SUBJECT: Ph.D. Proposal Presentation
BY: Giovanni Maronati
TIME: Wednesday, December 14, 2016, 1:30 p.m.
PLACE: Boggs, 3-47
TITLE: Modeling correlated random variables through Monte Carlo simulations: a novel approach to estimate Nuclear Power Plant Total Capital Investment Cost
COMMITTEE: Bojan Petrovic, Chair (ME)
Dan Kotlyar (ME)
Nolan E. Hertel (ME)
Chelsea C. White III (ISyE)
Paolo Ferroni (WEC)
Jurie J. Van Wyk (WEC)


Total Capital Investment Cost (TCIC) is the measure of merit that represents the cost of design, construction, and testing of a Nuclear Power Plant (NPP) up to commercial operation.
The recent construction of NPPs in the US and in Europe shows that uncertainties in the construction process lead projects over the initial budget. Wrong and inaccurate predictions in project duration and cost threatens the future nuclear industry and may push the nuclear energy production out of market.
In this work a methodology to estimate TCIC for an NPP is presented. A construction model based on a bottom-up approach was used to calculate TCIC for the Westinghouse Small Modular Reactor. The model shows the impact of different design variables on TCIC. A top-down approach was used to calculate TCIC for the Integral Inherently Safe Light Water Reactor (I2S-LWR).
The effect of uncertainties on TCIC and project duration for both cases is estimated through Monte Carlo simulations. The approach is based on the use of multi variate probability distributions to model correlations between random variables. Results of this methodology show a more accurate estimate and an improved prediction on TCIC uncertainty as correlations between variables are considered.
The estimating method presented here, used at an early stage of a reactor development, can inform both the design team and the decision maker and lead to more economically competitive designs.