Effect of Interval Censoring on the Probability of Test Success in Reliability Demonstration

SUMMARY & CONCLUSIONSWithin the scope of this work, investigations into the influence of interval censoring on the probability of success for reliability-demonstration tests were conducted. With the help of a simulation model based on the generation and evaluation of random numbers, a parameter study was carried out to determine the influence of the parameters on the probability of test success, test cost, and time. The probability of test success is a newly developed metric for planning life tests with Weibull-distributed life data. In addition to the length and frequency of inspection intervals, other parameters such as the reliability requirement, sample size, and the failure behavior of the product were also investigated. The results show that a high probability of test success can already be achieved with only a few (long) intervals under certain boundary conditions. However, product failure behavior and oversizing are also decisive influencing factors that must be considered when planning life tests. This underlines the fact that there is generally no best choice of inspection intervals, rather that optimal test planning can only be achieved when considering product-specific failure behavior.