ABSTRACT
Fault tree analysis calculates the failure rates of the complex systems in the event trees from more primitive events. A fault tree is a graph with two types of nodes, events and gates. The fault trees already constructed for many reactors contain natural graphical models. This chapter introduces a typical Probabilistic Risk Assessment (PRA) fault tree from the Interim Reliability Evaluation Program study, also used as an example in Spencer, Diegert, and Easterling. It reviews the history of PRA, particularly the WASH-1400 study which marks the beginning of the field. The chapter describes the type of data available as inputs to fault trees, and discusses the end results of a risk assessment and some of the uses to which they could be put. Event trees are models for a sequence of events. The event tree is drawn as a series of branches starting from the initiating event, each branch point corresponding to a particular safety system.
