The impact of fault expansion on the interval estimate for faultdetection coverage

摘要

A high fault detection coverage is very critical for systems withultra-safe requirements and fault injection is an effective techniquefor estimating the coverage. One difficulty of fault injection lies inthe huge number of injections that need to be carried out in order toobtain statistically significant results. Fault expansion has beenproposed as a means of reducing the number of injections by dividingfaults into equivalence classes. We show that this intuitively appealingapproach is only effective when each fault equivalence class is asignificant portion of the fault population