An integrated methodological approach for optimising complex systems subjected to predictive maintenance. (December 2021)

• Record Type: Journal Article
• Authors: Ahmed, Umair; Carpitella, Silvia; Certa, Antonella
• Is Part Of: Reliability engineering & system safety. Volume 216(2021)
• Publication Date: 2021-12
• Subjects: Complex systems -- Maintenance management -- Risk management -- Predictive maintenance -- FMECA -- ELECTRE TRI -- DEMATEL; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2021.108022
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: This paper treats the topic of complex systems subjected to predictive maintenance. A hybrid MCDM approach is integrated within the framework of traditional FMECA. The ELECTRE TRI aims to sort the identified failure modes to ordered risk classes. The DEMATEL analyses relations of dependence among failures within each class. Abstract: The present paper addresses the relevant topic of maintenance management, widely recognised as a fundamental issue involving complex engineering systems and leading companies towards the optimisation of their assets while pursuing cost efficiency. With this regard, our research aims to provide companies with a hybrid methodological approach based on Multi-Criteria Decision-Making (MCDM) capable to deal with the main failures potentially involving complex systems subjected to predictive maintenance. Such an approach is going to be integrated within the framework of traditional Failure Mode Effects and Criticality Analysis (FMECA), whose strengths and weaknesses are considered. In particular, the ELimination Et Choix Traduisant la REalité (ELECTRE) TRI is suggested to sort failure modes into risk priority classes while the Decision Making Trial and Evaluation Laboratory (DEMATEL) is proposed to highlight the most influencing failures within each risk class. The approach is applied to a real service system, whose critical components are monitored by sensors and subjected to predictive maintenance. Final results clearly demonstrate as highlighting the elements impacting the occurrence of other failures within specific risk classes is a significant driver towards the implementation of effective maintenance, maximising the whole level of performance of the analysed system over its lifecycle.