Extension of the GERT network model for describing, evaluating and optimizing information processes in distributed systems

The article is devoted to the development of an extended GERT network model adapted for the analysis and optimization of information processes in distributed systems. The scientific novelty of the study lies in the formalization of nodal logic that takes into account timestamps, probabilistic transitions and resource distribution, as well as in the introduction of a mechanism for hierarchical nesting of subnets while preserving stochastic characteristics. The practical significance is confirmed by the results of the model implementation, which showed a reduction in transaction processing time, increased fault tolerance and reduced infrastructure costs in the banking sector. The methodology is based on the integration of stochastic programming, game theory and machine learning, which allows overcoming the limitations of classical GERT networks. Particular attention is paid to modeling the spatio-temporal characteristics of distributed architectures through the concept of a "virtual time continuum", which ensures the universality of the approach. The research prospects are related to the development of cross-platform tools for various applications.