Predicting residual active life of a spacecraft

The paper presents mathematical statistics techniques that make it possible to monitor the quality of spacecraft during their production and provide early indications of potential equipment quality control troubles. It presents an algorithm for determining the average number of failures in total for all the specified operating hours, and for operating hours for which a prediction needs to be specified. It was found that it is necessary to predict the average number of failures for space hardware products as a whole, when a low number of failures occurs in individual components and predictions of specified operating time for these components have significant errors. It describes general principles, which make it possible to check the manufacturing process for accuracy and stability, as well as to determine distribution parameters for non-standard deviations in case of a stable manufacturing process.