CAN WE TRUST PRA: TAKE 3

摘要

The purpose of this study is to investigate the potential of Binary Decision Diagram (BDD) quantification methods to evaluate the current set of large linked fault trees used in nuclear power plant probabilistic risk assessments. The need for a BDD quantification method stems from desire to eliminate the simplifications that are inherent in the large linked fault tree quantification. These simplifications include the use of the rare event approximation, the use of truncation limits, and the simplified treatment of success terms. As part of a broader initiative to develop the next generation of the logic modeling tools, the BDD approach to the evaluation of large linked fault tree models plays a significant role. The successful development and application of BDD quantification methods can reduce the simplifications in the model thereby reducing resources associated with the model development, use, and maintenance at the same providing a more accurate estimate.