A2.10

Algorithms and Networks
Part II, 2003

(i) Consider a network with node set NN and set of directed arcs AA equipped with functions d+:AZd^{+}: A \rightarrow \mathbb{Z} and d:AZd^{-}: A \rightarrow \mathbb{Z} with dd+d^{-} \leqslant d^{+}. Given u:NRu: N \rightarrow \mathbb{R} we define the differential Δu:AR\Delta u: A \rightarrow \mathbb{R} by Δu(j)=u(i)u(i)\Delta u(j)=u\left(i^{\prime}\right)-u(i) for j=(i,i)Aj=\left(i, i^{\prime}\right) \in A. We say that Δu\Delta u is a feasible differential if

d(j)Δu(j)d+(j) for all jAd^{-}(j) \leqslant \Delta u(j) \leqslant d^{+}(j) \text { for all } j \in A

Write down a necessary and sufficient condition on d+,dd^{+}, d^{-}for the existence of a feasible differential and prove its necessity.

Assuming Minty's Lemma, describe an algorithm to construct a feasible differential and outline how this algorithm establishes the sufficiency of the condition you have given.

(ii) Let ES×TE \subseteq S \times T, where S,TS, T are finite sets. A matching in EE is a subset MEM \subseteq E such that, for all s,sSs, s^{\prime} \in S and t,tTt, t^{\prime} \in T,

(s,t),(s,t)M implies s=s(s,t),(s,t)M implies t=t\begin{array}{lll} (s, t),\left(s^{\prime}, t\right) \in M & \text { implies } & s=s^{\prime} \\ (s, t),\left(s, t^{\prime}\right) \in M & \text { implies } & t=t^{\prime} \end{array}

A matching MM is maximal if for any other matching MM^{\prime} with MMM \subseteq M^{\prime} we must have M=MM=M^{\prime}. Formulate the problem of finding a maximal matching in EE in terms of an optimal distribution problem on a suitably defined network, and hence in terms of a standard linear optimization problem.

[You may assume that the optimal distribution subject to integer constraints is integervalued.]