Mengyu Jin , Peng Zhang , Youlong Lv , Ming Wang , Wenbing Xiang , Hongsen Li , Jie Zhang
{"title":"A hybrid surrogate-assisted dual-population co-evolutionary algorithm for multi-area integrated scheduling in wafer fabs","authors":"Mengyu Jin , Peng Zhang , Youlong Lv , Ming Wang , Wenbing Xiang , Hongsen Li , Jie Zhang","doi":"10.1016/j.swevo.2025.102016","DOIUrl":null,"url":null,"abstract":"<div><div>In wafer fabrication, multiple areas handle different processes and production flows. To maintain the desired chemical and physical properties of wafers, strict time window constraints (TWCs) must be observed as wafers progress through these areas. However, independent scheduling within each area without collaboration complicates resource allocation and hinders overall production optimization. Implementing multi-area integrated scheduling is thus essential for effective production management, aiming to reduce total lead time and production costs. This paper proposes a hybrid surrogate-assisted dual-population co-evolutionary algorithm (HSA-DPEA) to efficiently tackle the multi-area integrated scheduling problem under multiple TWCs. The algorithm employs a dual-population co-evolutionary mechanism, consisting of normal and auxiliary populations, to balance convergence and diversity while ensuring feasibility. The normal population focuses on feasible solutions to maintain overall quality, while the auxiliary population explores infeasible regions to identify promising individuals that can guide the normal population's evolution. To enhance evolutionary efficiency and reduce the number of time-consuming real fitness evaluations, a hybrid surrogate-assisted model is introduced. This model adapts by training regression or classification models at different stages of population evolution. Additionally, an online learning strategy based on convergence and diversity is employed for continuous model updating to improve accuracy. The proposed algorithm is tested on 18 instances and validated through six months of continuous testing on a wafer fab simulation system. The results demonstrate that HSA-DPEA obtains better Pareto optimal sets, effectively reducing total lead time and production costs in multi-area integrated scheduling under multiple TWCs.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 102016"},"PeriodicalIF":8.2000,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Swarm and Evolutionary Computation","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2210650225001749","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0
Abstract
In wafer fabrication, multiple areas handle different processes and production flows. To maintain the desired chemical and physical properties of wafers, strict time window constraints (TWCs) must be observed as wafers progress through these areas. However, independent scheduling within each area without collaboration complicates resource allocation and hinders overall production optimization. Implementing multi-area integrated scheduling is thus essential for effective production management, aiming to reduce total lead time and production costs. This paper proposes a hybrid surrogate-assisted dual-population co-evolutionary algorithm (HSA-DPEA) to efficiently tackle the multi-area integrated scheduling problem under multiple TWCs. The algorithm employs a dual-population co-evolutionary mechanism, consisting of normal and auxiliary populations, to balance convergence and diversity while ensuring feasibility. The normal population focuses on feasible solutions to maintain overall quality, while the auxiliary population explores infeasible regions to identify promising individuals that can guide the normal population's evolution. To enhance evolutionary efficiency and reduce the number of time-consuming real fitness evaluations, a hybrid surrogate-assisted model is introduced. This model adapts by training regression or classification models at different stages of population evolution. Additionally, an online learning strategy based on convergence and diversity is employed for continuous model updating to improve accuracy. The proposed algorithm is tested on 18 instances and validated through six months of continuous testing on a wafer fab simulation system. The results demonstrate that HSA-DPEA obtains better Pareto optimal sets, effectively reducing total lead time and production costs in multi-area integrated scheduling under multiple TWCs.
期刊介绍:
Swarm and Evolutionary Computation is a pioneering peer-reviewed journal focused on the latest research and advancements in nature-inspired intelligent computation using swarm and evolutionary algorithms. It covers theoretical, experimental, and practical aspects of these paradigms and their hybrids, promoting interdisciplinary research. The journal prioritizes the publication of high-quality, original articles that push the boundaries of evolutionary computation and swarm intelligence. Additionally, it welcomes survey papers on current topics and novel applications. Topics of interest include but are not limited to: Genetic Algorithms, and Genetic Programming, Evolution Strategies, and Evolutionary Programming, Differential Evolution, Artificial Immune Systems, Particle Swarms, Ant Colony, Bacterial Foraging, Artificial Bees, Fireflies Algorithm, Harmony Search, Artificial Life, Digital Organisms, Estimation of Distribution Algorithms, Stochastic Diffusion Search, Quantum Computing, Nano Computing, Membrane Computing, Human-centric Computing, Hybridization of Algorithms, Memetic Computing, Autonomic Computing, Self-organizing systems, Combinatorial, Discrete, Binary, Constrained, Multi-objective, Multi-modal, Dynamic, and Large-scale Optimization.