In this thesis, we try to predict the reliability of a complex system from the initial stage, namely, predicting the reliability of the components of the system to the final stage, predicting the reliability of the system itself step by step and by choosing the appropriate methods. The predictability of the reliability of the system components is accomplished by using two methods and the Poohson heterogeneous process, which, by comparing the results of both methods, chooses the most appropriate method. In the prediction phase of the system reliability, we use the predictive reliability method of the reliability of the matrix-based system, the reason for choosing this method to predict the reliability of the system, its low computation volume, and its innovative method for prediction. Trustworthiness is reliable. The results of this thesis are important due to the application of complex systems in the industry, especially the electronics and automotive industries, as well as application in the problem of allocating surplus components.