Contractor Selection for Construction Project: A Genetic Fuzzy Model Approach

Abstract

The selection of a contractor is a pivotal aspect in civil engineering projects, determining their success and efficiency. This article introduces a comprehensive multi-criteria model aimed at selecting the most suitable contractor by considering both qualitative and quantitative factors essential in contractor evaluation. Addressing the challenge of assessing credit risk among contractors in civil contractor firms, an expert model integrating fuzzy logic principles and a genetic optimization algorithm is proposed. Utilizing MATLAB software, the model is simulated, and the credit risk estimation accuracy is evaluated through expert opinions from construction employer companies. Metrics such as MRE, MMRE, VAF, VARE, and MARE are computed and compared with existing models including Rao_CCPQ, Li_OCICS, and hierarchical analysis, employing the Friedman test via SPSS software. Analysis reveals significant enhancements in these metrics within the proposed genetic fuzzy system compared to other models, indicating improved accuracy in credit risk assessment of construction contractors. Consequently, this improvement enhances validation accuracy of contractors, mitigates the risk of contractual default in construction employer companies, and prevents potential financial and temporal burdens in civil projects.