Publications — Yuriy Dmitrishin | Articles, Research, Reports

Publications

Peer-reviewed articles, technical reports, and research by Yuriy Dmitrishin.

Recommendations for Implementing Artificial Intelligence Technologies in Nuclear Power Plant Construction Processes

Dmitrishin Yuriy · September 10, 2026

The report presents application areas for artificial intelligence technologies in nuclear power plant construction based on two complementary models developed by the author: a formal NPP life-cycle model for artificial intelligence and machine learning applications (2025), and a probabilistic optimization and forecasting model for NPP construction projects (2018). For each application area, the report describes the task, the underlying problem, the proposed solution, and examples of model-based implementation.

GitHub repository DOI: 10.5281/zenodo.22683709 Zenodo

Formalization of a Nuclear Power Plant Life-Cycle Model for AI and Machine Learning Applications

Dmitrishin Yuriy · February 8, 2026

This paper presents a formal, research-oriented model designed to support a broad range of artificial intelligence and machine learning (AI/ML) tasks related to managing the life cycle of nuclear power plants (NPPs). The model consists of a hierarchical graph structure, a semantic layer, and a classification system, with links to a corpus and associated metrics. While concentrated on NPP life cycles, the proposed principles and methods are transferable to other complex systems.

GitHub repository DOI: 10.5281/zenodo.18527396 Zenodo

Mathematical Optimization and Forecasting of Nuclear Power Plant Construction Progress using Probabilistic Models and Machine Learning Methods

Dmitrishin Yuriy · May 9, 2025

This publication presents a generalized probabilistic project management model that combines PERT, GERT, Monte Carlo methods, mathematical optimization, and modern machine learning and data analysis techniques. The approach aims to build predictive network models considering uncertainty, identify critical and subcritical paths, perform local and global optimization, and ensure model adaptation based on empirical data.

GitHub repository DOI: 10.5281/zenodo.15373676 Zenodo