Abstract
Within the context of advancing environmentally sustainable practices in the domain of seaport operations, the incorporation of electric transfer vehicles (EVs) has surfaced as a plausible substitute for traditional diesel-powered counterparts in the realm of cargo transfer. A two-stage energy-transport scheduling framework is proposed, which encompasses vehicle-job matching, EV dispatch, and energy management. The goal of this framework is to enhance the efficiency of transfer operations while concurrently reducing the charging cost. The first stage of the framework focuses on the optimization of the correspondence between vehicles and transfer jobs, followed by the dispatching of EVs in the second stage to mitigate job delay and reduce charging costs. The formulation and resolution of this two-stage model are achieved through the utilization of a deep Q-network method. Empirical validation, conducted through simulation experiments with real-world data from a northern Chinese seaport, confirms the efficacy of the proposed approach and highlights the inherent flexibility of the system. The comparative findings illustrate the efficacy of the proposed approach in diminishing both the energy procurement costs and the idle driving rate for vehicles. Notably, the utilization of this method yields substantial cost reductions of 25.8%, 50.37%, and 6.51% in comparison to the application of mathematical programming, a rule-based matching scenario, and a rule-based scheduling scenario, respectively.
Funding Information
  • Technology Project of State Grid Corporation of China (SGJSCZ00KJJS2400578)

This publication has 18 references indexed in Scilit: