It should be
P(A or B) = P(A) + P(B) - P(A and B)
calculated as 1 - P(!A and !B) = 1 - (1 - P(A))*(1-P(B))
It currently calculates it as P(A) + P(B)
the NRC handbook calls this "rare event approximation" (see page VI-5) for P(A or B), which argues the probability is accurate to within about ten percent of the true probability when P(Ei) < 0.1.
It should be
P(A or B) = P(A) + P(B) - P(A and B)
calculated as 1 - P(!A and !B) = 1 - (1 - P(A))*(1-P(B))
It currently calculates it as P(A) + P(B)
the NRC handbook calls this "rare event approximation" (see page VI-5) for P(A or B), which argues the probability is accurate to within about ten percent of the true probability when P(Ei) < 0.1.