Reliability-Test Planning Considering Multiple Failure Mechanisms and System Levels – an Approach for Identifying the Optimal System-Test Level, Type, and Configuration with Regard to Individual Cost and Time Constraints

Abstract

To demonstrate the reliability of a product, engineers are forced to find the best feasible compromise between cost, time, and accuracy. In addition, the test configuration must be set according to the lowest possible resource requirements. The approach proposed in this paper utilizes the concept of probability of test success presented by Dazer et al. [1] to assess different test plans. In addition to the probability of test success, the test time and costs are used in the assessment. In order to obtain the distribution of the probability of test success, costs, and time, the different test types are simulated on the respective system levels using a Monte Carlo simulation. Two-parameter Weibull distributions of the failure mechanisms of the system serve as a prerequisite for the simulation. For real-world application, these can stem from similar products and applications, previous product generations, preliminary tests, expert knowledge, or even simulations. Based on these distributions, pseudo-random numbers of failure times are generated.