Prior distribution elicitation is a challenging problem for a Bayesian inference-based mean time to repair (MTTR) demonstration because if inaccurate prior information is introduced into the prior distribution, the results become unreliable. This paper proposes a novel maintenance task representation model based on the similarity of attributed maintenance items. A novel similarity computation algorithm for maintenance tasks is then formulated on the basis of this model. Optimistic and pessimistic values are ascertained from the time data for similar maintenance tasks to obtain a prior distribution. The main idea is to separate maintenance tasks into distinct items and use attribute sets to extract key features. Each pair of items is then compared to quantify the differences between reference and candidate tasks. Candidate tasks with an acceptable difference from the reference task are taken as prior information sources for constructing the prior distribution. A case study involving a high-frequency (HF) transceiver MTTR Bayesian demonstration shows that the proposed method can effectively obtain maintenance tasks similar to those of information sources for prior distribution elicitation.
Maintainability verification is an important work in system development. However, two problems arise when the maintenance time of complex system conforms to a mixture distribution. The first problem is the occurrence of assuming log-normal distribution instead of mixture distribution and the second one is the limited sample size for maintainability verification. Regarding the problems mentioned above, firstly, the maintenance sample generation process of complex system is analyzed and the mixture distribution model of maintenance time is proposed. Then, the changes in two types of test risks are analyzed when assuming log-normal distribution instead of mixture distribution. Finally, a method to perform maintainability verification using similar maintenance task's samples is proposed. A case study using the maintenance time data of satellite communication interface system shows the efficiency of the proposed method.
During maintainability demonstration, the maintenance time for complex systems consisting of mixed technologies generally conforms to a mixture distribution. However existing maintainability standards and guidance do not explain explicitly how to deal with this situation. This paper develops a comprehensive maintainability demonstration method for complex systems with a mixed maintenance time distribution. First of all, a K-means algorithm and an expectation-maximization (EM) algorithm are used to partition the maintenance time data for all possible clusters. The Bayesian information criterion (BIC) is then used to choose the optimal model. After this, the clustering results for equipment are obtained according to their degree of membership. The degree of similarity for the maintainability of different kinds of equipment is then determined using the projection method. By using a Bootstrap method, the prior distribution is obtained from the maintenance time data for the most similar equipment. Then, a test method based on Bayesian theory is outlined for the maintainability demonstration. Finally, the viability of the proposed approach is illustrated by means of an example.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.