The effectiveness of contractor selection and tender bid evaluation significantly influences the success of large multinational projects, where procurement decisions directly impact cost efficiency, schedule performance, and overall project delivery outcomes. Traditional procurement evaluation methods typically rely on additive weighted scoring models and expert judgment, which often fail to capture complex relationships among operational variables and lack the ability to incorporate real-time enterprise data. This study proposes a machine learning–driven framework for tender bid evaluation and contractor selection using operational data extracted from enterprise resource planning (ERP) systems. The framework integrates financial, procurement, workforce, and operational datasets to construct predictive models capable of evaluating contractor performance and estimating delivery risk. A quantitative predictive modeling approach is implemented using multiple machine learning algorithms, including Random Forest, Support Vector Machine, Gradient Boosting, and Deep Neural Networks. Mathematical formulations are developed to transform ERP-derived variables into predictive contractor evaluation scores, enabling a shift from static scoring methods to dynamic, data-driven decision-making. The proposed framework is evaluated through a comparative analysis against traditional procurement scoring models, demonstrating improved predictive accuracy and enhanced capability in identifying high-risk contractor selections. The findings highlight the importance of integrating enterprise data analytics with artificial intelligence techniques to improve transparency, reduce subjectivity, and enhance efficiency in procurement decision processes. The study contributes to the advancement of intelligent procurement systems by providing a scalable and analytically rigorous approach to contractor selection in complex multinational project environments.
Onuh Matthew Ijiga2 Nnenna Linda Akunna*1 (Mon,) studied this question.