文献类型：期刊文献

中文题名：面向B2B电商供需匹配优化的遗传人类学习混合算法

英文题名：Genetic Human Learning Hybrid Algorithm for Optimizing Supply and Demand Matching in B2B E-commerce

作者：寇嘉敏[1];肖明明[1];吕兆峰[2];王文文[2];张磊[2];李克[1]

第一作者：寇嘉敏

机构：[1]北京联合大学智慧城市学院,北京100101;[2]鲁班(北京)电子商务科技有限公司,北京102308

第一机构：北京联合大学智慧城市学院

年份：2025

卷号：43

期号：5

起止页码：97-106

中文期刊名：系统工程

外文期刊名：Systems Engineering

收录：;北大核心:【北大核心2023】；

基金：国家自然基金资助项目(61972040);北京联合大学校级科研项目(ZK30202304);中铁物贸集团鲁班公司科技研究开发计划课题。

语种：中文

中文关键词：供应链网络设计;自适应人类学习优化算法;遗传算法;供需匹配;B2B电商

外文关键词：Supply Chain Network Design(SCND);Adaptive Human Learning Optimization Algorithm;Genetic Algorithms;Matching Supply and Demand;B2B E-commerce

摘要：在日益复杂多变的B2B大宗商品电商交易中,传统基于人工和经验的供需匹配严重制约了电商企业经营利润和服务水平的提升。本文将上述供需匹配问题建模为综合考虑供给需求匹配、产能、交付时间和信用风险等复杂约束条件下的供应链网络设计模型,并针对模型特点设计多维编码方式,提出一种改进的遗传人类学习混合算法对问题进行求解。以建筑业真实电商业务场景中的数据为例开展的实验表明,本文算法的求解精度与寻优能力在大、中、小三种不同规模的数据中均优于遗传算法、人类学习优化算法等基准算法,可有效替代行业内现行的基于经验的人工决策方式,提升B2B电商业务的智能化水平和经营能力。
In the increasingly complex and changeable B2B bulk commodity e-commerce transactions,the traditional supply-demand matching based on labor and experience seriously restricts the operating profits and service levels of e-commerce enterprises.In this paper,the above supply-demand matching problem is modeled as a supply chain network design model that comprehensively considers the complex constraints of supply-demand matching,production capacity,delivery time and credit risk,and designs a multi-dimensional coding method according to the characteristics of the model,and proposes an improved genetic human learning hybrid algorithm to solve the problem.Experiments based on data in real e-commerce business scenarios in the construction industry show that the solving accuracy and optimization ability of the proposed algorithm are better than benchmark algorithms such as genetic algorithm and human learning optimization algorithm in large,medium and small data,which can effectively replace the current experience-based manual decision-making method in the industry and improve the intelligence level and operation ability of B2B e-commerce business.