中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊
重庆交通大学学报（自然科学版）

重庆交通大学学报（自然科学版） ›› 2025, Vol. 44 ›› Issue (6): 82-96.DOI: 10.3969/j.issn.1674-0696.2025.06.09

• 交通+大数据人工智能 • 上一篇    

时间依赖型同时取送货车辆路径优化策略

陈仕军1，骆维2,3，吴华伟2,3，夏良才1，王鸿禹1   

  1. (1.湖北文理学院 数学与统计学院，湖北 襄阳 441053； 2.湖北文理学院 湖北隆中实验室，湖北 襄阳 441053； 3.纯电动汽车动力系统设计与测试湖北省重点实验室，湖北 襄阳 441053)
  • 收稿日期:2024-06-24 修回日期:2024-12-13 发布日期:2025-06-30
  • 作者简介:陈仕军（1980—），男，湖北襄阳人，副教授，博士，主要从事交通优化与调度方面的研究。E-mail：csj@hbuas.edu.cn 通信作者：骆维（2000—），男，湖南岳阳人，硕士研究生，主要从事车辆路径优化方面的研究。E-mail：1986874366@qq.com
  • 基金资助:
    国家自然科学基金项目(71501064)；湖北文理学院科研能力培育基金科技创新团队项目(2020kypytd006)；湖北文理学院研究生创新计划项目(YCX202421)；湖北文理学院大学生创新创业项目(20230100243，202410519005)

Routing Optimization Strategy of Time-Dependent Simultaneous Pickup and Delivery Vehicle

CHEN Shijun1, LUO Wei2,3, WU Huawei2,3, XIA Liangcai1, WANG Hongyu1   

  1. (1. School of Mathematics and Statistics, Hubei University of Arts and Science, Xiangyang 441053, Hubei, China; 2. Hubei Longzhong Laboratory, Hubei University of Arts and Science, Xiangyang 441053, Hubei, China; 3. Hubei Key Laboratory of Pure Electric Vehicle Power System Design and Testing, Xiangyang 441053, Hubei, China)
  • Received:2024-06-24 Revised:2024-12-13 Published:2025-06-30

PDF

摘要： 时间依赖型同时取送货车辆路径问题(TDVRPSDP)研究存在着采用与实际存在偏差的“阶跃时变速度”以及忽略车速对能耗影响的缺陷。为此，在考虑连续时变车速和载重对油耗影响的基础上，建立了以车辆使用、油耗及碳排放成本之和最小化为目标的数学模型，并采用混合人工蜂群算法(HABC)进行求解；该算法采用改进的最近邻法生成优质初始蜜源，设计了多种自适应大邻域搜索算子代替标准算法中蜜蜂的随机搜索机制，添加了劣解接受准则并运用一系列优化策略提高寻优能力；通过多种算例和案例来验证该算法的有效性。研究结果表明：对于TDVRPSDP子问题测试算例，HABC优于对比算法；对于TDVRPSDP测试算例，较标准人工蜂群算法和结合大邻域搜索的人工蜂群算法，所提出的算法平均配送成本分别降低18.3%、 1.7%；在实际案例求解中，HABC也展现出较强的寻优能力和收敛速度，能为企业有效降低配送成本。

关键词: 交通运输工程；自适应大邻域搜索；人工蜂群算法；车辆路径；优化策略

Abstract: The study on the time-dependent vehicle routing problem with simultaneous delivery and pickup (TDVRPSDP) has two defects, that is, using “step time-varying speed” that deviates from reality and ignoring the impact of vehicle speed on energy consumption. To address these issues, a mathematical model considering continuous time-varying speeds and impact of load on fuel consumption was established, aiming to minimize total costs including vehicle usage, fuel consumption and carbon emissions, and a hybrid artificial bee colony (HABC) algorithm was proposed for solution. The proposed algorithm used an improved nearest-neighbor method for generating high-quality initial honey sources, designed multiple adaptive large neighborhood search operators to replace random search mechanisms in standard artificial bee colony (ABC) algorithm, added an inferior solution acceptance criterion, and adopted a series of optimization strategies to enhance search capability. The effectiveness of the proposed algorithm was verified by multiple computational experiments and case studies. The research results show that for test case of TDVRPSDP sub-problems, HABC algorithm outperforms comparative algorithms. For TDVRPSDP test case, the proposed algorithm reduces average distribution costs by 18.3% and 1.7% respectively compared with standard ABC algorithm and ABC algorithm combined with large neighborhood search. In practical case solving, HABC has also demonstrated strong optimization ability and convergence speed, which can effectively reduce distribution costs for enterprises.

Key words: traffic and transportation engineering; adaptive large neighborhood search; artificial bee colony algorithm; vehicle routing; optimization strategy

中图分类号: