Multi-objective optimization for preventive maintenance of offshore safety critical equipment integrating dynamic risk and maintenance cost

Abstract

Safety-critical equipment (SCE) are installed on offshore installations as important safety barriers to prevent the occurrence of major accident. In order to ensure the specified safety function of SCE, preventive maintenance (PM) is an indispensable activity. However, PM decision making is still a challenge as it is influenced by various factors such as the risk in association with SCE degradation, risk in association with PM, and maintenance-related cost. Especially, the relevant risks are inherently dynamic in nature and the effects of these influence factors are conflicting. Therefore, this study focuses on the optimization of PM interval for SCE integrating the dynamic characteristics of risk and the conflicting effects among the increased risks due to PM and SCE degradation as well as maintenance-related cost. A systematic multi-objective optimization framework is proposed for the PM interval optimization. Two dynamic risk models are established for the evaluation of dynamic risk profiles of SCE in the operational phase and PM phase respectively. In combination with the lifecycle maintenance-related cost model of SCE, the multi-objective optimization with three objective functions is conducted for the PM interval optimization. The proposed framework is applied in the case study on a typical SCE (pressure safety valve, PSV) to optimize the PM interval. The effectiveness and availability of this framework is verified. It is noted that the common industrial practice of conducting PM once a year is acceptable for PSV once the risks in both operational and PM phases are given equal importance weight. However, a larger PM interval is recommended when the lifecycle maintenance-related cost is included in the optimization goal and further validation procedures are conducted, i.e. statistical analysis of optimal candidates and reassessment of risk profile with the optimal candidates. The proposed multi-optimization framework provides a useful tool for the optimal PM interval determination for the SCE on offshore installations.