Efficacy of Using Support Vector Machine in a Contractor Prequalification Decision Model

Abstract

Contractor prequalification is basically a nonlinear two-group classification problem. A robust contractor prequalification decision model should include the ability of handling both quantitative and qualitative data. Support vector machine (SVM) is a set of related supervised learning methods which can handle data in a high dimensional feature space for nonlinear separable problems. A new contractor prequalification decision model using SVM is proposed to assist clients to identify qualified contractors for tendering in this study. A case study was used to validate the proposed decision model and the classification ability was compared with neural networks (NNs) and principal component analysis (PCA). The results show that the proposed SVM model outperforms NN and PCA and the merits of using SVM to mitigate the limitations of using NN are elaborated. The proposed decision model is an ideal alternative for supporting clients to perform contractor prequalification decision making.