BDD algorithms based on modularization for fault tree analysis

Highlights: •A new idea integrating truncation means and BDD based on modularization technique. •Reducing memory consumption and computation time. •Using the new insights, Risk Monitor or On-line Risk Monitor becomes more efficient and fast. -- Abstract: This paper explains a Binary Decision Diagram (BDD) algorithm based on modularization to solve a Fault Tree (FT) or Event Tree (ET) in Probabilistic Safety Assessment (PSA) of the nuclear power plant. Fault Tree Analysis (FTA) is a non-deterministic polynomial-time hard problem, both the complexity of the calculation and the large size of real models exponentially increase along with the number of variables, as well as the BDD structure which is transferred from FT model. When solving a large FT in BDD algorithm, two difficulties will be met: the memory is not enough to store large BDD structure and the process generates too much Minimal Cut Sets (MCSs) that is time consuming. In order to overcome these difficulties, this paper presents new insights on integrating truncation means and BDD based on modularization technique. The results indicate that the algorithm is efficient to solve the FT or ET with less memory consumption and in shorter time. This new idea can be applied to Risk Monitor or On-line Risk Monitor of nuclear power plant, which makes the computing process more efficient and fast.