Ant Colony Optimisation for vehicle routing problems: from theory to applications

Ant Colony Optimisation is a metaheuristic for combinatorial optimisation problems. In this paper we show its successful application to the Vehicle Routing Problem (VRP). First, we introduce VRP and its many variants, such as VRP with Time Windows, Time Dependent VRP, Dynamic VRP, VRP with Pickup and Delivery. These variants have been formulated in order to bring the VRP as close as possible to the kind of situations encountered in real-world distribution processes. Two case studies are presented: the application of Ant Colony Optimisation to the solution of the Time Dependent VRP, where the travel times depend on the time of the day, and Ant Colony Optimisation for Dynamic VRP, where customers’ orders arrive during the delivery process. Finally, two real-world, industrial-scale applications are presented. The former is an application solving a VRP with Time Windows for a major supermarket chain in Switzerland; the latter is an application solving a VRP with Pickup and Delivery for a leading distribution company in Italy. The results for these two real-world cases, in particular the increase in the vehicle routes performances and their potential use as strategic planning tools, are presented and discussed.