Joint production, quality and condition-based maintenance strategy for manufacturing system considering the dynamic customer demand under delivery performance

•Joint strategy integrates production, quality and condition-based maintenance.•The novel model considering the dynamic customer demand under delivery performance.•Nonconforming products intercepts serve as a threshold for maintenance intervention during processing.•Improved deep reinforcement learn...

Full description

Saved in:
Bibliographic Details
Published in:Computers & industrial engineering Vol. 207; p. 111271
Main Authors: Li, Sen, Yang, Zhaojun, He, Jialong, Ma, Chi, Wang, Liang, Li, Guofa, Liu, Yan, Nie, Xinyu
Format: Journal Article
Language:English
Published: Elsevier Ltd 01.09.2025
Subjects:
Deep reinforcement learning
Delivery performance
Dynamic customer demand
Manufacturing system optimization
Production-quality-maintenance integration
Deep reinforcement learning
Manufacturing system optimization
Production-quality-maintenance integration
Dynamic customer demand
Delivery performance
ISSN:0360-8352
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Joint strategy integrates production, quality and condition-based maintenance.•The novel model considering the dynamic customer demand under delivery performance.•Nonconforming products intercepts serve as a threshold for maintenance intervention during processing.•Improved deep reinforcement learning algorithm is proposed to address “curse of dimensionality”. Modern manufacturing systems face the critical challenge of fulfilling customer demand while maintaining product quality, delivery performance, and equipment health under uncertainty. Frequent maintenance improves quality but may interrupt production and cause delivery delays. Conversely, minimizing downtime to ensure timely delivery can accelerate degradation and reduce quality. These trade-offs are further complicated by dynamic customer demand—poor delivery performance erodes trust and reduces demand, while high-quality delivery builds trust and increases future orders. To address these challenges, this paper proposes a joint strategy that integrates production, quality control, and condition-based maintenance (CBM), with a particular focus on dynamic customer demand influenced by delivery performance. The strategy is modeled using a Markov decision process (MDP), where product quality acts as an indirect indicator of system health, enabling timely maintenance decisions during production. To solve this high-dimensional joint decision problem, an improved deep reinforcement learning algorithm is developed, incorporating a dueling double Q-network, enhanced sampling, and a revised loss function. A numerical study validated the effectiveness and superiority of the proposed algorithm and strategy, and examined the impact of customer emphasis on delivery performance through sensitivity analysis.
ISSN:0360-8352
DOI:10.1016/j.cie.2025.111271