| 多目标公交车辆与司机调度问题元启发算法设计 |
| |
| 引用本文： | 孔云峰.多目标公交车辆与司机调度问题元启发算法设计[J].交通信息与安全,2021,39(3):50-59. |
| |
| 作者姓名： | 孔云峰 |
| |
| 作者单位： | 河南大学黄河中下游地理信息技术教育部重点实验室 河南 开封 475000 |
| |
| 基金项目： | 河南省自然科学基金项目182300410132 |
| |
| 摘 要： | 公交车辆与司机调度问题是智慧公交管理中的核心问题之一。针对我国人车固定作业模式下，相关研究中成本考虑不周全、算法通用性差和算法测试不充分等局限，设计了1个多目标公交车辆与司机调度问题元启发算法。算法支持电动车辆调度，适用于单线或跨线运营管理，满足人车固定或人车分离的调度模式，也支持灵活的车辆与司机相关参数设置。算法顾及车辆停车间隔、电池充电、司机休息与就餐等约束条件，优化目标包括车辆固定成本、车辆行驶成本、司机固定成本和司机津贴成本。算法首先生成初始解、再迭代使用班次链算子改进当前解，并通过群解、扰动和可变邻域下降等策略改进解的质量。使用62个单线案例和11个跨线案例进行算法测试，验证了算法的性能，并比较了不同运营模式下调度结果的差异。结果表明，使用续航里程150 km电动车辆取代燃油车辆，单线运营车辆数量增幅为0.8%，跨线运营增幅为1.6%；与单线运营相比，跨线运营所需车辆和司机数量分别减少4.6%和2.4%；与燃油车辆人车固定调度模式相比，人车分离能显著减少所需车辆，单线运营减少3.6%，跨线运营减少1.8%，所需司机数量基本保持不变，但司机需要换车驾驶，平均约为2次。
|
| 关 键 词： | 城市公共交通 车辆与司机调度问题 算法设计 算法测试 |
| 收稿时间： | 2020-10-14 |
A Metaheuristic Algorithm for Multi-objective Transit Bus and Driver Scheduling Problems |
| |
| KONG Yunfeng.A Metaheuristic Algorithm for Multi-objective Transit Bus and Driver Scheduling Problems[J].Journal of Transport Information and Safety,2021,39(3):50-59. |
| |
| Authors: | KONG Yunfeng |
| |
| Affiliation: | Key Laboratory of Geospatial Technology for the Middle and Lower Yellow River Regions of Ministry of Education, Henan University, Kaifeng 475000, Henan, China |
| |
| Abstract: | This article introduces a metaheuristic algorithm for multi-objective transit bus and driver scheduling problems, such as fuel or electronic vehicles, single route/multiple routes, and driving the same bus on the same day in most transit companies in China.The work aims to minimize the fixed bus cost, the bus travel cost, the fixed driver cost, and the allowance for drivers and to satisfy various operational rules on vehicles and drivers.The algorithm starts from an initial solution and iteratively improves the solution by local search and perturbation.It is also enhanced by two search strategies such as population-based search and variable neighborhood decent search.The performance of the proposed algorithm is tested on 62 single-route instances and 11 multi-route instances.There are three important findings for transit operations in China from the experimentation.Electronic vehicles may replace fuel buses by increasing 0.8% and 1.6% vehicles for single-route and multi-route instances, respectively.Compared with single-route scheduling, multi-route scheduling has the potentials to reduce 4.6% of vehicles and 2.4% of drivers.If the drivers are allowed to drive different buses in their daily works, the number of vehicles required can be reduced significantly, especially for the single-route instances.The general-purpose metaheuristic algorithm in the work is essential for developing intelligent public transit systems in China. |
| |
| Keywords: | |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《交通信息与安全》浏览原始摘要信息 |
|
点击此处可从《交通信息与安全》下载全文 |
|
