Scientific approaches to the development of digital technologies for assessing economic efficiency and managing risks in energy production are presented using small nuclear plants as an example at the main stages of their life cycle. Efficiency indicators of the primary process equipment are determined based on the index method. It is shown that the calculation of the index enables a mathematically sound solution to the problem of efficiency assessment and risk management at the main stages of the plant life cycle. To evaluate the effectiveness of measures taken to ensure the safety of energy production, methods of predictive analysis and management in this area of modern production are proposed. The production system for operating complex technical systems is considered as an aggregate of three components: primary process equipment, resources, and human capital. A detailed set of factors influencing the reliability of personnel's production activities is presented. These factors include experience, safety culture, competence, and psychophysiological adaptation. Standard approaches to risk calculation are adopted, ensuring high reliability of the obtained results. The application of fuzzy set theory, as presented in the article, allows a mathematically correct solution to the optimization of the utility function and the determination of the possibilities and probabilities of risk realization. The damage resulting from risk realization is represented according to the composition and parameters accepted in the energy sector. The comprehensive risk analysis of energy production presented in the article enables fuller application of generally accepted methods of functional and cost analysis of complex technical systems at the main stages of their life cycle. The proposed technologies form the scientific and methodological foundation for neural network programming, machine learning, and predictive analysis in the creation and operation of small nuclear power plants using elements of artificial intelligence.