Abstract
The development of the nuclear industry’s Probabilistic Risk Assessment (PRA) has significantly contributed to the design, reliability, and safety of nuclear power plants (NPP). Today, PRAs form integration in all nuclear power plant development stages, including design, licensing, operation, maintenance, and decommissioning. Many legacy PRA tools have been used in the nuclear field; however, more powerful tools are needed to cope with the rapid development of the nuclear industry and NPP designs. These tools are now required to analyze new aspects of the plants that were never envisioned before, and more computational resources are needed.
This study uses synthetically generated fault trees of various sizes and specifications to benchmark fault tree solvers; XFTA version 1.3.1 and SCRAM version 0.16.2. The analysis is performed in two steps. The first step is to compare the probabilities, minimal cut sets, and importance factors. The second step is measuring CPU time, wall time, and the memory usage of each engine for benchmarking. The result for the first step for the same configuration is the same for each calculation engine. Overall, XFTA can complete most of the given runs in a shorter time with less memory usage than SCRAM. All computations and measurements are done on a specified computer.
