An analysis of risk perception and the RPN index within Failure Modes and Effects Analysis
Davie, James L.
MetadataShow full item record
Keywords. Failure Modes and Effects Analysis, Risk Analysis, Risk Perception. Purpose. to determine and if the Failure Modes and Effects Analysis (FMEA) and specifically the Risk Priority Number (RPN) calculation provides a robust method of risk prioritization that accurately reflects natural risk perception. Design/Methodology/Approach. In order to gather a baseline measure of natural risk perception, a survey of questions was provided to individual participants with varying degrees of experience in the use of the FMEA tool. The survey questions proposed situations with varied levels of risk with regard to severity, occurrence, and detection using high, mid, and low value points within a fractional orthogonal array. The survey consisted of two separate surveys containing the same questions arranged in different orders. The questions were first posed as a narrative description of each event and then presented within a FMEA matrix format. The results of the two separate surveys were then compared to determine if the RPN calculation method reflects the natural perception of the risk associated with the identical situation. Findings. The current method of RPN calculation involves the multiplication of three ordinal scale measurement values that are derived from a Likert scale. Because of the limitations of the ordinal scale, many mathematical operations, including multiplication, cannot be performed on these values and hold any useful meaning indicating that the current calculation method is not a robust design. Based on the results of the survey analysis it was discovered that the RPN as it exists within the FMEA model does not reflect the natural perception of risk. It was noted that the perceived severity influences the perception of occurrence and that detection had little impact at all. Therefore the RPN calculation at a minimum should reflect this interaction and no longer weight each of the three components equally. Research Limitations/Implications. This research focused on the perception of risk between our natural tendencies and the RPN methodology in order to generate a base line for comparison. Because the FMEA is intended to be a group effort, a similar study will need to be conducted to determine if the interactions observed at the individual level are reflected at the group level. Practical Implications. The proposed revision to the RPN calculation utilizes easy-to-understand mathematical operations that can be applied through the use of a spreadsheet program. With the removal of the detection component from the determination of required action, the FMEA team cannot depend on additional inspections to mitigate risks associated with the design of a system. Originality/Value. The value of this research is in the increase in the accuracy of the RPN to more closely reflect the natural perception of risk by identifying and incorporating the severity / occurrence interaction.