A contractor selection decision support system

Abstract

Proper contractor selection is a critical aspect for successful project management practice. In the light of world economic crises, the need for proper practices in this respect has become paramount. The construction industry today constitutes a {dollar}7.5 trillion market. Between the years 2005 and 2012, the construction industry was one of the most prominent sectors, with an average annual growth of 11%. This rapid growth of the industry means that more projects will be on the market and hence more bid opportunities will come into existence. Contractor selection is a difficult task due to the risks it entails. Therefore there is a need for a decision support system that can aid decision makers with this challenging task. This area has been extensively examined by research. Numerous models have been developed. However, these models fall short of industrial applicability because they remain primitive, data intensive, or risk oblivious. The aim of this research is to develop a decision support system that takes into account the uncertainty pertaining to the nature of the problem. Two Phase Contractor Selection Decision Support System is developed to take into account key qualifying criteria according to literature and expert survey. The first stage utilizes Multi Attribute Utility Theory (MAUT). In the second stage a combination of Analytic Hierarchy Process (AHP) and Preference Ranking Organization Method for Enrichment Evaluation PROMETHEE II is used. The PROMETHEE II method features modifications that are uniquely applied in this model.A second step is to perform stochastic analysis to evaluate a Stochastic Success Index (SSI) based on Monte Carlo Simulation of the probability distributions of sub criteria weights. The model requires data pertaining to weights for criteria and thresholds for the PROMETHEE method. This information is collected from a survey of industry professionals. Statistical data analysis was performed on the collected data to study its significance. The model was applied on a case study and was found to be fully functional and capable of giving significant results. Overall, the model is stable and can be widely applied in different applications. However, there is room for further research work in terms of model development and validation.