“…Notice that, although (5) and ( 6) are feedforward loops, the existing techniques (Chang and Hwang, 1992;Hunt et al, 1992c) are not suitable for developing fault trees corresponding to the event r(-l). This is due to the fact that the starting nodes of both loops, m2 and s6, are located on the same NFBL (3) and, thus, their effects cannot be considered independently.…”