•  
  •  
 

Keywords

Artificial intelligence, Contractor pre-qualification, Machine learning, Multi criteria decision making, Contractor selection, Construction procurement, Risk assessment

Document Type

Article

Abstract

This study develops and validates an AI-driven contractor pre-qualification optimization framework to improve the accuracy, consistency, and reliability of contractor selection for institutional construction projects. The study addresses limitations in conventional approaches that place strong emphasis on bid price and historical completion performance while providing limited assessment of future project risks. A case study design was adopted using contractor pre-qualification data for an institutional building project. Four contractors were evaluated using cost competitiveness, success rate, completion time reliability, quality performance, safety performance, and financial stability. A weighted multi-criteria decision-making model was combined with supervised machine learning to predict the probability of on-time completion, risk of cost overrun, and expected quality performance. The resulting contractor ranking was compared with what was obtained using the existing pre-qualification approach. The existing approach ranked the contractors as Firm A, Firm D, Firm C, and Firm B, while the AI-driven framework ranked them as Firm A, Firm C, Firm D, and Firm B. Firm A remained the preferred contractor, achieving a pre-qualification score of 93.81, a 95% predicted probability of on-time completion, an 8% cost overrun risk, and an expected quality score of 8.8/10. The results indicate stronger differentiation of contractor capability and project risk under the proposed framework.

DOI

10.30684/2412-0758.2434

First Page

267

Last Page

289

Share

COinS