<![CDATA[Enterprise Resource Planning (ERP) implementation is a strategic initiative to improve business process integration and organizational performance; however, many projects that are operationally successful fail to deliver significant business value. This study proposes a Benefit Realization Prediction in ERP Project Management approach using a machine learning model based on Extreme Gradient Boosting (XGBoost) with a quantitative explanatory–predictive design and a dataset of 300 ERP projects. The model integrates multi-dimensional variables, including project characteristics, project management, and organizational readiness, to predict benefit realization using a multi-class classification approach. The dataset is split into 80% training data and 20% testing data, where the model is trained to capture complex relationships among variables and evaluated for generalization performance. Hyperparameter tuning is applied to optimize key parameters such as n_estimators, learning_rate, and max_depth. The results show that the XGBoost model achieves strong performance with an accuracy of 91% and an AUC of 0.95. Feature importance analysis identifies top management support, organizational readiness, project manager experience, user training intensity, and system integration complexity as the most influential factors. Additionally, experimental scenarios such as ablation study and early prediction demonstrate the model’s ability to identify key factors and provide early-stage predictions, highlighting its potential as a decision-support tool for proactive and value-based ERP project management]]>
Copyrights © 2026
