ECONOMETRIC MODEL OF THE IMPACT OF MANAGERIAL DECISIONS ON THE LEVEL OF ENERGY CONSUMPTION

Abstract

The study addresses the critical issue of energy consumption in medical enterprises by developing an econometric model that quantifies the influence of managerial decisions and technical factors on overall energy use. Energy consumption in healthcare institutions is shaped by a complex interaction of organizational, technological, and resource-related determinants, including electricity, heat, gas, and water consumption; the efficiency of medical equipment; the quality of thermal insulation of buildings and networks; the level of personnel training in energy efficiency; the age of facilities; and the use of renewable energy sources. Based on expert surveys, ten core factors were identified as the most significant drivers of energy consumption. These factors were normalized and incorporated into a multiple regression model constructed with real monitoring data over a 100-month observation period. The model integrates quantitative and qualitative indicators, ensuring comparability across heterogeneous variables through min–max normalization. Regression coefficients were estimated using the least squares method, and model adequacy was validated by calculating determination coefficients and analyzing residual distributions. The findings reveal that heat consumption, water use, and the efficiency of medical and climatic equipment exert the strongest impact on energy demand. The constructed model demonstrates high reliability in forecasting energy consumption, with deviations mainly occurring under peak-load conditions, which are challenging to capture using linear approaches. Nevertheless, the model provides practical value for decision-making in energy management by allowing managers to estimate the effects of optimizing specific factors or assess potential risks of energy inefficiency. The research highlights the importance of integrating statistical modeling into healthcare energy management to improve resource allocation and reduce operational costs. Furthermore, the study outlines perspectives for applying advanced modeling approaches such as neural networks and neuro-fuzzy methods, which may enhance forecasting accuracy but require greater computational resources. The proposed model thus serves as a practical and resource-efficient tool for guiding managerial strategies aimed at energy saving and sustainable operation of medical enterprises.