On the Wiener index of orientations of graphs (Q6103481)

\textit{M. Knor} et al. [Appl. Math. Comput. 284, 260--267 (2016; Zbl 1410.05079)] conjectured: ``Let \(T\) be a tree. If \(D\) is an orientation of \(T\) that maximises \(W(D)\) (the Winer index), then there exists a vertex \(v\) in \(D\) such that for every vertex \(u\) there exists a \((u,v)\)-path or a \((v,u)\)-path in \(D\).'' The author of this paper proves that this conjecture is not true in general. The counter-example of this conjecture appears on page 127. It is not a simple example because the author proves a long result (Theorem 2) that shows that the conjecture does not hold. Using the proof of Theorem 2, the author proves in Section 3 that it is not easy to find an orientation of a given graph that maximises the Wiener index. This is an NP problem. The techniques used to prove the main result, Theorem 2, are easy. However, it contains good claims.