Hide/Show Apps

A knowledge-based risk mapping tool for cost estimation of international construction projects

2014-07-01
Yildiz, Acelya Ecem
Dikmen Toker, İrem
Birgönül, Mustafa Talat
Ercoskun, Kerem
Alten, Selcuk
Effective risk assessment and management is critical for success in international construction projects. This paper proposes a knowledge-based risk mapping tool for systematically assessing risk-related variables that may lead to cost overrun in international markets. The tool uses an ontology that relates risk and vulnerability to cost overrun [1] and a novel risk-vulnerability assessment methodology [2] to estimate potential risk paths that may emerge in international construction projects. The tool has been developed in collaboration with an industrial partner, a construction management company that gives risk management consultancy services to Turkish contractors working in international markets. The tool has been designed by using the previous projects of the partner firm as test cases and preferences of company professionals are taken into account while determining the functions of the tool. As the reliability and usability of the tool significantly depend on the subjective evaluations of users about level of vulnerability and magnitude of potential risk events, a lessons learned database has been incorporated into the tool so that decision-makers may refer to risk event histories of previous projects to make estimations about forthcoming projects. In order to evaluate the usability of the tool, a usability test has been conducted by eight construction experts. Usability test results demonstrated that the tool can be utilized for prediction of probable risk paths and their impact on cost. Results of a case study, a building project in Serbia, have also been reported in this paper to demonstrate the functions and the performance of the tool. Although the usability and reliability test results are satisfactory, the tool should be seen as an initial platform which should be improved by increasing its "intelligence" by high quality and enough number of risk event histories, and also customized according to user preferences as well as company policies.