44) = 0
Y5  (X51 + X52 + X53 + X54) = 0

Goal to maximize full professors in departments

X11  1
X21  2
X31  2
 X41  2
X51  3

Implement student-to-faculty ratio goals:

100  15Y1  0
250  20Y2  0
400  22Y3  0
125  17Y4  0
375  25Y5  0

Teaching load  In department courses (3 course per student)

15  (3X11 + 3X12 + 2X13 + 2X14)  0
38  (3X21 + 3X22 + 2X23 + 2X24)  0
60  (3X31 + 3X32 + 2X33 + 2X34)  0
19  (3X41 + 3X42 + 2X43 + 2X44)  0
56  (3X51 + 3X52 + 2X53 + 2X54)  0

Teaching load  Outside department courses (250 courses per department)

13  (X12 + 2X13 + 2X14)  0
13  (X22 + X23 + 2X24)  0
13  (2X34 + 2X34)  0
13  (2X43 + 2X44)  0
13  (1X52 + 2X53 + 2X54)  0

Add new staff members for each 15 new faculty members added
15Z1  Y1  0
15Z2  Y2  0
15Z3  Y3  0
15Z4  Y4  0
15Z5  Y5  0

Xij non-negative and integer, i department, j faculty type

Yi non-negative and integer of i department type
Zi  1, integer for i department type

b. Implement the MOLP in Excel and use Solver to find solutions.

Department
Faculty Type 1 2 3 4 5
Minimize Faculty Cost Full Professor 1 2 2 2 3
$3,837,000 Associate Professor 0 0 0 0 1
265 Assistant Professor 0 0 0 0 0
90 Faculty Members Instructor 7 16 27 7 22
Staff Member 1 2 2 1 2
Maximize Faculty Quality Full Professor 10 7 47 15 34
$11,513,000 Associate Professor 3 2 0 0 1
1258 Assistant Professor 1 2 0 3 2
150 Faculty Members Instructor 4 4 7 4 4
Staff Member 2 1 4 2 3

As an additional consideration, the Dean might be interested in the minimum number of

faculty required to meet the objectives. That solution is the following:
 Department
Faculty Type 1 2 3 4 5
Minimize Faculty Cost Full Professor 1 2 7 2 9
$4,578,000 Associate Professor 0 6 7 0 6
421 Assistant Professor 1 0 0 1 0
76 Faculty Members Instructor 6 7 9 6 6
Staff Member 1 1 2 1 2

c. Formulate the GP for this problem

Minimize

Subject to:
Equal instructor to (associate + assistant) professor numbers:
X12 + X13  X14 + d1  d1+ = 0
X22 + X23  X24 + d2  d2+ = 0
X32 + X33  X34 + d3  d3+ = 0
X42 + X43  X44 + d4  d4+ = 0
X52 + X53  X54 + d5  d5+ = 0

Compute the total faculty in each of the departments

Y1  (X11 + X12 + X13 + X14) = 0
Y2  (X21 + X22 + X23 + X24) = 0
Y3  (X31 + X32 + X33 + X34) = 0
Y4  (X41 + X42 + X43 + X44) = 0
Y5  (X51 + X52 + X53 + X54) = 0

Goal to maximize full professors in departments

X11  d6+ = 1
X21  d7+ = 2
X31  d8+ = 2
X41  d9+ = 2
X51  d10+ = 3

Implement student-to-faculty ratio goals:

100  15Y1 + d11  d11+ = 0
250  20Y2 + d12  d12+ = 0
400  22Y3 + d13  d13+ = 0
125  17Y4 + d14  d14+ = 0
375  25Y5 + d15  d15+ = 0

Teaching load  In department courses (3 course per student)

15  (3X11 + 3X12 + 2X13 + 2X14) + d16  d16+ = 0
38  (3X21 + 3X22 + 2X23 + 2X24) + d17  d17+ = 0
60  (3X31 + 3X32 + 2X33 + 2X34) + d18  d18+ = 0
19  (3X41 + 3X42 + 2X43 + 2X44) + d19  d19+ = 0
56  (3X51 + 3X52 + 2X53 + 2X54) + d20  d20+ = 0

Teaching load  Outside department courses (250 courses per department)

 13  (X12 + 2X13 + 2X14) + d21  d21+ = 0
13  (X22 + 2X23 + 2X24) + d22  d22+ = 0
13  (2X33 + 2X34) + d23  d23+ = 0
13  (2X43 + 2X44) + d24  d24+ = 0
13  (X52 + 2X53 + 2X54) + d25  d25+ = 0

Add new staff members for each 15 new faculty members added
15Z1  Y1 = 0
15Z2  Y2 = 0
15Z3  Y3 = 0
15Z4  Y4 = 0
15Z5  Y5 = 0

Goal constraint for faculty salary goal:

85X11 + 68X12 + 54X13 + 42X14 + 87X21 + 75X22 + 58X23 + 35X24 + 91X31
+ 77X32 + 58X33 + 37X34 + 87X41 + 75X42 + 63X43 + 35X44 + 80X51
+ 70X52 + 57X53 + 37X54 + d26  d26+ = 3837

Goal constraint for faculty quality goal:

10X11 + 7X12 + 5X13 + 2X14 + 10X21 + 7X22 + 5X23 + 2X24 + 10X31 + 7X32
+ 5X33 + 2X34 + 10X41 + 7X42 + 5X43 + 2X44 + 10X51 + 7X52 + 5X53
+ 2X54 + d27  d27+ = 1258

d. Implement the GP in Excel and solve the model using Solver. What is the
recommended staffing for Maxwell College? The initial results are summarized in
the following table.

Department
Faculty Type 1 2 3 4 5
Full Professor 2 6 9 2 15
Associate Professor 0 1 0 0 0
Assistant Professor 3 0 3 0 0
Instructor 2 5 3 5 0
Staff Member 1 1 1 1 1

e. Examine the solution obtained in Part d. If any of the goals seem to far out of line,
adjust your goal programming weights, resolve and re-examine the suggested
solution.

These answers will vary depending upon the initial weights employed by the student.

Department
Faculty Type 1 2 3 4 5
Full Professor 1 2 2 2 3
Associate Professor 1 2 2 1 15
Assistant Professor 1 1 1 1 1
Instructor 5 12 23 5 0
Staff Member 1 2 2 1 2

This faculty mix is more reasonable. Full professor requirements are met, both associates
and assistants are employed, all course requirements are met and the cost for this option is
$4,298,000, with a quality score of 362 employing a total 81 faculty members.
PTS: 1