Thesis (M.Eng.) - Asian Institute of Technology, 2004

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Civil Engineering

The consequence of failure can be catastrophic for project based constrnction organizations in today's competitive world. Therefore, these organizations are constantly seeking ways to avoid failure and contain risks early on in the project life cycle. One such approach to handle risks is through the allocation of 'suitable' contingency or reserve. However, when it comes to determining this suitable contingency fund project personnel often face difficulty. They end up assigning a rather arbitrary contingency fund which does not have any rational basis for the decision that they make. Consequently, either of the two things is bound to happen. In first case, a more than what is required contingency fund is allocated which gets locked up and may make the project economically unath·active. In the second case, a low contingency allocation means that the funding for the project is inadequate for successful completion. In order to provide a logical basis to allocate contingency to projects, a methodology is developed here that incorporates risk assessment. Based on subjective judgment sought from the project personnel regarding the likelihood and severity of risks in their project, the degree of risk of the project is dete1mined. Contingency fund is then allocated based on this quantified degree ofrisk. The developed methodology was applied to two real hydropower projects in Nepal to check the adequacy of the contingency fund allocated. All the constrnction risks were identified for these projects. Their effects were quantified by dete1mining the probability and severity by using Analytical Hierarchy Process (AHP). Ranging was also performed on Bill of Quantity (BOQ) items. Activity/ Risk factor matrix was developed which was then used as an input to the Monte Carlo Simulation to combine the risk distributions of various BOQ items. This resulted in a risk adjusted cost for the project from which the contingency was dete1mined. The application of the methodology showed that the contingencies allocated for the projects under study were insufficient. The result obtained from the risk analysis of the projects applied in this study is realistic. This is because the total constrnction cost obtained after allowing for contingencies determined through this methodology roughly corresponded to the actual cost of the project.