Scribd
Upload
116 views2 pages

Homework 5: Ij Ij I Ij

This homework assignment contains 6 problems related to optimization and network flows. Students are encouraged to discuss the problems but must submit individual work. The problems include: 1) solving an uncapacitated network flow problem using the network simplex algorithm; 2) showing the equivalence between capacitated network flows and transportation problems; 3) formulating an edge orientation problem as a minimum cost network flow; 4) properties of node-arc incidence matrices and the max-flow min-cut theorem; 5) representing circulations as cycles and formulating a minimum average cost cycle problem; and 6) finding the maximum flow and associated cut in a given network.

Uploaded by

Morley Calo Mahinay Jr.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
116 views2 pages

Homework 5: Ij Ij I Ij

This homework assignment contains 6 problems related to optimization and network flows. Students are encouraged to discuss the problems but must submit individual work. The problems include: 1) solving an uncapacitated network flow problem using the network simplex algorithm; 2) showing the equivalence between capacitated network flows and transportation problems; 3) formulating an edge orientation problem as a minimum cost network flow; 4) properties of node-arc incidence matrices and the max-flow min-cut theorem; 5) representing circulations as cycles and formulating a minimum average cost cycle problem; and 6) finding the maximum flow and associated cut in a given network.

Uploaded by

Morley Calo Mahinay Jr.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Optimization Khaled Elbassioni

Homework 5
Summer 2011 Anke van Zuylen

This assignment is due on June 16 in class. You are allowed (even encouraged) to
discuss these problems with your fellow classmates. All submitted work, however, must
be written individually without consulting someone else’s solutions or any other source
like the web.

Problem 1: Consider the uncapacitated network flow problem shown in Figure 1. The
label next to each arc is its cost. Solve the problem using the network simplex algorithm.
Start with the tree indicated by the dashed arcs in the figure.
1
2 3 6

0 5 2
2 1
0 3
8 1
2 1
3 2
0
1 4
2 1 4 7 5

Figure 1: The network flow problem in Exercise 1.

Problem 2: (Equivalence of capacitated network flow and transportation


problems) Consider a capacitated network problem defined by a graph (N , A) and the
data uij , cij , bi . Assume that the capacity uij of every arc is finite. We construct a related
transportation problem as follows. For every arc (i, j) ∈ A, we form a source node in
the transportation P problem with supply uij . For every node i ∈ N , we construct a sink
node with demand {k|(i,k)∈A} uik − bi . At every supply node (i, j) there are two outgoing
infinite capacity arcs: one goes to demand node i, and its cost coefficient is 0; the other
goes to demand node j and its cost coefficient is cij . See Figure 2 for an illustration.
Show that there is a one-to-one correspondence between feasible flows in the two
problems and that the cost of the two corresponding flows is the same.

P
i k uik − bi
0

uij i, j

cij P
j k ujk − bj

Figure 2: The transportation problem in Exercise 2.


Problem 3: Given an undirected (multi-)graphPG = (V, E), with weights w : V → Z, it
is required to orient the edges so as to minimize v max{w(v) − in(v), 0}, where in(v) is
the in-degree of v in the orientation. Formulate this problem as a minimum cost network
flow problem.

Problem 4: (i) Show that the node-arc incidence matrix of a directed graph is totally
unimodular (that is, the determinant of every square submatrix has value in {0, ±}).
(ii) Consider the standard LP formulation of the maximum flow problem. Write the
dual LP and then use the duality theorem of LP to prove the max-flow min-cut theorem.

Problem 5: (i) (Vertices of the circulation polytope) Let G = (N , A) be a


directed graph, with node-arc incidence matrix A. Show that the vertices of the polytope
{f ∈ RA : Af = 0, f ≥ 0, e′ f = 1} (where e is the vector of all ones) are in one-to-one
correspondence with the directed cycles of G.
(ii) Let G = (N , A) be a directed graph with costs c : A → R. Consider the problem
of finding a directed cycle in G with minimum average cost. Show that this problem can
be formulated as a linear program (Hint: Use part (i)).

Problem 6: Find the maximum flow from node 1 to node 7 in the network G = (N , E),
where N = {1, . . . , 7}, E = {(1, 2), (1, 3), (2, 3), (2, 4), (2, 5), (3, 4), (3, 6), (4, 5), (4, 6), (5, 7),
(6, 5), (6, 7)} with arc capacities: u12 = 6, u13 = 7, u23 = 1, u24 = 3, u25 = 4, u34 = 2,
u36 = 5, u45 = 3, u46 = 2, u57 = 7, u65 = 2, u67 = 4. Identify the associated minimal
cut-set.

You might also like