General variable neighborhood search for electric vehicle routing problem with time-dependent speeds and soft time windows

With the growing environmental concerns and the rising number of electric vehicles, researchers and companies are paying more and more attention to green logistics. This paper studies the Electric Vehicle Routing Problem with time-dependent speeds and soft time windows. The purpose is to minimize the total distance travelled, while penalizing early or late arrivals at the customers' locations. For this purpose, we formulated the Mixed Integer Linear Program (MILP) and developed a General Variable Neighborhood Search (GVNS) metaheuristic, an efficient way to tackle this problem. To prove the efficiency of our approach, we tested the GVNS against the Adaptive Large Neighborhood Search (ALNS) algorithm and our MILP model, using a set of available benchmark instances. After an extensive experimental evaluation, we concluded that GVNS can find better quality solutions than other methods considered in this research or the same quality solution in less time.(c) 2023 by the authors; licensee Growing Science, Canada

Automated summary

This paper studies the Electric Vehicle Routing Problem with time-dependent speeds and soft time windows. The authors formulated a Mixed Integer Linear Program (MILP) and developed a General Variable Neighborhood Search (GVNS) metaheuristic to tackle the problem. Experimental results showed that GVNS can find better quality solutions in less time compared to other methods.