Functional risk-oriented integrated preventive maintenance considering product quality loss for multistate manufacturing systems. Issue 4 (16th February 2021)

• Record Type: Journal Article
• Authors: Zhao, Yixiao; He, Yihai; Zhou, Di; Zhang, Anqi; Han, Xiao; Li, Yao; Wang, Wenzhuo
• Is Part Of: International journal of production research. Volume 59:Issue 4(2021)
• Page Start: 1003
• Page End: 1020
• Publication Date: 2021-02-16
• Subjects: multistate manufacturing systems -- functional risks -- product quality loss -- integrated maintenance -- maintenance level; Factory management -- Periodicals; 658.57
• DOI: 10.1080/00207543.2020.1713416
• Languages: English
• ISSNs: 0020-7543

Abstract

Abstract : During the abnormal operations before breakdown of a multistate manufacturing system, product quality loss could systematically characterise the affected economic cost in advance. However, few studies have considering the valuable quality loss data, thus, a novel risk-oriented optimisation approach of integrated maintenance is proposed considering the quality loss of work-in-process (WIP) obtained from the manufacturing system. First, the functional risks in manufacturing systems are expounded and classified by detectable deviations and potential manufacturing defects of the finished product based on machine performance state, and product quality loss is therefore applied for preventive maintenance activities. Second, the degradation path of machine performance and its effect on WIP are clarified. Explicit risk is presented by dimensional deviations of key quality characteristics, whereas implicit risk is quantified by the influence of component- and system-level manufacturing defects on the early product usage phase. Third, different levels of maintenance actions are proposed, and the integrated preventive maintenance plan for the entire production horizon is obtained by minimising product quality loss and PM investments. The optimal decision combination is solved by genetic algorithm. Finally, the optimal maintenance strategy is presented to verify the superiority of this method through the analysis of a cylinder head manufacturing system.