This study proposes a simulation-based optimization approach to address the recharging scheduling problem of electric buses to minimize charging waiting time. Poor scheduling could lead to longer waiting times and potentially affect operation schedules regarding time and service quality. This study addresses a simulation-based optimization framework to evaluate various performance metrics during electric bus service, including waiting times, charging costs, and the utilization of charging piles. In this study, we propose a hybrid approach, simplified swarm optimization (SSO), which is an evolutionary algorithm with a backtracking (BT) mechanism and dynamic charging in a simulation framework. Based on the dynamic charging, SSO is used to determine the additional charging in terms of battery capacities, and a BT mechanism is employed to enhance algorithm efficiency and achieve breakthroughs in solution quality. A case study from Taiwan with 43 generated datasets was conducted in deterministic and stochastic situations to compare the effectiveness and efficiency among three charging rules (i.e., full charging rule, flexible charging rule, dynamic charging rule) and two algorithms (i.e., particle swarm optimization and SSO) The results indicate the superior performance in all scenarios by using a statistical test, which offers effective decision support for bus operators’ electric bus recharging scheduling.

中文翻译：

---

一种基于仿真的电动公交车充电调度问题优化方法


本研究提出了一种基于仿真的优化方法来解决电动公交车的充电调度问题，以最大限度地减少充电等待时间。糟糕的调度可能会导致更长的等待时间，并可能影响时间和服务质量方面的运营调度。本研究提出了一个基于仿真的优化框架，用于评估电动公交车服务期间的各种性能指标，包括等待时间、充电成本和充电桩利用率。在这项研究中，我们提出了一种混合方法，即简化集群优化 （SSO），这是一种在模拟框架中具有回溯 （BT） 机制和动态充电的进化算法。基于动态充电，使用 SSO 来确定电池容量方面的额外充电，并采用 BT 机制来提高算法效率，实现解决方案质量的突破。在确定性和随机性情况下，使用来自 Taiwan 的 43 个生成数据集进行案例研究，比较三种充电规则（即满充规则、柔性充电规则、动态充电规则）和两种算法（即粒子群优化和 SSO）的有效性和效率，结果表明在所有场景中的性能都优于 为公交运营商的电动公交车充电调度提供了有效的决策支持。