Scholarlink

<p>To tackle the inherent complexity of scheduling large-scale, high-mix, low-volume orders, this paper formulates a high-dimensional reentrant hybrid flowshop scheduling problem with incompatible job families (HRHFSP-IJF). The proposed model is tailored to satisfy multi-objective engineering requirements from both service and production perspectives. Given the NP-hard nature of HRHFSP-IJF, a multi-strategy adaptive cultural algorithm (MSACA) is developed. This algorithm leverages three strategy pools—multi-heuristic decoding, multi-level group interaction, and success–failure memory in the belief space-to guide population initialization, global search, and local search. First, four decoding strategies are designed to manage priority rules, machine assignments, idle area insertions, and job sequencing for both initialization and iterative updating. Second, interactive individuals are chosen based on non-dominated ranks and Hamming distance, with knowledge transfer facilitating effective cultural interactions. Third, the algorithm employs a dominance-based measure of success/failure and probability learning to adaptively choose variable neighborhood search operators, thereby enhancing the local exploitation of the population. Scheduling experiments on various scales demonstrate that MSACA outperforms several competitive algorithms in terms of convergence, diversity, and dominance, offering a viable solution framework for HRHFSP-IJF.</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

A multi-strategy adaptive cultural algorithm for scheduling high-dimensional reentrant hybrid flowshops with incompatible job families

  • 期刊论文

  • 发表日期：

    2025年7月8日

查看全文
  • Jiawei Wu,
  • Yong Liu

摘要

To tackle the inherent complexity of scheduling large-scale, high-mix, low-volume orders, this paper formulates a high-dimensional reentrant hybrid flowshop scheduling problem with incompatible job families (HRHFSP-IJF). The proposed model is tailored to satisfy multi-objective engineering requirements from both service and production perspectives. Given the NP-hard nature of HRHFSP-IJF, a multi-strategy adaptive cultural algorithm (MSACA) is developed. This algorithm leverages three strategy pools—multi-heuristic decoding, multi-level group interaction, and success–failure memory in the belief space-to guide population initialization, global search, and local search. First, four decoding strategies are designed to manage priority rules, machine assignments, idle area insertions, and job sequencing for both initialization and iterative updating. Second, interactive individuals are chosen based on non-dominated ranks and Hamming distance, with knowledge transfer facilitating effective cultural interactions. Third, the algorithm employs a dominance-based measure of success/failure and probability learning to adaptively choose variable neighborhood search operators, thereby enhancing the local exploitation of the population. Scheduling experiments on various scales demonstrate that MSACA outperforms several competitive algorithms in terms of convergence, diversity, and dominance, offering a viable solution framework for HRHFSP-IJF.