Contractor Performance Prediction Model for the United Kingdom Construction Contractor: Study of Logistic Regression Approach

An accurate prediction of contractor potential is of vital importance during contractor selection and evaluation process. Such prediction enables identification and classification of contractor performance to ease the selection process. This paper outlines the use of clients' tender evaluation preferences for predicting a contractor performance via a logistic regression (LR) approach. A total of 31 clients’ tender evaluation criteria were selected to develop a LR model for predicting contractor performance. The proposed model was developed based on 48 of United Kingdom public and private construction projects and validated in 20 independent cases. It was found that 75% of the cases correctly and the model statistically accurate for contractor performance prediction, where the input variables consist of nominal and interval data. The paper summarized techniques and advantages of LR analysis and discussed literature findings of contractor selection and evaluation methodologies undertaken by construction researchers and commentators from the United Kingdom and Northern America.