3
$\begingroup$

I am having trouble understanding the proof for why this problem is NP-complete.

"Network reliability problem"

Given an undirected graph $G$, we have a positive integer $T(X,Y)$ associated with every pair of vertices $(X,Y)$. We finally have an upper limit $B$. Can we find a subgraph including $B$ edges where every pair of vertices $(X,Y)$ has (at least) $T(X,Y)$ disjoint paths between $X$ and $Y$?

My lecturer showed that we can reduce the Hamiltonian cycle problem to this problem using the following pseudocode:

HamiltonianCycle(G[V,E]):
 return Reliability(G[V,E], T(X,Y)=2 forall (X,Y), B=|V|)

I do not understand why we set $T(X,Y)=2$.

Could someone help me understand?

asked Nov 8 at 10:11
$\endgroup$

1 Answer 1

3
$\begingroup$

Suppose that you have an algorithm for Reliability.

I want to solve Hamiltonian Cycle.

What do I do? I ask: can you find a subgraph consisting of at most $n$ edges such that between every pair of vertices there are 2 disjoint paths?

Given $n$ vertices and $n$ edges, there is only one possible graph that satisfies the criterion: the cycle graph.

Reduction from Ham Cycle to Reliability. Let $B=n$ and $T(X, Y) = 2$ for every pair of vertices $X, Y$.

answered Nov 8 at 19:20
$\endgroup$

Your Answer

Draft saved
Draft discarded

Sign up or log in

Sign up using Google
Sign up using Email and Password

Post as a guest

Required, but never shown

Post as a guest

Required, but never shown

By clicking "Post Your Answer", you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.