Hybrid Diagnostics - The ability to diagnose designs that contain both functions and failure modes—in particular, the ability to infer the status of functions from known status of failure modes (and vice versa).
For example, if it has been verified that a particular failure mode has not occurred, then the likelihood of each of its affected functions occurring must be reduced accordingly. When all of the failure modes that affect a particular function have been ruled out, then that function can be inferred to be good. Conversely, if a particular function has been proven to be good, then any failure mode that always affects that function can be inferred to have not occurred. Failure modes that sometimes affect a function cannot be ruled out until all affected functions have been proven good.
Hydrid diagnostics require that the failure probabilities associated with functions and failure modes be dynamically adjusted as diagnostics are processed—for example, if it has been determined that a particular failure mode has not occurred, the probability associated with that failure mode must somehow be subtracted from the failure probabilities of its affected functions. In order for this adjustment to be performed, functional and failure mode probabilities must be correlated (based on the Failure Probability Calculation Method) that has been specified for each object).
