Neural network selection mechanism for BDD construction

The binary decision diagram (BDD) methodology is the latest approach used to improve the analysis of the fault tree diagram, which gives a qualitative and quantitative assessment of specified risks. To convert the fault tree into the necessary BDD format requires the basic events of the tree to be placed in an ordering. The ordering of the basic events is critical to the resulting size of the BDD, and ultimately affects the performance and benefits of this technique. A number of heuristic approaches have been developed to produce an optimal ordering permutation for a specific tree, however they do not always yield a minimal BDD structure for all trees. Latest research considers a neural network approach used to select the ‘best’ ordering permutation from a given set of alternatives. To use this approach characteristics are taken from the fault tree as guidelines to selection of the appropriate ordering permutation. This paper looks at a new method of using the Jacobian matrix to choose the most desired characteristics from the fault tree, which will aid the neural network selection procedure.