ISSN Approved Journal || eISSN: 2582-8185 || CODEN: IJSRO2 || Impact Factor 8.2 || Google Scholar and CrossRef Indexed
Fast Publication within 48 hours || Low Article Processing Charges || Peer Reviewed and Referred Journal || Free Certificate
Dynamic reliability-centered maintenance modeling integrating failure mode analysis and Bayesian decision theoretic approaches
1 Department of Mechanical and Industrial Systems Engineering, Tagliatelle College of Engineering, University of New Haven, USA.
2 Department of Electronics and Information Technology, University of South Wales, UK.
Research Article
International Journal of Science and Research Archive, 2023, 08(01), 1117-1135.
Article DOI: 10.30574/ijsra.2023.8.1.0136
Publication history:
Received on 14 January 2023; revised on 21 February 2023; accepted on 26 February 2023
Abstract:
As industries face increasing pressure to maintain complex systems with minimal downtime and optimized cost structures, traditional static maintenance strategies fall short of addressing real-time uncertainty and evolving operational conditions. Reliability-Centered Maintenance (RCM), while foundational, must adapt to incorporate probabilistic reasoning and continuous decision-making to remain effective. This paper presents a dynamic RCM modeling framework that integrates Failure Mode and Effects Analysis (FMEA) with Bayesian decision-theoretic approaches to enable real-time, risk-informed maintenance interventions. The model begins with a comprehensive failure mode mapping using FMEA to identify critical assets, failure causes, and effects. Each failure mode is assigned dynamic risk priority numbers (RPNs) that evolve based on operational data, sensor inputs, and environmental variability. A Bayesian belief network is layered onto this framework to update prior failure probabilities as new data becomes available, capturing the stochastic nature of degradation. Decision nodes within the Bayesian structure enable cost-risk trade-offs to be evaluated in real time, ensuring that optimal maintenance actions are selected under uncertainty. Furthermore, the model accommodates adaptive learning through posterior updates, refining both failure predictions and policy recommendations over time. A case study involving an energy generation plant demonstrates a 31% improvement in mean time between failures (MTBF) and a 24% reduction in maintenance costs. The dynamic fusion of qualitative failure analysis with quantitative Bayesian inference allows for smarter, context-aware decision-making and predictive readiness. This study provides a robust framework for implementing intelligent RCM strategies in Industry 4.0 environments where responsiveness, resilience, and safety are paramount.
Keywords:
Reliability-Centered Maintenance; Failure Mode Analysis; Bayesian Networks; Predictive Modeling; Decision Theory; Dynamic Maintenance Planning
Full text article in PDF:
Copyright information:
Copyright © 2023 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0
