Multi-objective Human-robot collaborative batch scheduling in distributed hybrid flowshop via automatic design of local search-reconstruction-feedback algorithm

•The model of multi-objective DHFBSP_HC is constructed.•A configurable local search-reconstruction-feedback algorithm is developed.•An automated algorithm design is embedded into the algorithm designing. The emergence of distributed production models has spurred extensive research on distributed hyb...

Full description

Saved in:

Bibliographic Details
Published in:	Computers & industrial engineering Vol. 203; p. 110983
Main Authors:	He, Peng, Jiang, Xuchu, Wang, Qi, Zhang, Biao
Format:	Journal Article
Language:	English
Published:	Elsevier Ltd 01.05.2025
Subjects:	Automated algorithm design Distributed hybrid flowshop scheduling Human-robot collaborative scheduling Multi-objective optimization Distributed hybrid flowshop scheduling Multi-objective optimization Automated algorithm design Human-robot collaborative scheduling
ISSN:	0360-8352
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	•The model of multi-objective DHFBSP_HC is constructed.•A configurable local search-reconstruction-feedback algorithm is developed.•An automated algorithm design is embedded into the algorithm designing. The emergence of distributed production models has spurred extensive research on distributed hybrid flowshop scheduling. Despite advancements in resource allocation for flowshops, most studies overlook human-robot collaboration, which remains essential for complex manufacturing processes in real-world production. Additionally, the rise of multi-variety, small-batch production has driven widespread adoption of batch scheduling. Therefore, this paper introduces a multi-objective distributed hybrid flowshop batch scheduling problem with human-robot collaboration (DHFBSP_HC), aiming to minimize makespan and total energy consumption simultaneously. To address this issue, we propose a local search-reconstruction-feedback (LSRF) algorithm, which consists of four core components: population initialization, local search, reconstruction, and feedback mechanism. Additionally, the algorithm incorporates three configurable strategies, including fitness evaluation approaches, initialization approaches, and objective normalization approaches. These configurable strategies are regarded as categorical parameters, whereas the other parameters are referred to as numerical parameters. To select categorical and numerical parameters that can optimize the results of multi-objective DHFBSP_HC, we introduce the automated algorithm design and use I/F-Race to optimize parameter settings. Through comparisons with several state-of-the-art algorithms, we demonstrate the effectiveness and superiority of the LSRF algorithm in solving the multi-objective DHFBSP_HC.
ISSN:	0360-8352
DOI:	10.1016/j.cie.2025.110983