Cost Effective Insurance: The FMEA Process

You prepare for the unexpected in life. Here’s how to do something similar for the reliability of your equipment and processes.

Failure Modes & Effects Analysis (FMEA) is a structured procedure for achieving improved reliability and continuity of operation in critical equipment and processes. This analysis can be used in many applications, but here the emphasis will be on new equipment or processes proposed for manufacturers that operate on a 24/7 basis. Briefly, the method consists of defining conceivable failure modes, determining the effect of each, assigning weighting factors and deriving numerical results that allow a logical ranking of corrective measures.

It’s worth noting that FMEA is most effective when used early in the design or implementation sequence, when modifications are far easier and less expensive than later — when costs of changes and, more importantly, lost production time can be major. This tool can help users move from a reactive to a proactive mode of manufacturing management. Handled correctly, it could be viewed as a form of cheap insurance for the reliability of manufacturing processes.

FMEA is certainly not one of those notorious “flavor-of-the-month” processes or programs. First used in the aircraft industry in the 1950s, it was later adopted by the automotive and aerospace industries and medical equipment manufacturers. These days, software developers and financial analysts use it. It also has become part of the SMRP (Society of Reliability & Maintenance Professionals) “Body of Knowledge” recommended for study by anyone preparing for the Certified Maintenance and Reliability Professional (CMRP) examination. There is justifiable recognition of FMEA’s value in many types of applications.

The FMEA procedure
The FMEA evaluation is intended to provide quantitative indicators of failure risk that can be used to effectively rank order implementation of corrective actions. The key derived parameter is known as the Risk Priority Number (RPN). Procedural steps follow:

• The vital, first task in developing RPN values is to identify all of the failure modes that apply to the proposed equipment or component being evaluated. The nature of these failure modes will depend on the attributes of the proposed design and the service conditions in which the equipment will be expected to function. Some typical failure modes might include static stress overload, mechanical fatigue, corrosion, wear or elevated electrical resistance. (See Reference 1 for descriptions and illustrations of many varieties of failure in mechanical and metallurgical areas.)
• The second step is to define the primary effect on operations of each failure mode. Some typical effects associated with the previous examples of failure include permanent (plastic) deformation or rupture of a given component, reduced life of rotating equipment, pin-hole leaks or rupture of a given component and premature replacement caused by wear or failure of a safety alarm to function at the available operating voltage.
• Next, three numerical weight or influence factors are defined by the analyst to determine the primary effect on operations produced by each failure mode. A scale of relative importance of 1 to 10 is used, with 1 denoting no negative results and 10 indicating maximum negative results. The first factor is the severity (S) of the effect on operations. The second factor is the probability of occurrence (O) of that effect-1 denotes low probability and 10 the highest probability. The third factor is the probability of that effect being detected (D) before a major failure occurs. Here, a small selected value for D indicates a high probability of early detection. A large value from the range indicates a low probability of early detection.
• The value of RPN is the product of these three factors for each failure mode: RPN=S x O x D. Larger RPN values indicate failure-mode effects that present the most risk to reliable, continuous operations.
• The next steps require identification and implementation of effective corrective actions or changes to the proposed design for each failure mode. The relative magnitudes of the different RPN values permit the analyst to prioritize corrections so as to have the greatest potential influence on the resulting reliability of the new equipment.

Some guidelines for using FMEA
FMEA should not be limited to new equipment or processes. Such an analysis also can provide useful information on applying corrective actions when either significant modifications are made to existing equipment or when more-severe service conditions become necessary with the same equipment. In each case, new failure modes or effects that can adversely affect operational continuity, in the absence of compensating changes, may be created.

It can be useful to repeat the FMEA procedure — often more than once — after initial corrective changes resulting from the first iteration have been implemented in the design. These design changes are, of course, done before any fabrication, installation or final purchasing tasks begin. The complexity of the proposed equipment or process will indicate whether repetition of the procedure is justified. Follow-up evaluations may include modified failure modes, effects and values of the S, O and D parameters, created because of design changes from the prior FMEA.

Depending on the given application and availability of funds, corrective actions for all identified failure-mode effects may not be implemented. Those with the highest RPN values will generally produce the biggest “bang for the buck.” Keep in mind, however, that application of informed engineering judgment is essential in deciding whether to delete some corrective action(s).

Advantages of FMEA…

• The FMEA procedure provides a logical, structured format for working through the difficult thinking and decision-making required to produce significant improvements in operational reliability.

FMEA is certainly not one of those notorious “flavor-of-the-month” processes or programs. There is justifiable recognition of its value in many types of applications.

• Typically, FMEA is completed early in the design/development process when changes to improve reliability are most cost-effective. Thus, expensive change orders or time delays that occur later, when the new design is being fabricated, installed or, worse yet, after an unplanned shutdown has occurred, are avoided or minimized.
• Design changes can be ranked and implemented efficiently by treating the most important first.
• Getting honest input for the FMEA analysis from all groups that have a vested interest in the proposal (e.g., management, operations, project engineering, reliability engineering and maintenance) is both useful and smart. It’s useful because it reinforces the quality of the analysis. It’s smart because these various “buy-ins” increase the probability of positive support, once the proposed equipment is in service.
• FMEA documents relationships between specific design features, service conditions and failure modes that are captured by the structured, concise format of the procedure. Retention of this information then becomes valuable historical data in the evaluation of future proposals.
• Completion and retention of FMEA data provides a definitive record of the organization’s due diligence in its planning and engineering work. These records could later become invaluable in refuting accusations of inadequate attention to personnel safety or environmental effects (which are actually caused by other causes), in the event of future litigation.

Requirements and limitations of FMEA…

• More than approval from management is required (i.e., strong advocacy is needed).
• Considerable commitment by the person leading the FMEA task and his or her (typically small) team is essential. This is especially true for complex new proposals. (It is in just these types of applications, where thorough analyses are not conducted, that failures may be most probable and costly.)
• The ideal FMEA task leader will demonstrate a rare combination of skills and talents. First, he or she should have diverse engineering experience, as well as a solid, practical knowledge of the overall manufacturing operation and the specific technology being proposed. Equally important, this person needs the objectivity and the soft skills to obtain and apply the truth from all vested interests. Management’s advocacy is critical in helping motivate cooperation of all parties. The FMEA leader and necessary team members may exist in-house or an outside consultant may be needed.
• In complex applications, where individual components in the overall system need separate analysis, it may be most efficient to use specialized software designed to work through the FMEA process. This also simplifies record retention.
• Clearly, each new proposal — or modification to existing equipment — will first require a quick evaluation to establish the major potential failure modes and their consequences before FMEA is conducted. If these consequences are minor, then FMEA likely is not going to be justified.

Weighing your options
This discussion has only scratched the surface of the Failure Modes & Effects Analysis procedure and some of its characteristics. The references cited at the conclusion of this article present far more information on the subject. Reference 4, in particular, provides cogent arguments to support FMEA and other proactive approaches to reliability.

The major value of up-front FMEA — before the design process or equipment modification has been completed — is that it forces a thorough, logical thinking process at the optimal point in the new design (or design-modification) phase, when corrective changes can and should be made. The keys to success in implementation are management’s commitment and establishing a suitable task leader or champion.

In appropriate applications, investment in the FMEA procedure will be insignificant compared with not doing a thorough evaluation and then experiencing major disruptions and expenses during operation — via compromised personnel safety, equipment failure and/or lost production. Although you hope against hope that you’ll never need it, it’s in anticipation of those types of situations that FMEA can be a very good type of insurance policy. As the TV pitchman said, “You can pay me now or you can pay me later.”

Which option will you choose for your operations? MT

References
1. API Recommended Practice 571, “Damage Mechanisms Affecting Fixed Equipment in the Refining Industry.” (This document describes multiple failure modes that can occur in a variety of industries and equipment).
2. Bowles, J.B., “Failure Modes and Effects Analysis,” ASM International Handbook, Vol. 11, Failure Analysis and Prevention, 2002, pgs. 50-59.
3. Useful Websites: www.fmeainfocentre.com, www.reliasoft.com, www.weibull.com and www.npd-solutions.com
4. Williamson, Bob, “Getting Operations’ Buy-In for Reliability,” Maintenance Technology, March 2008, pgs. 6-8.

Gerald O. “Jerry” Davis, P.E., is a principal in Davis Materials & Mechanical Engineering, Inc. (DMME), a consulting engineering firm based in Richmond, VA. He holds graduate degrees in both engineering and business and spent 31 years working in mechanical, metallurgical and corrosion engineering functions for several organizations, including the U.S. Air Force, Honeywell and Battelle Memorial Institute. Telephone: (804) 967-9129; Internet: www.dmm-engr.com; E-mail: dmme@verizon.net