A branch-and-cut algorithm for the Time Window Assignment Vehicle Routing Problem

This paper presents a branch-and-cut algorithm for the Time Window Assignment Vehicle Routing Problem (TWAVRP), the problem of assigning time windows for delivery before demand volume becomes known. A novel set of valid inequalities, the precedence inequalities, is introduced and multiple separation heuristics are presented. In our numerical experiments the branch-and-cut algorithm is 3.8 times faster when separating precedence inequalities. Furthermore, in our experiments, the branch-and-cut algorithm is 193.9 times faster than the best known algorithm in the literature. Finally, using our algorithm, instances of the TWAVRP are solved which are larger than the small scale instances previously presented in the literature.