An ILS heuristic for the waste collection vehicle routing problem with time windows

Abstract

This study addresses the Waste Collection Vehicle Routing Problem with Time Windows (WCVRPTW) with multiple disposal sites and drivers’ lunch breaks. The problem consists of routing garbage trucks to collect customers waste within given time window in order to minimize the total travel cost. The waste is driven to disposal sites, multiple trips to disposal sites are allowed per day for the vehicles, and the vehicles must be empty when returning to the depot. To find near-optimal routes for the waste collecting vehicles, we propose a heuristic algorithm based on the meta-heuristics Iterated Local Search (ILS) and Variable Neighborhood Descent (VND). The performance of the proposed algorithm is evaluated by comparing its results to CPLEX commercial solver, and a variable neighbourhood tabu search (VNTS) algorithm addressed in the literature. The results obtained, on benchmark problems from the literature, are highly competitive and a number of new improved solutions are reported.