R-FMEA. Using Reverse FMEA to Improve Production Processes

Introduction

The Process FMEA is one of the key tools in the modern engineer's toolkit. It helps to identify potential problems in the production process, assess their risk level, and set priorities for actions that reduce that risk. Thanks to PFMEA, we can identify areas for improvement before failures occur, which translates into higher product quality, and therefore, a greater chance of market success for the company.

That's the theory...

In practice, PFMEA is most often developed "from behind a desk" or "in a conference room," without direct contact with the production process, relying instead on data from similar processes or products (assuming we are launching a new product). For this reason, a PFMEA may sometimes turn out to be incomplete. Some potential failure modes, their causes, or detection methods may be overlooked, and control methods may be evaluated too optimistically. Ultimately, it may lead to an incorrectly estimated risk level. As a result, such a PFMEA differs from the actual production process and is not very useful..

This is exactly where R-FMEA (Reverse FMEA) comes to the rescue.

R-FMEA is about "confronting theory with practice," that is, physically verifying whether the developed PFMEA truly reflects the reality on the production floor.

So, let's dive in...

OEM Requirements

Some automotive manufacturers (OEMs) have included in their Customer Specific Requirements (CSR) the obligation to review the P-FMEA, typically using the R-FMEA method. Some examples:

• Renault Group. The use of R-FMEA is required for updating the P-FMEA.[1]
• Ford Motor Company. R-FMEA is required for all new launches, first as a preliminary activity at the equipment supplier, and later after its final installation in the production site.[2]
• Stellantis. The R-PFMEA process must be developed, planned, and monitored. Verification is carried out by intentionally inducing defects at the production station, following a defined R-FMEA execution frequency.[3]
• GM. Tools such as Reverse PFMEA, or other equivalent methods supporting P-FMEA reviews, should be used. The review should be conducted at least once per year.[4]

R-FMEA Planning

Reverse FMEA requires prior planning to allocate resources such as personnel, time, and production line (machine or workstation) availability. The plan should also take into account any customer specific requirements, if defined.

For example, Ford requires performing R-FMEA during equipment implementation, while other manufacturers expect R-FMEA to be carried out on a regular basis. A common approach is to assume that R-FMEA should be performed at least once a year (as in the case of GM).

In summary, R-FMEA reviews should be planned in such a way as to meet customer requirements (its the absolute minimum) while remaining practical and achievable without overextending available resources. Unfortunately, this is often challenging, particularly in factories with hundreds of machines, multiple production lines, and, hmm.. as usual.. too little time :(

The R-FMEA planning form (a document) may take any format, unless specific requirements have been established with a particular customer.

R-FMEA Execution

The execution of R-FMEA can be based on various methods developed within a given factory. I suggest the following approach:

• R-FMEA Team. The team should be multidisciplinary (similar to a PFMEA team). I recommend including a person who knows the process and the operation of the specific workstation or machine, as well as a maintenance technician who can assist in testing possible failures and detection methods, and ensure that the team does not damage the equipment during experiments.
• Documentation. The current P-FMEA analysis (within the scope of the R-FMEA) should be available for review. I also suggest collecting data on known issues, customer complaints, process defect rates (e.g., PPM), and similar information.
• R-FMEA Checklist. Prepare an R-FMEA checklist for the given process. This checklist will be helpful during the actual verification of the process in terms of potential failures, their causes, and detection methods.
• Process Observation. I recommend first observing the production process operating at normal (full) capacity for a period of time, e.g., 30 minutes. Do not interfere with or disturb operators. During this time, identify issues such as breakdowns, stoppages, adjustments, errors, unusual operator actions, dropped parts, etc.
• On-Line Verification. At this stage, stop the production process and perform the activities listed in the checklist. For example, try skipping an operation, inserting a part incorrectly, interrupting the process mid-cycle, or “tricking” the error-proofing or mistake-proofing system. Check whether the machine detects the error and how it reacts. Record the results, for example, in the R-FMEA checklist or form.
• P-FMEA Update. After completing the observation and testing on the line, return to the P-FMEA analysis to review and update the entries. Add any newly discovered issues and reassess the associated risk levels.
• Risk Reduction Actions. Following the P-FMEA update, new risks requiring mitigation may appear. Appropriate actions should therefore be implemented to reduce these risks. After implementation, update the P-FMEA again to ensure that it remains consistent with actual conditions.

A record of the performed R-FMEA should be retained—at a minimum, documentation confirming that a review of the P-FMEA was conducted as part of the R-FMEA. It is also recommended to keep supporting documents, such as the completed checklist and process observation results.

Remember that after updating the P-FMEA, you should review and, if necessary, update the control plan.

Summary

Improving a process (and consequently the product manufactured within that process) requires that the process operates reliably. The occurrence (or shipment) of nonconforming products should be minimized through appropriate solutions such as Poka-Yoke, Error-Proofing, or Mistake-Proofing, as well as through the use of SPC, maintaining high process capability Cp, Cpk, process performance Pp, Ppk, and machine capability Cm, Cmk.

A key element in this framework is the P-FMEA analysis, which should accurately reflect the actual manufacturing process. That's why R-FMEA is such an effective approach. It requires manufacturing plants to compare the P-FMEA documentation with real production conditions, discover new risks, and take action to reduce them.

The purpose of R-FMEA is to promote a proactive approach, improving the process through its review, physical testing, and identification of weaknesses before actual defects occur, which could otherwise lead to complaints, higher costs, and damage to the company's reputation. Therefore, I believe that the R-FMEA is an excellent tool for process improvement.