基于问题性质的装配式预制件配送优化算法

Property-based assembled prefabricated components scheduling algorithm

中文关键词  异构车辆  装配式预制件  车辆调度  多频访问

英文关键词  heterogeneous vehicle  prefabricated components  vehicle scheduling  multi-frequency

中文摘要

      在传统建筑业转型升级的背景下, 新型建筑工业化推动了装配式建筑的发展. 本文针对预制件生产企业物流管理的关键环节进行了研究, 考虑异构车辆约束、装载机约束和时间窗约束, 以最小化总运输成本为目标, 提出了考虑多频次访问特性的车辆调度问题, 进而设计了一种针对问题特性的混合基因算法. 为求解该问题, 算法基于问题性质设计了三维编码方式, 以增强算法探索解空间的能力; 引入基于需求划分的多样性评估模型和自适应调整机制来保持种群多样性; 提出了4种不同的邻域策略进一步增强局部搜索能力. 最后, 在不同规模算例上进行了仿真实验与算法对比, 验证了算法的有效性, 同时, 对车辆购买类型和数量对总物流成本的影响进行了分析.

英文摘要

      In the context of the transformation and upgrading of the traditional construction industry, the new construction industrialization has promoted the development of assembled buildings. This study investigates the key aspects of logistics management in precast manufacturing enterprises. The multi-frequency vehicle scheduling problem is proposed to minimize the total transportation cost, considering the heterogeneous vehicle constraints, loader constraints, and time windows constraints. And then, a property-based hybrid genetic algorithm is designed to solve it. To solve the problem, a property-based three-dimensional encoding is designed to enlarge the search scope, a diversity assessment model based on demand partitioning and an adaptive adjustment mechanism are introduced to maintain population diversity and four different neighborhood strategies are proposed to further intensify the search in local regions. Finally, the effectiveness of the algorithm is verified by simulations experiments and algorithm comparisons on different instances. And the impact of vehicle purchase type and quantity on the transportation cost is also analyzed.