A Two-stage Reliability Allocation Method under Uncertainty

Reliability allocation is a dynamic process and is generally implemented in the early development stage. Most existing reliability allocation methods have neglected the two important characteristics for the reason that they simply evaluate the reliability allocation factors in a precise way and cannot ensure a valid reliability improvement after allocation. To overcome these limitations, we propose a two-stage reliability allocation method based on fuzzy set with grey number elements. In the first stage, reliability allocation factors are assessed by linguistic variables which are expressed by fuzzy set with grey number elements. The operations of both fuzzy set and grey number are employed in this stage to obtain the allocation weight. During the second stage, a reliability improvement rate is developed based on the system reliability improvement target, current reliability level and allocation weight. An illustrative example with regard to a newly-developed civil aviation product is presented to demonstrate that the proposed approach is more rational and effective on dynamically assigning the system reliability requirements to its subsystems in the context of uncertainty.