THE DEVELOPMENT OF A PREDICTIVE AND ANALYTICAL MANAGEMENT MODEL FOR THE DEVELOPMENT OF EDUCATIONAL PROGRAMS OF AN ENGINEERING AND ECONOMIC PROFILE

Abstract

The modern labor market imposes high demands on graduates of engineering and economics specialties, requiring not only fundamental knowledge but also advanced digital competencies, project management skills, and the ability for interdisciplinary interaction. The article presents the results of the development and testing of an integrated predictive-analytical system for assessing the competency profile of students in engineering and economics. The goal of the work was to create a tool capable of not only diagnosing current gaps between the training of students and the requirements of the digital economy but also forming forecast-based recommendations for proactive modernization of educational programs. The methodological basis of the study was the synthesis of six key approaches: three competency models -T-Shaped Skills (assessing the depth of specialization and breadth of interdisciplinary competencies), the 4K concept (diagnosing soft skills - critical thinking, creativity, communication, cooperation), and Bloom’s taxonomy (gradation of cognitive levels); as well as three analytical methods - the Delphi method for structured expert assessment, Gap Analysis for quantitative identification of competency gaps, and predictive analytics algorithms including time series analysis (ARIMA, Prophet) and gradient boosting (XGBoost). The empirical base was formed through data triangulation, including a survey of 1459 students from three leading Russian universities, as well as independent expert evaluations from the academic community (20 professors) and the professional market (20 employers). A unique component of the methodology is the integration of traditional pedagogical assessment models with predictive analytics methods: time series analysis (ARIMA and Prophet models) and machine learning algorithms (XGBoost) for constructing scenariobased forecasts of competency demand up to 2035. The testing results revealed a persistent structural imbalance. While a high level of soft skills development was confirmed, particularly in cooperation (4.35 out of 5) and critical thinking (3.78), a critical deficit was diagnosed in key hard skills essential for Industry 4.0/5.0: programming (2.94) and project management (3.08). Comparative analysis of the three groups of assessments (self-assessment, professor assessment, and employer assessment) showed that it is business representatives who assign the strictest ratings for digital competencies, indicating a significant gap between educational outcomes and practical requirements. The first scientific outcome of the study is the developed algorithm constituting a closed-loop competency management system whose key distinguishing feature is the integration of time series analysis and machine learning methods. The second scientific outcome is the designed cycle of predictive and analytical management for the development of engineering-economics educational programs, which visually illustrates the transition of educational program renewal toward data-driven governance principles, thereby ensuring the training of graduates who are competitive in the dynamic digital economy. Future research will focus on developing predictive models of demand for key competencies up to the year 2035 to enable proactive educational management, as well as on formulating a targeted set of recommendations for the education system based on the synthesis of diagnostic and predictive data.