Replacement Policy for Heterogeneous Items Subject to Gamma Degradation Processes

• Published in: Methodology and computing in applied probability Vol. 24; no. 3; pp. 1323 - 1340
• Main Authors: Cha, Ji Hwan, Finkelstein, Maxim, Levitin, Gregory
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.09.2022; Springer Nature B.V
• Subjects: Business and Management; Cost analysis; Degradation; Economics; Electrical Engineering; Failure; Frailty; Human error; Life Sciences; Mathematics and Statistics; Parameter sensitivity; Population; Preventive maintenance; Random variables; Sensitivity analysis; Statistics; Stochastic models; 62P30; Probabilistic analysis; Hybrid preventive maintenance policy; Heterogeneous degradation process; 60 K10; Dependent increments; Mixed gamma process
• ISSN: 1387-5841; 1573-7713
• DOI: 10.1007/s11009-021-09859-5

A hybrid preventive maintenance policy for heterogeneous degrading items is discussed. It combines the classical age-replacement strategy, when a system is replaced either on failure or at the predetermined age, with replacement of the system when degradation reaches the predetermined level at some intermediate time. Items come from two subpopulations with different reliability characteristics. Non-homogeneous gamma processes model degradation of an item from each subpopulation. We justify probabilistically the superiority of the proposed policy over that for homogeneous populations and over the policy without possibility of additional replacement. The corresponding long-run cost rate is derived for the suggested cost structure. Some detailed numerical illustrations are presented and relevant sensitivity analysis for the main parameters of the model is performed.