MANAGEMENT DECISION-MAKING MECHANISM IN SHORT-TERM EDUCATIONAL PROJECTS BASED ON LEARNING ANALYTICS

Abstract

This paper conceptualizes a Learning Analytics-based managerial framework for short-term, military educational initiatives. The study is prompted by digital learning ecosystems generating vast performance data without systematic methodologies to convert metrics into empirical management strategies. Within high-stakes environments where errors yield severe consequences, operational choices predominantly rely on instructor intuition. This subjectivity undermines the consistency, transparency, and replicability of project oversight, ultimately impeding the scalability of intensive training models. The study validates a structured decision-making framework for instructional oversight using Learning Analytics. Combining project management with adaptive systems and risk governance, it enables instructors to execute targeted pedagogical interventions. The architecture was verified using performance logs from 120 trainees in high-fidelity military simulations. Standardized corrective workflows evaluate six key composite indicators: Skill Index, Completed Missions, Error Rates, MARCH Protocol Compliance, Knowledge Retention, and Practice Scores. The empirical findings demonstrate that managerial decision-making within intensive training can be formalized into a multi-stage algorithm covering anomaly detection, risk assessment, classification, intervention selection, execution, and monitoring. Field data enabled a taxonomy mapping instructional scenarios to calibrated administrative interventions, ranging from training repetition to immediate module advancement. Ultimately, this framework transitions Learning Analytics from passive monitoring into an active, prescriptive decision-support engine for short-term educational project oversight. The scientific novelty centers on a standardized decision-making framework pairing Learning Analytics with educational risk thresholds. Unlike conventional subjective assessments, this algorithmic blueprint triggers administrative interventions based purely on objective performance data. Practically, the mechanism integrates into digital learning ecosystems to optimize high-stakes military training. Furthermore, these findings establish a scalable methodological foundation for automated decision-support systems and predictive dashboards across broader higher education architectures.