Second Exam Solutions ♦ Principles of Optimization (0301271) ♦ Fall 2016

Note: This exam is composed of 30 questions. You have 60 minutes to finish.

Question Answer
1 B
2 B
3 B
4 B
5 B
6 B
7 A
8 E
9 D
10 E
11 A
12 A
13 C
14 E
15 A
16 B
17 A
18 E
19 B
20 D
21 A
22 A
23 C
24 B
25 C
26 B
27 E
28 B
29 B
30 C

No. of incorrect answers

Total score = No. of correct answers − .
4

1
Circle the correct answer to each question. Wrong answers will be penalized by deduction of a
quarter of a mark for each incorrect answer.

1. All linear programming problems may be solved using graphical method.

A. True.
B. False.

2. The composition of two convex functions is convex.

A. True.
B. False.

3. XYZ Inc. produces two types of paper towels, called regular and super-soaker. Marketing
has imposed a constraint that the total monthly production of regular should be no more
than twice the monthly production of super-soakers. Letting x1 be the number of units of
regular produced per month and x2 represent the number of units of super-soaker produced
per month, the appropriate constraint(s) will be

A. 2x1 ≤ x2 .
B. x1 ≤ 2x2 .
C. x1 ≤ 0.5x2 .
D. x1 − x2 ≤ 0.
E. x1 0.5x2 ≥ 0.

4. Problem A is a given formulation of a linear program with an optimal solution. Problem B

is a formulation obtained by multiplying the objective function of Problem A by a positive
constant and leaving all other things unchanged. Problems A and B will have

A. the same optimal solution and same objective function value.

B. the same optimal solution but different objective function values.
C. different optimal solutions but same objective function value.
D. different optimal solutions and different objective function values.

5. Consider the following linear programming problem:

max 12x + 10y

s.t. 4x + 3y ≤ 480,
2x + 3y ≤ 360,
x, y ≥ 0.

Which of the following points (x, y) could be a feasible corner point?

A. (40,48).
B. (120,0).
C. (180,120).
D. (30,36).
E. none of the above.

2
 6. A two variable linear programming problem cannot be solved by the simplex method.

A. True.
B. False.

7. The intersection of two convex sets is convex.

A. True.
B. False.

8. XYZ Inc. produces two types of printers, called regular and high-speed. Regular uses 2
units of recycled plastic per unit and high-speed uses 1 unit of recycled plastic per unit of
production. The total amount of recycled plastic available per month is 5,000. A critical
machine is needed to manufacture the printers and each unit of Regular requires 5 units of
time in this machine and each unit of high-speed requires 3 units of time in this machine.
The total time available in this machine per month is 10000 units. Letting x1 be the number
of units of regular produced per month and x2 represent the number of units of high-speed
produced per month, the appropriate constraint(s) will be

A. 2x1 + x2 = 5000.
B. 2x1 + x2 ≤ 5000.
C. 5x1 + 3x2 ≤ 10000.
D. A and C.
E. B and C.

9. Problem A is a given formulation of a linear program with an optimal solution and its
constraint 1 is ≤ type. Problem B is a formulation obtained from Problem A by replacing the
≤ constraint by an equality constraint and leaving all other things unchanged. Problems A
and B will have

A. the same optimal solution and same objective function value.

B. the same optimal solution but different objective function values.
C. different optimal solutions but same objective function value.
D. same or different solution profile depending on the role of the constraints in the solutions.

10. Consider the following linear programming problem:

max 12x + 10y

s.t. 4x + 3y ≤ 480,
2x + 3y ≤ 360,
x, y ≥ 0.

Which of the following points (x, y) is not in the feasible region?

A. (30,60).
B. (105,0).
C. (0,110).
D. (100,10).
E. none of the above.

3
11. In any graphically solvable linear program, if two points are feasible, then any weighted
average of the two points where weights are non-negative and add up to 1.0 will also be
feasible.

A. True.
B. False.

12. In a two variable graphical linear program, if the coefficient of one of the variables in the
objective function is changed (while the other remains fixed), then slope of the objective
function expression will change.

A. True.
B. False.

13. XYZ Inc. produces two types of printers, called regular and high-speed. Regular uses 2
units of recycled plastic per unit and high-speed uses 1 unit of recycled plastic per unit of
production. XYZ is committed to using at least 5,000 units of recycled plastic per month.
A critical machine is needed to manufacture the printers and each unit of Regular requires
5 units of time in this machine and each unit of high-speed requires 3 units of time in this
machine. The total time available in this machine per month is 10000 units. Let x1 be the
number of units of regular produced per month and x2 represent the number of units of
high-speed produced per month. Imposing both of these constraints, and non-negativity
constraints one of the feasible corner points is (assuming the first number in parenthesis is
x1 and the second number in the parenthesis is x2 ):

A. (0, 0).
B. (2000, 0).
C. None exists.
D. (0, 5000).
E. (2500, 0).

14. A point that satisfies all of a problem’s constraints simultaneously is a(n)

A. optimal solution.
B. corner point.
C. intersection of the profit line and a constraint.
D. intersection of two or more constraints.
E. None of the above.

15. If, when we are using a Simplex table to solve a maximization problem, we find that the
ratios for determining the pivot row are all negative, then we know that the solution is

A. unbounded.
B. infeasible
C. degenerate.
D. optimal.
E. none of the above.

4
16. If a graphically solvable linear program is unbounded, then it can always be converted to a
regular bounded problem by removing a constraint.

A. True.
B. False.

17. In a two variable graphical linear program, if the RHS of one of the constraints is changed
(keeping all other things fixed) then the plot of the corresponding constraint will move in
parallel to its old plot.

A. True.
B. False.

18. Two models of a product - Regular (x) and Deluxe (y) - are produced by a company. A linear
programming model is used to determine the production schedule. The formulation is as
follows:
max 50x + 60y (maximum profit)
s.t. 8x + 10y ≤ 800 (labor hours),
x+y ≤ 120 (total units demanded),
4x + 5y ≤ 500 (raw materials),
x, y ≥ 0 (non-negativity).
The optimal solution is x = 100, y = 0. How many units of the labor hours must be used to
produce this number of units?

A. 400.
B. 200.
C. 500.
D. 5000.
E. none of the above.

19. LP theory states that the optimal solution to any problem will lie at

A. the origin.
B. a corner point of the feasible region.
C. the highest point of the feasible region.
D. the lowest point in the feasible region.
E. none of the above.

20. In converting a greater-than-or-equal constraint for use in a simplex table, we must add

A. an artificial variable.
B. a slack variable.
C. a slack and an artificial variable.
D. an excess and an artificial variable.
E. a slack and an excess variable.

5
21. Dual of a linear programming problem with maximize objective function, all constraints and
non-negative variables has minimize objective function, all constraints and non-negative
decision variables.

A. True.
B. False.

22. The two objective functions (max 5x + 7y, and min −15x − 21y) will produce the same solution
to a linear programming problem.

A. True.
B. False.

23. In order for a linear programming problem to have a unique solution, the solution must exist

A. at the intersection of the non-negativity constraints.

B. at the intersection of the objective function and a constraint.
C. at the intersection of two or more constraints.
D. none of the above.

24. For a minimization problem using a Simplex table, we know we have reached the optimal
solution when the row R0

A. has no numbers in it.

B. has no positive numbers in it.
C. has no negative numbers in it.
D. has no nonzero numbers in it.
E. none of the above.

25. A feasible solution requires that all artificial variables are

A. greater than zero.

B. less than zero.
C. equal to zero.
D. there are no special requirements on artificial variables; they may take on any value.
E. none of the above.

6
26. If the right-hand side of a constraint is changed, the feasible region will not be affected and
will remain the same.

A. True.
B. False.

27. If a minimization problem has an objective function of 2x1 + 5x2 , which of the following
corner points is the optimal solution?

A. (0,2).
B. (0,3).
C. (3,3).
D. (1,1).
E. (2,0).

28. In a linear programming problem, when the objective function is parallel to one of the
constraints, then

A. the solution is not optimal.

B. multiple optimal solutions may exists.
C. a single corner point solution exists.
D. no feasible solution exists.
E. none of the above.

29. A linear programming problem cannot have

A. no optimal solutions.
B. exactly two optimal solution.
C. as many optimal solutions as there are decision variables.
D. an infinite number of optimal solutions.
E. none of the above.

30. With Bland’s rule, the simplex algorithm solves feasible linear minimization problems with-
out cycling when

A. we choose the rightmost nonbasic column with a negative cost to select the entering
variable.
B. we choose the rightmost nonbasic column with a negative cost to select the leaving
variable.
C. we choose the leftmost nonbasic column with a negative cost to select the entering
variable.
D. we choose the leftmost nonbasic column with a negative cost to select the leaving variable.