< Back to previous page


A cost-based failure prioritization approach for selecting maintenance strategies for thermal power plants: a case study context of developing countries

Journal Contribution - Journal Article

The failure mode and effect analysis (FMEA) employing the risk priority numbers (RPN) has been used extensively for identifying and prioritizing failure modes with a view of mitigating their impact on equipment failure. However, in its traditional form, the prioritization approach through the RPN lacks the objectivity required for robust risk assessment, more so, where maintenance data is available, which could enhance such objectivity. This paper extends a quantitative approach for prioritizing failure modes and component failures in facilities, and more specifically, leverages on maintenance data often recorded in such facilities. To enhance the objectivity of the risk prioritization process, the proposed approach integrates three objective measures—the cost of failure, failure occurrence rate and percentage downtime effects of equipment failure. The integrated measures are demonstrated as more robust for prioritizing risks as opposed to ordinal indices as the case in the conventional FMEA approach. Using historical maintenance records, a three-step ranking approach is proposed for prioritizing critical failure modes in a thermal power plant where a case study is discussed. Moreover, the study compares the results derived from the prioritization approach with that derived utilizing the conventional RPN method. The comparative study demonstrates the added value of a more objective and quantitative prioritization approach for maintenance decision support. Ultimately, the critical failure modes are evaluated using a decision scheme to allocate appropriate maintenance strategies as the final step of risk assessment (i.e. risk treatment). The proposed approach is viewed as generalizable, intuitive and offering insights to the maintenance practitioners.
Journal: International Journal of System Assurance Engineering and Management
ISSN: 0975-6809
Issue: 5
Volume: 10
Pages: 1 - 19
Publication year:2019