Vehicle route planning in e-waste mobile collection on demand supported by artificial intelligence algorithms

Mobile collection of e-waste on demand is one of the methods that can contribute to an increase in the collection rate of waste. In this method, a person requests the waste pick up from a household at a preferred time. To support such a collection method an efficient algorithm and information system for convenient waste disposal for residents has to be applied. Our study investigates using artificial intelligence algorithms for solving the vehicle routing problem with time windows for a heterogeneous fleet of waste collection vehicles. We present an algorithm and a productive model of the online system enabling comprehensive communication for people that request waste equipment for collection, registering of data and solving the VRPTW. The system includes parametric models of four algorithms (simulated annealing, tabu search, greedy, bee colony optimization). The result of the optimization is the assignment of a minimal number of collection vehicles, a vehicle routing plan, timely collection of waste from a household and collection cost reduction. The study includes the simulation of e-waste collection requests in Tokyo, Philadelphia and Warsaw to compare algorithms for various urban arrangements of streets and buildings. The results show that the best of the four algorithms, to facilitate e-waste mobile collection on demand, is simulated annealing and the worst is tabu search. The proposed model and algorithm can bring significant improvement in planning the routes of the vehicles in the e-waste collection, including a positive social impact on the new method of waste collection, especially in urban areas.