Tanwar, Monika

Purpose: The idea is to develop a condition-based maintenance (CBM) model for continuously monitored degrading systems under imperfect maintenance. The system is subjected to random shocks, inspection, corrective maintenance (CM) and preventive maintenance (PM); respective thresholds are set to make decisions. Design/methodology/approach: The generalized renewal process is considered for degradation modeling, and the maximum likelihood estimation method is used for parameter estimation. Optimum replacement time is derived by minimizing downtime per unit time (DT). Findings: Optimum replacement time is derived by minimizing DT. The paper provides insight into the virtual age model’s applicability and efficacy in CBM policy-based decision-making. Practical implications: This paper assesses the CBM policy for continuously degrading systems under random shocks and imperfect maintenance. An operating system faces five events, i.e. CM, PM, inspection, random shock and sudden failure, during its lifetime. The combination and timing of these mutually exclusive events decide the degradation path in addition to the system's natural degradation (wear-out) dynamics when it is not tampered with in any way. Sequential modeling of such events for a degradation dataset will provide maintenance estimation and prediction parameters. Originality/value: In general, CBM models do not consider CM to be part of the maintenance policy. It is debatable whether to include CM as part of a CBM policy. In our case, we consider CM in CBM policy formulation, assuming: (1) fail-repair: the system is restored to its operating state after failure with repair; (2) salvage: disposal of non-repairable part or material from the system, i.e. CM actions are a result of minor component-level failures. System failure is referred to as a case involving rebuild/complete system replacement. We assume CM for minor failures and acknowledge that failure can be delayed using CBM but cannot be avoided entirely. Therefore, the consideration of CM in the CBM model will make the model more generic and comprehensive.

Purpose: The idea is to develop a condition-based maintenance (CBM) model for continuously monitored degrading systems under imperfect maintenance. The system is subjected to random shocks, inspection, corrective maintenance (CM) and preventive maintenance (PM); respective thresholds are set to make decisions. Design/methodology/approach: The generalized renewal process is considered for degradation modeling, and the maximum likelihood estimation method is used for parameter estimation. Optimum replacement time is derived by minimizing downtime per unit time (DT). Findings: Optimum replacement time is derived by minimizing DT. The paper provides insight into the virtual age model’s applicability and efficacy in CBM policy-based decision-making. Practical implications: This paper assesses the CBM policy for continuously degrading systems under random shocks and imperfect maintenance. An operating system faces five events, i.e. CM, PM, inspection, random shock and sudden failure, during its lifetime. The combination and timing of these mutually exclusive events decide the degradation path in addition to the system's natural degradation (wear-out) dynamics when it is not tampered with in any way. Sequential modeling of such events for a degradation dataset will provide maintenance estimation and prediction parameters. Originality/value: In general, CBM models do not consider CM to be part of the maintenance policy. It is debatable whether to include CM as part of a CBM policy. In our case, we consider CM in CBM policy formulation, assuming: (1) fail-repair: the system is restored to its operating state after failure with repair; (2) salvage: disposal of non-repairable part or material from the system, i.e. CM actions are a result of minor component-level failures. System failure is referred to as a case involving rebuild/complete system replacement. We assume CM for minor failures and acknowledge that failure can be delayed using CBM but cannot be avoided entirely. Therefore, the consideration of CM in the CBM model will make the model more generic and comprehensive. © 2024, Emerald Publishing Limited.