Progression of the binary decision diagram conversion methods

The Binary Decision Diagram method has been formulated over the last decade. It is the latest methodology developed to overcome the deficiencies of the common fault tree analysis technique. Its advantages include increased efficiency in determining the qualitative characteristics of a failure mode represented using a fault tree, and improved accuracy when calculating the corresponding quantitative performance measures. The disadvantage of the approach however, is that the conversion from the fault tree can not be guaranteed to be optimal, reducing the advantages of using the method. Until this is rectified the inclusion of this methodology within commercial packages is not possible. The conversion process involves selecting the basic events from the fault tree and generating an ordering from which the Binary Decision Diagram is constructed. It is this ordering which is crucial. A number of conversion alternatives exist in the research literature, from heuristic approaches examining the positioning of the basic events in the fault tree, to the latest developments which look at selection mechanisms using pattern recognition approaches. This paper reviews the conversion approaches available, the advantages and disadvantages of each, and the latest methodologies being investigated.