DEVELOPMENT OF A MACHINE LEARNING MODEL FOR PREDICTING ENROLMENT TRENDS AND RESOURCE ALLOCATION OPTIMISATION FOR THE UNIVERSITY OF IBADAN, NIGERIA

Effective enrolment forecasting and resource allocation are critical challenges for higher education 
institutions in developing countries, where financial constraints, infrastructural limitations, and 
volatile admission patterns significantly complicate strategic planning. This research investigated 
the application of machine learning techniques to improve student enrolment prediction accuracy 
and optimise institutional resource allocation at the University of Ibadan. 
The research adopted a quantitative, data-driven approach guided by the Cross-Industry 
Standard Process for Data Mining (CRISP-DM) framework. The dataset comprised of 
comprehensive institutional records from 2014 to 2024, including undergraduate admissions, 
enrolment figures, faculty staffing, departmental budgets, hostel allocations, and macroeconomic 
indicators. A hybrid modelling strategy benchmarked multiple supervised learning algorithms, 
including Random Forest, Support Vector Machines, Extreme Gradient Boosting (XGBoost), 
Logistic Regression, and Long Short-Term Memory (LSTM) networks, across classification and 
regression tasks. Rigorous feature engineering, multicollinearity assessment, and 5-fold cross
validation ensured model robustness and generalisability. 
Results showed that ensemble methods, particularly Random Forest, delivered superior 
performance, achieving classification accuracy of approximately 95% and regression R² values 
close to 0.93. Feature importance and SHAP (SHapley Additive exPlanations) analysis revealed 
that internal institutional factors, specifically faculty staffing levels, departmental budget 
allocations, student–staff ratios, and admission selectivity, exert substantially stronger predictive 
influence than external macroeconomic variables. The models also identified structural 
inefficiencies in resource distribution and quantified enrolment sensitivity to capacity constraints 
and disruptions. 
Building on these predictive insights, an optimisation framework was developed to 
recommend data-driven resource allocation strategies aligned with forecasted enrolment patterns. 
An interactive dashboard was designed to deliver actionable forecasts and optimisation 
recommendations to university administrators. The research demonstrated that integrating 
predictive analytics with resource allocation modelling produces a reliable decision-support 
system capable of guiding faculty staffing, budget distribution, and infrastructure planning under 
conditions of uncertainty. The findings underscore the strategic value of data-driven governance 
in higher education and provide a scalable analytical blueprint for similarly constrained 
universities in developing contexts seeking to enhance operational efficiency, financial 
sustainability, and long-term academic planning.