Scholarlink

<p>Multi-compartment vehicles are widely used in logistics to transport products with incompatible qualities or types, which necessitate separation into distinct compartments. This practice tackles the fundamental challenge of the multi-compartment vehicle routing problem (MCVRP), a complex logistical issue that this study seeks to analyze in depth. Our research extends the core MCVRP framework by integrating a broader range of real-world complexities, including product variety (multicommodity), divisible customer demand, time windows, and the use of a heterogeneous fleet of vehicles. To tackle the intricacies of the rich MCVRP (rMCVRP), we propose an enhanced variable neighborhood search (VNS) algorithm that ensures both scalability and robustness. This study pioneers the evaluation of the algorithm's performance against established VRP benchmarks, specifically adapted to address the unique challenges of the rMCVRP. Prior to addressing the rMCVRP, the VNS algorithm is validated on standard multi-compartment VRP scenarios to establish baseline performance. The VNS algorithm demonstrates effectiveness and efficiency on specific test instances, as evidenced by our comprehensive analysis and comparison with core MCVRP benchmarks. The results indicate that the proposed VNS algorithm constitutes a notable advancement in logistics optimization, showcasing its ability to effectively manage the complexity and diversity inherent in contemporary routing problems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An enhanced variable neighborhood search algorithm for rich multi-compartment vehicle routing problems

  • 期刊论文

  • 发表日期：

    2025年12月12日

查看全文
  • Olcay Polat,
  • Can B. Kalayci,
  • Duygu Topaloğlu

摘要

Multi-compartment vehicles are widely used in logistics to transport products with incompatible qualities or types, which necessitate separation into distinct compartments. This practice tackles the fundamental challenge of the multi-compartment vehicle routing problem (MCVRP), a complex logistical issue that this study seeks to analyze in depth. Our research extends the core MCVRP framework by integrating a broader range of real-world complexities, including product variety (multicommodity), divisible customer demand, time windows, and the use of a heterogeneous fleet of vehicles. To tackle the intricacies of the rich MCVRP (rMCVRP), we propose an enhanced variable neighborhood search (VNS) algorithm that ensures both scalability and robustness. This study pioneers the evaluation of the algorithm's performance against established VRP benchmarks, specifically adapted to address the unique challenges of the rMCVRP. Prior to addressing the rMCVRP, the VNS algorithm is validated on standard multi-compartment VRP scenarios to establish baseline performance. The VNS algorithm demonstrates effectiveness and efficiency on specific test instances, as evidenced by our comprehensive analysis and comparison with core MCVRP benchmarks. The results indicate that the proposed VNS algorithm constitutes a notable advancement in logistics optimization, showcasing its ability to effectively manage the complexity and diversity inherent in contemporary routing problems.