The construction industry is crucial to economic growth as it significantly impacts infrastructure, urbanization, and employment. However, selecting suitable contractors remains challenging, often causing delays, budget overruns, and compromised quality. Traditional methods relying on subjectivity and familiarity are insufficient for modern projects. This study enhances contractor selection by incorporating advanced decision strategies and forecasting methods, using fuzzy TOPSIS and gradient boosting machine (GBM). The framework aims to robustly recognize skilled, dependable contractors, ultimately improving project delivery. The multi-criteria decision-making approach employs fuzzy TOPSIS to assess and rank contractors based on criteria like price, deadlines, finances, safety, quality, knowledge, and skills. GBM also predicts success probabilities, augmenting decisions with predictive analytics. The combination of fuzzy TOPSIS and GBM considerably strengthens selection by enabling thorough evaluations of subjective and numerical factors. The framework performs better than traditional techniques through more precise, reliable contractor choices. In conclusion, combining fuzzy TOPSIS and GBM proves effective for construction contractor selection. This method enhances selection dependability and efficiency by addressing multi-criteria complexity and incorporating forecasting. Furthermore, it establishes a basis for future sophisticated analytics use across diverse project management phases.