On Solving the Time Window Assignment Vehicle Routing Problem via Iterated Local Search

In this paper, we propose a combined algorithm based on an Iterated Local Search (ILS) and a mathematical model to solve the Time Window Assignment Vehicle Routing Problem (TWAVRP). The TWAVRP appears when the volume of customer demands is uncertain and time windows should be allocated to customers so as to minimize expected travel costs. Our goal is to find a heuristic strategy that can efficiently improve the current TWAVRP solution methods in the literature. For this purpose, we first use an ILS algorithm to generate feasible sets of routes. Then, we invoke a Mixed Integer Linear Programming formulation that assigns time windows to customers and selects the subset of routes of minimum expected cost. Computational results performed on benchmark instances show that our algorithm is competitive with respect to the literature, especially for instances with more than 45 clients.