| dc.description.abstract | The lack of effective and comprehensive risk identification leads to ineffective risk management, which accordingly leads to failure in achieving project goals. Previous research efforts related to simulating risks in construction projects mostly have been limited to viewing risks as segregated factors, and thus providing static analysis. This paper addressed this knowledge gap through a risk path simulation–driven approach that addresses construction risks as a continuum of risk sources, events, and project characteristics. The authors (1) identified 131 construction project risks by conducting an extensive literature review, (2) designed a risk path by analyzing risk paths available in the literature, capitalizing on identified benefits, and avoiding drawbacks, (3) conducted a project-based survey to collect data from 53 construction projects related to the magnitude and impact of occurrence of the identified risks on cost overruns, (4) developed a series of two genetic algorithm (GA) models and one artificial neural network (ANN) model to quantify the impact of construction risks on cost overruns, and (5) created an ontology model to facilitate data structuring and retrieval. An ontology is a data modeling tool used to represent unstructured information into structured forms. Results show that 80% of the risks identified in this research were found to have a significant impact on cost overruns, and 89% can be recognized at project initiation. The contribution of this framework lies in its ability to represent the cost impact of a chain of conditions and events, rather than a single-point risk factor using GA and ANN modeling techniques. The proposed framework may help industry practitioners assess the impact of significant construction risks on cost overruns. | |
