Study on Statistical Model of Reparable System Failure Data

Either for hardware or software repairable system, reliability analysis is very important. However, failure data analysis for repairable system is not an easy task. The widely used AMSAA model (power law process) is a kind of NHPP, corresponding to minimal repair. According to the variety of the reliability of repairable systems, accounting for the influences of the modification and repair, the repair effectiveness index is introduced, and a reliability growth model based on virtual age is developed. The model can be applied to various situations, such as perfect repair, imperfect repair, minimal repair, worse repair and worst repair. The failure rate function and the expression of the cumulate failure numbers are developed, and the MLE of the model parameters are also presented. Finally, the results of the analysis of the example show that the model is practical, and the parameters of the model have more exact engineering meaning.