A cost-informed component maintenance index and its applications

All systems and components are unreliable in the sense that they will fail. While a failed component in a system is being repaired (i.e., corrective maintenance), preventive maintenance (PM) may be conducted on the other components to improve the reliability of the system. The selection of different components for PM may result in a variety of maintenance policies with different cost implications. It is therefore necessary to develop appropriate tools such as importance measures to guide engineers in selecting components for PM in order to minimize relevant costs. There is little research, nevertheless, that jointly minimizes the total expected cost of maintenance and maximizes the number of components for PM. To fill in this gap, this paper proposes an importance index, Cost-Informed Component Maintenance Index (CICMI). It then derives some propositions of the proposed index and different maintenance policies, respectively. A method to optimize the number of components for PM subject to cost constraints is then proposed. A case study on a reactor coolant system is performed to illustrate the applicability of the proposed methods.

Automated summary

All components and systems are prone to failure. While repairing a failed component, preventive maintenance can be conducted on other components to enhance system reliability. Choosing different components for preventive maintenance may result in different maintenance policies with varying costs. To minimize costs, engineers need appropriate tools such as importance measures to guide their selection of components for preventive maintenance. However, research that simultaneously minimizes maintenance costs and maximizes the number of components for preventive maintenance is limited. To address this gap, this paper proposes the Cost-Informed Component Maintenance Index (CICMI) as an importance index and derives propositions for this index and different maintenance policies. A method is also proposed to optimize the number of components for preventive maintenance within cost constraints. A case study on a reactor coolant system is conducted to demonstrate the applicability of the proposed methods.