Solving green vehicle routing problem using self-organizing map

Abstract

Road transport systems constitute a significant source of emissions that adversely affects the environment and human health. Vehicular emissions can be reduced by various means, one of which is efficient vehicle-route planning. Accordingly, the green-vehicle routing problem (GVRP) was developed to minimize the distance covered and emissions released by vehicles via the use of alternate-fuel vehicles (AFVs), which can be refueled at alternate fueling stations (AFSs) and depots.

This research proposes the use of a self-organizing map (SOM) to address the green-vehicle routing problem. Numerical experiments were performed on randomly generated problem instances to evaluate the effectiveness of the SOM algorithm. The results revealed that the SOM yielded better GVRP solutions than the other heuristic approaches reported in the literature. We also found that problems involving uniformly distributed customers are easier to solve than those involving clustered customers. Additionally, increasing the number of AFSs reduces transportation costs and allows vehicles to cater to more customers. However, realization of all the benefits offered by additional AFSs depends on their respective installation locations. This study is expected to provide organizations possessing fleets of AFV with an alternative approach to eliminate the risks associated with the shortage of fuel and time while operating on cost-effective routes. This would help such organizations overcome the challenges associated with limited vehicle-driving range and fuel availability.