Root cause analysis based maintenance policy

Abstract

Purpose

The paper seeks to ease the implementation of predictive maintenance policy in industry using the root cause analysis technique, and to compare the reliability and cost effectiveness of root cause based maintenance (RCBM) relative to conventional corrective maintenance (CM).

Design/methodology/approach

The system is modularized into its components and maintenance schedules are developed based on each component's individual degradation trends. The effectiveness of RCBM over CM is studied by analyzing system reliability patterns and total maintenance cost functions obtained through empirical cost models, accounting for yield and production loss, maintenance, replacement and catastrophic failure costs. Cost variations for various possible failure distribution parameter values (β, η) under the CM and RCBM policies are also obtained. The proposed methodology is tested in a real aircraft failures case study.

Findings

RCBM is generally more effective over CM in achieving timely maintenance at optimal cost (savings up to 65 percent) while keeping high system reliability, for a wide range of (β, η) values. However, CM could still be beneficial for a restricted range of large (β, η).

Practical implications

Industry should consider shifting from CM to adopt the proposed RCBM policy, which is proved to be more efficient in most cases. The implementation is not necessarily complex.

Originality/value

The effectiveness of RCBM over CM in terms of reliability and cost considerations is clearly illustrated. This paper justifies the need to shift from CM to RCBM, which brings us closer to a practical implementation of predictive maintenance. This work also serves as a simple and valuable guide to implementation for maintenance and operational managers in production industries.

Citation

Ming Tan, C. and Raghavan, N. (2007), "Root cause analysis based maintenance policy", International Journal of Quality & Reliability Management, Vol. 24 No. 2, pp. 203-228. https://doi.org/10.1108/02656710710722293