This paper uses the China AP1000 project as an example to exhibit the application of quantitative risk management in nuclear power plant construction projects. For those lump sum contracts, one of the most significant purposes of quantitative risk management is to determine the contingency, i.e. the reserved money and time for projects. This paper studies the application of Monte Carlo simulation in determining the contingency, taking into account the distinctive features of nuclear power projects.
Most nuclear power projects, especially advanced ones such as Generation III and above, meet one common obstacle in estimating key economic indicators — the absence of historical data due to its avant-garde design. As cost estimators of the coal power plant contractors may collect their data from thousands of previous cases, nuclear power plant contractors, especially in many developing countries, do not have a shared database of financial data. Some first-of-a-kind nuclear power plants have absolutely no historical data to look up.
This paper aims to provide a resolution to this problem. First, the feasibility and representativeness of different probability distributions are compared based on their respective skewness and kurtosis to determine the best-suited distribution in nuclear power projects. This paper also analyzes the use of second-order Monte Carlo simulation in reducing the error caused by the biased estimation of inexperienced risk assessment engineers.