An Initialized ACO for the VRPTW

The Vehicle Routing Problem with Time Windows is an important task in logistic planning. The expenditure on employing labor force, i.e., drivers for vehicles, accounts for most of the costs in this domain. We propose an initialized Ant Colony approach, IACO-VRPTW, with the primary goal (f1) to reduce the number of vehicle needed to serve the customers and the second-priority goal (f2) of decreasing the travel distance. Compared with methods that optimize f2, IACO-VRPTW can reach or reduce f1 in 8 out of 18 instances of the Solomon benchmark set, at the cost of increasing travel distance slightly. IACO-VRPTW can effectively decrease the number of vehicles, travel distance and runtime compared with an ACO without initialization.