Maintenance Strategy Optimization Using Failure Rate Analysis and Risk-Based Inspection: Review, Gap Analysis, Framework Development, and Case Study

Abstract

Optimizing maintenance in asset-intensive industries requires integrating reliability, safety, and cost under uncertain, time-dependent degradation. Conventional strategies often fail to capture stochastic failure behavior, limiting predictive capability and leading to suboptimal inspection and replacement decisions. Although Risk- Based Inspection (RBI) enhances asset prioritization, its implementation often relies on static assumptions, fragmented data structures, and limited adaptability. A review of maintenance optimization, failure-rate modeling, and RBI applications reveals key gaps: isolated deployment of Reliability-Centered Maintenance (RCM), Total Productive Maintenance (TPM), and RBI; insufficient integration of predictive analytics; limited quantification of cost–risk–reliability trade-offs; and inadequate modeling of system-level interdependencies. These gaps constrain the evolution toward dynamic, risk-informed maintenance systems. This study proposes an Integrated Reliability-Centered Framework (IRCF) that unifies probabilistic reliability modeling, RBI principles, and predictive condition monitoring in a continuous, data-driven architecture. The framework integrates time-dependent failure rate estimation, quantitative risk assessment, and Remaining Useful Life (RUL) prediction, with feedback mechanisms that dynamically update risk rankings and inspection intervals. Application of the IRCF to a feedwater centrifugal pump station in a steam generation system reduced downtime by ~25% and maintenance costs by ~20%, while enhancing safety through risk-prioritized interventions. Inspection intervals derived from Probability of Failure (PoF) and RUL metrics optimized resource allocation and minimized unnecessary preventive maintenance.