A hybrid adaptive large neighborhood search for time-dependent open electric vehicle routing problem with hybrid energy replenishment strategies

As competition intensifies, an increasing number of companies opt to outsource their package distribution operations to professional Third-Party Logistics (3PL) fleets. In response to the growing concern over urban pollution, 3PL fleets have begun to deploy Electric Vehicles (EVs) to perform transportation tasks. This paper aims to address the Time-Dependent Open Electric Vehicle Routing Problem with Hybrid Energy Replenishment Strategies (TDOEVRP-HERS) in the context of urban distribution. The study considers the effect of dynamic urban transport networks on EV energy drain and develops an approach for estimating energy consumption. Meanwhile, the research further empowers 3PL fleets to judiciously oscillate between an array of energy replenishment techniques, encompassing both charging and battery swapping. Based on these insights, a Mixed-Integer Programming (MIP) model with the objective of minimizing total distribution costs incurred by the 3PL fleet is formulated. Given the characteristics of the model, a Hybrid Adaptive Large Neighborhood Search (HALNS) is designed, synergistically integrating the explorative prowess of Ant Colony Optimization (ACO) with the localized search potency of Adaptive Large Neighborhood Search (ALNS). The strategic blend leverages the broad-based solution initiation of ACO as a foundational layer for ALNS’s deeper, nuanced refinements. Numerical experiments on a spectrum of test sets corroborate the efficacy of the HALNS: it proficiently designs vehicular itineraries, trims down EV energy requisites, astutely chooses appropriate energy replenishment avenues, and slashes logistics-related outlays. Therefore, this work not only introduces a new hybrid heuristic technique within the EVRP field, providing high-quality solutions but also accentuates its pivotal role in fostering a sustainable trajectory for urban logistics transportation.