Investigating Edge-Reordering Procedures in a Tabu Search Algorithm for the Capacitated Arc Routing Problem

This paper presents two ideas to guide a tabu search algorithm for the Capacitated Arc Routing Problem to a promising region of the solution space. Both ideas involve edge-reordering, although they work in different ways. One of them aims to directly tackle deadheading cycles, and the other tries to reorder edges with the aim of extending a scope of solutions that can be reached from a given solution. Experiments were performed on 134 benchmark instances of various sizes, and the two ideas were shown to have an ability to guide the search to good solutions. Possible issues that may arise when implementing these ideas are also discussed.