UPDATES AND CURRENT APPLICATION OF THE MCDET SIMULATION TOOL FOR DYNAMIC PSA

摘要

The MCDET simulation tool is an appropriate tool for dynamic PSA and can essentially support risk-informed decision making (RIDM). Coupled with a deterministic simulator code, it applies Monte Carlo (MC) simulation in combination with the Dynamic Event Tree (DET) approach. This combination is able to treat any kind of epistemic and aleatory uncertainty. The so-called probabilistic module of the MCDET tool implements the routines for MC simulation and DET generation. It is linked to a scheduler supervising the alternate calculations of the module and the simulator code. Current developments are mainly focusing on this scheduler and the simulation control. They aim at essentially reducing the computing time by parallelization and at improving the ease of use by standardization. The MCDET tool and its extra Crew-module are currently applied to analyze the effects of manual fire fighting actions in the frame of a probabilistic fire risk analysis. This paper gives an overview of the updates of the MCDET tool and describes the MCDET analysis of fire fighting actions.