A Bayesian approach to the optimization reliability costs

The lack of reliability testing in a project finds its reason in financial considerations and deadlines. In this context, Bayesian techniques find their applications as they contribute to a significant reduction in the amount of reliability testing based on the classical approach, with the knowledge of the reliability data on a priori relevant components. This paper is devoted to a Bayesian approach to the optimization reliability costs of tests conducted on the electronic components installed in a lift. The Bayesian estimation provides a failure rate of 1,795·10^-6 failure/hr to 60% confidence instead of 2,771·10^-6 failure / hr after testing. A gain of 64, 77% in terms of time and therefore the cost of testing will be reduced considerably. When the number of failures increases (K₀ = 2; 3 and 5), the real tests are not added to virtual testing and a decrease in time to be won has been recorded.