PREDICTIVE MANAGEMENT OF THE ECONOMIC EFFICIENCY OF CONSTRUCTION PROJECTS BASED ON AI AND BIG DATA

Abstract

The article examines the theoretical and applied aspects of predictive management of the economic efficiency of construction projects in the context of digital transformation. The relevance of shifting from traditional reactive management, focused on identifying and correcting deviations after their occurrence, to predictive management aimed at anticipating potential changes and developing preventive managerial decisions is substantiated. The essence of predictive management of construction project economic efficiency is clarified, and its differences from traditional management approaches are identified. It is established that economic efficiency is determined by the interrelated influence of cost, time, resource, and financial parameters, which requires their integrated analysis and forecasting. The role of Big Data and artificial intelligence in creating an information and analytical basis for predictive management is determined. Big Data provides the collection, consolidation, structuring, and processing of heterogeneous project data, while AI and machine learning enable the identification of patterns, forecasting of key project parameters, detection of potential deviations, and assessment of emerging risks. The main components of predictive analysis are systematized according to the objects of analysis, data sources, forecast parameters, potential risks, and corresponding managerial decisions. A model of predictive management of the economic efficiency of construction projects based on the integrated use of AI and Big Data is proposed. The model combines five interconnected components: information, predictive and analytical, risk and scenario assessment, managerial, and control and adaptive contours. Their interaction ensures the transformation of project data into forecasts and risk assessments, and analytical results into preventive managerial decisions concerning costs, schedules, resources, procurement, and financing. Feedback mechanisms enable actual results to be incorporated into the information environment, thereby updating the data base, improving forecasts, and adapting managerial decisions to changing project conditions. The proposed approach creates prerequisites for the early identification of cost overruns and schedule delays, optimization of resource allocation and financial flows, timely response to potential deviations, and improvement of the economic efficiency of construction projects.