Review Questions, Final Exam: A Few General Questions
Review Questions, Final Exam: A Few General Questions
Review Questions, Final Exam: A Few General Questions
3. What is the main idea behind the Simplex Method? (Think about what it is doing
graphically- How does the algorithm start, how does it proceed?)
3. Convert the following LP to one in standard form. Write the result in matrix-vector
form, giving x, c, A, b (from our formulation).
min z = 3x − 4y + 2z
st 2x − 4y ≥ 4
x + z ≥ −5
y+z ≤1
x+y+z =3
with x ≥ 0, y is URS, z ≥ 0.
x1 = 0, x2 = 6, s1 = 4, s2 = 0, s3 = 6
1
Figure 1: Figure for the convex combinations, Exercise 5.
5. Consider Figure 1, with points A(1, 1), B(1, 4) and C(6, 3), D(4, 2) and E(4, 3).
2
(a) Is the tableau optimal (and did your answer depend on whether we are maximizing
or minimizing)? For the remaining questions, you may assume we are maximizing.
(b) Give the current BFS.
(c) Directly from the tableau, can I increase x2 from 0 to 1 and remain feasible? Can
I increase it to 4?
(d) If x2 is increased from 0 to 1, compute the new value of z, x1 , s1 (assuming s2
stays zero).
(e) Write the objective function and all variables in terms of the non-basic (or free)
variables, and then put them in vector form.
3
15. Consider a maximization problem with the optimal tableau given below. First, give
the optimal solution. Next, determine the second-best solution (think about the sets
of variables that could have come before the current set).
x1 x2 x3 x4 rhs
2 1 0 0 10
3 2 1 0 3
4 3 0 1 5
17. You are given the tableau below for a maximization problem. Give conditions on
a1 , a2 , a3 , b, c that make the following statements true:
x1 x2 x3 x4 x5 rhs
−c 2 0 0 0 10
−1 a1 1 0 0 4
a2 −4 0 1 0 1
a3 3 0 0 1 b
18. Suppose we have obtained the tableau below for a maximization problem. State con-
ditions on a1 , a2 , a3 , b, c1 , c2 that are required to make the following statements true:
(a) The current solution is optimal, and there are alternative optimal solutions.
(b) The current basic solution is not a BFS.
(c) The current basic solution is a degenerate BFS.
(d) The current basic solution is feasible, but the LP is unbounded.
(e) The current basic solution is feasible, but the objective function can be improved
by replacing x6 with x1 as a basic variable.
x1 x2 x3 x4 x5 x6 rhs
c1 c2 0 0 0 0 10
4 a1 1 0 a2 0 b
−1 −5 0 1 −1 0 2
a3 −3 0 0 −4 1 3
4
From the Group Work Problem
(Exam 2 Review Links on the Class Website)
In the textbook’s “Dakota Problem”, we are making desks, tables and chairs, and we want
to maximize profit given constraints on lumber, finishing and carpentry (resp).
For the primal, let x1 , x2 , x3 be the number of desks, tables and chairs we make (resp),
where the original (max) tableau is as given below:
x1 x2 x3 s1 s2 s3 rhs x1 x2 x3 s1 s2 s3 rhs
−60 −30 −20 0 0 0 0 0 5 0 0 10 10 280
8 6 1 1 0 0 48 → 0 −2 0 1 2 −8 24
3
4 2 2
0 1 0 20 0 −2 1 0 2 −4 8
3 1 5 1 3
2 2 2
0 0 1 8 1 4
0 0 −2 2
2
1. Write the down vectors/matrices that we typically use in our computations. Namely,
c, cB , B, and B −1 .
2. Using our vector notation, if B gives the optimal basis, how do we compute the dual,
y =?
4. Using the optimal Row 0 from the primal, write down the solution to the dual:
5. In our ”normal form”, we have Ax ≤ b for the primal and AT y ≥ c for the dual. We
will define two ”slacks”1
• The ”slack” for the primal, given x: b − Ax. Compute the current slack for the
primal.
• The ”slack” for the dual, given y: AT y − c. Compute the current slack for the
dual. You can use your answer to (3) if necessary.
7. Write down the inequalit(ies) we need for ∆ if we change the coefficient of x2 from 30
to 30 + ∆, and we want the current basis to remain optimal.
8. Write down the inequalit(ies) we need for ∆ if we change the coefficient of x3 from 20
to 20 + ∆, and we want the current basis to remain optimal.
9. How does changing a column of A effect the dual? Use this to see what would happen
if we change the column for x2 (tables) to be [5, 2, 2]T - Is it now worth it to make
tables?
1
Sorry, the vocabulary is related to the ”slack variable”, but we’re using ”slack” in a different context
now. Ask if you’re ever not sure which we’re talking about.
5
10. How does creating a new column of A effect the dual? Use this to see if it makes
sense to manufacture footstools, where we sell them for $15 each, and the resources
are [1, 1, 1]T .
Find Row 0 and the RHS for the optimal tableau (without performing row reductions!)
7. Give an argument why, if the primal is unbounded, then the dual must be infeasible.
8. In solving the following LP, we obtain the optimal tableau shown:
(a) If we add a new constraint, is it possible that we can increase z? Why or why
not?
(b) If we add the constraint 3x1 + x2 ≤ 10, is the current basis still optimal?
(c) If we add the constraint x1 − x2 ≥ 6, we can quickly see that the optimal solution
changes. Find out if we have a new optimal solution or if we have made the
problem infeasible.
(d) Same question as the last one, but let’s change the constraint to 8x1 + x2 ≤ 12.
(e) If I add a new variable x3 so that:
max z = 6x1 + x2 + x3
st x1 + x2 + 2x3 ≤ 5
2x1 + x2 + x3 ≤ 6
x1 , x2 ≥ 0
Does the current basis stay optimal? Answer two ways- One using the optimal
tableau, and the second using the dual.
6
From the Chapter 6 Review
1. Consider the following LP and its optimal tableau, shown.
7
(a) Find the dual of this LP and its optimal solution.
(b) Find the range of values of b2 = 15 for which the current basis remains optimal.
If b2 = 5, what is the new optimal solution?
max z = −4x1 − x2 − x3
st 4x1 + 3x2 + x3 ≥ 6
x1 + 2x2 + x3 ≤ 3
3x1 + x2 + x3 = 3
x1 , x2 , x3 ≥ 0
20. Use the Theorem of Complementary Slackness to find the optimal solution to the
following LP and its dual. To assist you, note that y1 = y2 = 1 is a solution to the
dual.
max z = 3x1 + 4x2 + x3 + 5x4
st x1 + 2x2 + x3 + 2x4 ≤ 5
2x1 + 3x2 + x3 + 3x4 ≤ 8
x1 , x2 , x3 , x4 ≥ 0
8
33. Consider the following LP:
max z = c1 x1 + c2 x2 x1 x2 s1 s2 rhs
st 3x1 + 4x2 ≤ 6 0 0 1 2 14
⇒
2x1 + 3x2 ≤ 4 1 0 3 −4 2
x1 , x2 ≥ 0 0 1 −2 3 0
34. Consider the following LP and its partially determined optimal tableau below (missing
values are denoted by ?).
2. Five workers are available to perform four jobs. The time it takes each worker to
perform each job is given below. The goal is to assign workers to jobs so as to minimize
9
the total time required. Use the Hungarian method to solve.
Job 1 2 3 4
Worker 1 10 15 10 15
2 12 8 20 16
3 12 9 12 18
4 6 12 15 18
5 16 12 8 12
3. A company must meet the demands shown below for a product. Demand may be
backlogged at a cost of $5 per unit per month. All demand must be met at the end of
March. Thus, if 1 unit of January demand is met during March, a cost of $5× 2=$10
is incurred. Monthly production capacity and unit production cost during each month
are shown below. A holding cost of $20 per unit is assessed on the inventory at the
end of each month.
HINT: To set this up, think of Jan, Feb and Mar as having supplies as 35, 30and35,
and demands as 30, 30, 20 (we’ll need a dummy to balance. For the costs, January can
supply January at a cost of $400 per unit, January can supply Feb at a cost of $420
per unit, and it can supply March at a cost of $440 per unit.
4. Appletree cleaning has five cleaning people. To complete cleaning the house, they
must vacuum, clean the kitchen, clean the bathroom, and do general straightening up.
The time it takes each person to do each job is shown below. If each person can be
assigned at most one job, set up and solve the table for the assignment problem using
10
the Hungarian method.
Job 1 2 3 4
Worker 1 6 5 2 1
2 9 8 7 3
3 8 5 9 4
4 7 7 8 3
5 5 5 6 4
8. Use the NW corner method to find a BFS, then solve the unbalanced (minimization)
problem shown below:
12 14 16
60
14 13 19
50
17 15 18
40
40 70 10
4 2 4
15
12 8 4
15
10 10 10
11
For the next two problems, we’ll need the optimal tableau for the PowerCo problem
(p 391):
8 6 10 9
10 25 35
9 12 13 7
45 5 50
14 9 16 5
10 30 40
45 20 30 30
15. For the PowerCo problem, find the range of values of c24 for which the current basis is
optimal.
16. For the PowerCo problem, find the range of values of c23 for which the current basis is
optimal.
From Chapter 8
The review for this chapter was pretty terrible, so here are some review questions.
1. Write the shortest path problem as (i) a transhipment problem, and (ii) a linear pro-
gram. For specificity, use the PowerCo network below (Figure 2, p 414). (Hints: For
transhipment, we have one supply, one demand, and a bunch of warehouses. For the
LP, you could write it from the transhipment problem.). Finally, find the shortest path
from Plant 1 to City 1 using Dijkstra’s algorithm.
2. Given the figure below (Fig 23 from the text), first write the maximum flow problem
as a linear program. (Hint: Think about the constraints on the flow for each edge,
12
then for each vertex). Solve the max-flow problem using Ford-Fulkerson. Be sure to
write out the residual graphs. Finally, find a cut giving the minimum capacity to show
that your solution is correct.
3. Continuing with Figure 23 from the previous question, with the maximum flow, if the
cut is:
A = {so, 2, 3} , B = {1, 4, si}
then what is the net flow across the cut? What is the capacity of the cut?
4. To use a max-flow for the assignment problem, recall that we have a source node that
goes to a node for each person (capacity of 1 each), then each person is attached to a
job (capacity of 1 for each edge), and then each job is attached to a sink. (NOTE: We
are maximizing the number of compatible pairs.)
(Problem 7, 8.3) Four workers are available to perform four jobs, but not all workers
may be assigned to every job (See the chart below, X marks compatible). Draw the
network for the maximum flow problem that can be used to determine whether all jobs
can be assigned to a suitable worker.
5. (Figure 4 from text below) (a) If we look at the edges as having costs, find the path
from node 1 to all other nodes. (b) If we look at the edges as capacities for a flow, find
the maximum flow.
13
14