Competing risk problems involving degradation failures are becoming increasingly common and important in practice. In this paper, we investigate the modeling of competing risk problems involving both catastrophic and degradation failures of oil circuit breaker under normal functioning conditions. Using a real case study, shocks highlighted by short-circuits frequency and magnitudes are modeled using non homogeneous Poisson process, however the random variables representing degradations (oil aging and contacts wear out) are modeled following Weibull, uniform and exponential distributions respectively. The aim is the evaluation of system reliability and state probabilities upon the mission time. In this paper it has been proved that one part of system reliability improvement may be supported throughout the relevance of the downstream performances.