The application of numerical probabilistic analysis for the reliable estimate of the characteristics equipment responsible appointment

The paper considers the problem to build reliable estimate of reliability indices for the special equipments in the condition of small statistic failure data. We propose a new approach to building reliable estimates. Its basis is a numerical probabilistic analysis, which allows to calculate reliable estimates in terms of different uncertainty types and the limited amount of information. The numerical probabilistic analysis organizes the reliable computing in cases when the cumulative distribution function is not known. To solve this problem, we developed the method to construction of the reliable estimates empirical cumulative distribution function and reliable estimate probability density function.