Predictive Maintenance Model for Centrifugal Pumps under Improper Maintenance Conditions

Nazmee Hashim, Adnan Hassan and Mohd Foad Abdul Hamid (2020), Predictive Maintenance Model for Centrifugal Pumps under Improper Maintenance Conditions, AIP Conference Proceedings 2217, 030170 (1-6); (Scopus)

(link to publisher)


Industrial equipment reliability can be improved through understanding the underlying root causes of system failures and by implementing the predictive maintenance strategy. Failures and maintenance activities data are normally available in industries; however proper failure causes analysis is still lagging. As such, the benefits of predictive maintenance strategy could not be realised. This has contributed to high operational and maintenance cost and not sustainable in the long term. The objective of the paper is to propose a customised predictive maintenance model for minimising the overall maintenance cost of centrifugal pumps in an oleo-chemical plant in Johor, Malaysia. About 14% of total maintenance cost in this plant was attributed to centrifugal pumps related maintenance. Currently, repeated components failures were solved through firefighting without addressing the actual root causes. In this study, the Failure Modes Effects and Criticality Analysis (FMECA) was used to identify critical components interactive failures, namely the mechanical seal, bearing and shaft of the centrifugal pumps. The failure distribution functions for the series reliability structure was formulated after thorough statistical tests on the historical data. The potential benefit of the proposed model is illustrated with alternative maintenance interventions to achieve acceptable pump reliability levels at minimum cost. The findings from this study should be beneficial to maintenance practitioners and researchers toward achieving sustainable operations.