Evaluating the cost of failure risk: A case study of the kang-wei-kou stream diversion project

Abstract

For water resource engineering projects, some potential risks that are unclear and difficult to investigate are the most crucial problem for engineers. This study presents a unique case of unknown risks, the Kang-Wei-Kou (KWK) Diversion project, which collapsed due to an unexpected hydraulic jump resulting from uneven regional rainfall. We used HEC-RAS incorporating Monte Carlo simulations for analyzing the potential risks to the original design of the KWK Diversion, concerning boundary conditions. An extreme value type I distribution was selected as the form of probability density function (PDF) to upstream discharge, whereas a normal distribution was selected as the form of PDF to downstream water depth, as boundary conditions. Applying the Monte Carlo method, the analysis revealed that the probability of failure was 7.5%, which was non-negligible and should have been considered in the design plan. This case study indicates that overlooking downstream conditions can lead a major facility to fail, and a stochastic analysis could be a way to find out potential risks. We also applied the cost–benefit analysis to review whether the design plans were appropriately estimated with careful risk based consideration.