FMEAs have been prioritized using RPN; however, a new standard has introduced AP for prioritization. This study seeks to determine if the number of required improvement actions increases, decreases, or stays the same when using AP in place of RPN. Statistical software was used to simulate 10,000 combinations of severity, occurrence, and detection. Both AP and RPN were calculated for the 10,000 combinations. Statistical hypothesis testing was performed to determine if there was a difference between RPNs when sorted by AP and to determine if there was a difference in actions required using RPN or AP. There is a statistically significant difference between RPNs when sorted by high, medium, and low AP. Using an RPN threshold equal to or greater than 100 would result in no change in the number of actions required if prioritizing by high and medium, but would result in fewer actions required if only high is used.

This paper describes the p-diagram (parameter-diagram) and its application in planning a DoE (Design of Experiments). A case study describing an actual problem from industry is presented where the planning phase started to go wrong as difficulties in selecting the right variables for the DoE were discovered. Furthermore, running these experiments became prohibitively expensive, due to the large number of such experiments that would be needed, and though the exploitation of a p-diagram it was then possible to come up with a feasible DoE.