目的 针对汽车零部件物流包装中因纸箱规格设计不合理导致的包装材料浪费和箱内空间利用率低等问题,考虑到实际场景中订单内多品类零部件异构混装的特性,构建以总包装成本最小化为目标的优化模型,探索高效求解物流纸箱规格设计方案方法。方法 首先,基于历史订单数据构建包装成本优化模型;其次,采用Sobol序列生成均匀初始种群,弥补随机初始化不足;接着,在差分进化算法中引入Q-Learning调控机制,实现对关键参数的动态自适应调整,从而平衡全局搜索与局部优化能力。最后,基于降序最佳适应策略,求解满足几何与重量约束下的混合装箱方案及实际用箱数量。结果 仿真实验表明,本文算法在收敛速度与寻优精度上均明显优于传统遗传算法、模拟退火算法及常规差分进化算法;与原有方案相比,优化物流纸箱规格后,同批订单总包装成本可降低约53%。结论 该方法适用于高频波动订单、产品尺寸跨度大、多规格产品等复杂物流包装场景,通过优化箱型设计实现降本增效并提高物流效率。
Abstract
To address the issues of packaging material waste and low container space utilization caused by unreasonable carton specification design in automotive parts logistics, the work aims to construct an optimization model with the goal of minimizing the total packaging costs, explicitly considering the characteristics of heterogeneous mixed packing of multi-category components in real-world scenarios, to explore efficient methods for solving logistics carton size design schemes. Firstly, a packaging cost optimization model was established based on enterprise orders and product dimension data. The Sobol sequence was then employed to generate a uniform initial population to compensate for the deficiencies of random initialization. Subsequently, a Q-Learning control mechanism was integrated into the Differential Evolution algorithm to achieve dynamic adaptive adjustment of key parameters, thereby balancing global search and local optimization capabilities. Furthermore, a constructive greedy packing strategy was utilized to solve for the mixed packing scheme and actual carton quantity in line with geometric and weight constraints. Simulation experiments demonstrated that the proposed algorithm significantly outperformed the traditional Genetic Algorithm, Simulated Annealing, and conventional Differential Evolution algorithm in terms of convergence speed and optimization accuracy. Compared with the original scheme, the total packaging cost for the same batch of orders was reduced by approximately 53% after the optimization of logistics carton specifications. This method is applicable to complex logistics packaging scenarios characterized by high-frequency fluctuating orders, large spans of product dimensions, and multi-specification products, effectively achieving cost reduction and efficiency enhancement through optimized carton type design.
关键词
物流纸箱 /
规格优化 /
差分进化算法 /
Q-Learning /
Sobol序列
Key words
logistics carton /
specification optimization /
differential evolution algorithm /
Q-Learning /
Sobol sequence
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 曾敏刚, 朱佳. 汽车零部件运输包装尺寸标准化研究[J]. 工业工程与管理, 2013, 18(2): 31-38.
ZENG M G, ZHU J.Research on Transport Packaging Dimension Standardization of Auto Parts[J]. Industrial Engineering and Management, 2013, 18(2): 31-38.
[2] 宇可, 王艳芳, 赵小兵, 等. 海上运输补给物资包装与物流模数研究[J]. 包装工程, 2014, 35(3): 143-147.
YU K, WANG Y F, ZHAO X B, et al.Packaging and Logistics Modular of the Military Material Transportation at Sea[J]. Packaging Engineering, 2014, 35(3): 143-147.
[3] BRINKER J, GÜNDÜZ H I. Optimization of Demand-Related Packaging Sizes Using a P-Median Approach[J]. The International Journal of Advanced Manufacturing Technology, 2016, 87(5): 2259-2268.
[4] WONG W K, LEUNG S Y S. Carton Box Optimization Problem of VMI-Based Apparel Supply Chain[C]// 2006 IEEE International Conference on Management of Innovation and Technology. Singapore. IEEE, 2006: 911-915.
[5] LI Y Y, ZHANG X D, WANG P.A Cost-Minimization Model to Optimal Packaging Size in E-Commerce Context[C]// Proceedings of the 2019 Annual Meeting on Management Engineering. Kuala Lumpur Malaysia. ACM, 2019: 35-41.
[6] SINGH M, ARDJMAND E.Carton Set Optimization in E-Commerce Warehouses: A Case Study[J]. Journal of Business Logistics, 2020, 41(3): 222-235.
[7] 柳雅真, 王利强. 面向批量订单包装的物流箱规格优化问题研究[J]. 包装工程, 2023, 44(17): 229-236.
LIU Y Z, WANG L Q.Optimization of Logistics Bin Specification for Batch Order Packaging[J]. Packaging Engineering, 2023, 44(17): 229-236.
[8] DAS J N, TIWARI M K, SINHA A K, et al.Integrated Warehouse Assignment and Carton Configuration Optimization Using Deep Clustering-Based Evolutionary Algorithms[J]. Expert Systems with Applications, 2023, 212: 118680.
[9] 朱姗, 张博, 胡祥培. “一地多仓”型网上超市多品订单的拆分优化决策方法[J]. 中国管理科学, 2024, 32(4): 250-260.
ZHU S, ZHANG B, HU X P.Order Splitting Optimization Method of Multi-Item Order Fulfillment in Online Supermarkets with Multi-Warehouses in a City[J]. Chinese Journal of Management Science, 2024, 32(4): 250-260.
[10] YANG Z F, YANG S, SONG S, et al.PackerBot: Variable-Sized Product Packing with Heuristic Deep Reinforcement Learning[C]// 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Prague, Czech Republic. IEEE, 2021: 5002-5008.
[11] 任庆欣, 冯锋. 基于Sobol-Halton序列ZOA-GWO的WSN覆盖研究[J]. 计算机技术与发展, 2025, 35(5): 1-8.
REN Q X, FENG F.Research on WSN Coverage Based on Sobol-Halton Sequence ZOA-GWO[J]. Computer Technology and Development, 2025, 35(5): 1-8.
[12] 陈佳雯, 祝欣, 汤正阳, 等. 基于RLDE算法的梯级水库发电优化调度方法[J]. 长江科学院院报, 2025, 42(6): 210-218.
CHEN J W, ZHU X, TANG Z Y, et al.Optimal Scheduling Method for Power Generation of Cascade Reservoirs Based on RLDE Algorithm[J]. Journal of Changjiang River Scientific Research Institute, 2025, 42(6): 210-218.
[13] MARTELLO S, MONACI M, VIGO D.An Exact Approach to the Strip-Packing Problem[J]. INFORMS Journal on Computing, 2003, 15(3): 310-319.
[14] YANG Y, WU Z L, HAO X D, et al.Two-Layer Heuristic for the Three-Dimensional Bin Design and Packing Problem[J]. Engineering Optimization, 2024, 56(10): 1601-1638.
PDF(913 KB)
渝公网安备 50010702501716号 渝ICP备15012534号-2
