Analysis methodology of failure data analysis for technical objects by the maximum plausibility method with the use of Mathcad program

The methodology for assessing the reliability indicators of technical objects based on a priori information about operat-ing time to failure, presented in the form of complete or censored data, is considered. The methodology is founded up-on a developed algorithm of sequential analysis, utilising maximum likelihood estimates of the studied samples distrib-uted according to the Weibull distribution. This approach is advantageous over many of the existing statistical methods with a large amount of data, in that it allows for the determination of parameter estimates taking into account various types of censoring (suspensions). The application of the maximum likelihood method is contingent upon a preliminary evaluation of the criterion of agreement between the presented data set and the two-parameter Weibull distribution law. This is due to the latter's universality, which facilitates its application in the description of a wide range of resource data. Utilizing estimates of the shape parameter and scale parameter of the distribution, the proposed method facilitates the determination of point and interval estimates of the probability of failure-free operation, failure intensity, B-resource, and mean time between failures. A study was conducted on the errors of the obtained estimates of reliability indicators for data in the form of a time-censored sample and a failure-censored sample. The study also examined the depend-ence of the error on the size of the array. The description is accompanied by a numerical example of obtaining point and interval estimates of reliability indicators for a technical object using the Mathcad computer algebra system.