An analysis of the use of numerical–analytical methods for modeling diagnostics and prediction of failures of digital twins for specific applications in maintenance and repair in various industries is presented. The relevance of this work is determined by the need to create highly detailed physical models of units and assemblies of mechanical engineering products, allowing the study of wear and failure processes under conditions of dynamically changing external factors. The methods used, the results obtained, the agreed conclusions, and open problems are considered. An assessment of the reliability and demonstrativeness of the studies reviewed is given. The results obtained are of interest to technical researchers and specialists involved in the management of complex technical systems.
Erofeev et al. (Sun,) studied this question.