Yakovyna V. Methods and tools of software reliability analysis considering lifecycle stages

The thesis addressed the actual scientific and applied problem of the development of models, methods and tools for software reliability analysis. The developed solutions take into account software complexity, architecture and life cycle stages, making it possible to improve the accuracy of the estimation of software reliability measures. The software reliability models in the form of nonhomogeneous Poisson process, as well as in the form of high-order Markov chains have been developed to analyze the reliability of complex software systems. The methods of software failures forecasting using regression analysis and neural networks have been constructed.