I agree. FMEA has separate fields for likelihood of occurrence and severity of occurrence. I agree with this in that the severity of occurrence is high/critical

I agree. FMEA has separate fields for likelihood of occurrence and severity of occurrence. I agree with this in that the severity of occurrence is high/critical. The likelihood, I don't think it described that well. To put an analogy, I've participated in FMEA's for software systems and the underlying physical installation. The severity of losing your class A data center due to flooding or complete power loss for over 48 hours is critical. The likelihood is low depending on the data centers location. Question is how likely were the unintentional discharge scenarios in this FMEA study?