Development of a Contractor Failure Prediction Model Using Analytic Network Process

Abstract

Construction contractor failure is one of the most critical and costly risks for a employer. Despite the ability of the employer to terminate the construction contract due to the contractor's failure to achieve crucial contractual objectives, the employer still suffers adverse impacts on time, cost, and goodwill. Under price-driven selection, construction contracts are usually awarded to the lowest bidder, with little attention to a bidder's capabilities. Therefore, this study attempted to develop a model to assist construction professionals in selecting the bidder with the lowest failure potential. The analytic network process (ANP) was used to analyze the data collected from a prepared fuzzy questionnaire. The results concluded a ranking for the reasons of contractor failure, which were initially identified from the literature and categorized into five categories. The results showed that “corporate governance” and “financial position” are the first and second most influential categories indicating contractor failure potential, respectively. Furthermore, “cost control,” “tender approach,” and “technical competency” are ranked as the third, fourth, and least influential categories, respectively. Construction practitioners can utilize the model developed by this study to evaluate bidders to minimize the probability of contractor failure and, consequently, to maximize the probability of successful project delivery.