A branch-and-cut algorithm for the multidepot rural postman problem

Abstract

This paper considers the Multidepot Rural Postman Problem, an extension of the classical Rural Postman Problem in which there are several depots instead of only one. The aim is to construct a minimum cost set of routes traversing each required edge of the graph, where each route starts and ends at the same depot. The paper makes the following scientific contributions: (i) It presents optimality conditions and a worst case analysis for the problem; (ii) It proposes a compact integer linear programming formulation containing only binary variables, as well as a polyhedral analysis; (iii) It develops a branch-and-cut algorithm that includes several new exact and heuristic separation procedures. Instances involving up to four depots, 744 vertices, and 1,315 edges are solved to optimality. These instances contain up to 140 required components and 1,000 required edges.

Peer Reviewed

Postprint (author's final draft)