PFMEA RPN Rating and Mitigation

Locked lesson.

Upgrade

About this lesson

The severity, occurrence, and detection ratings are combined to create a Risk Priority Number (RPN). When that RPN exceeds the organization's risk threshold for Process FMEAs, a mitigation strategy is developed and implemented to lower the scores until the RPN is acceptable.

Quick reference

PFMEA RPN Rating and Mitigation

When the risk priority number, which is the product of the three rating scores, exceeds the business threshold, a mitigation action should be made to improve the process and reduce the risk.

When to use

After the scoring for severity, probability of occurrence and detection are complete, the risk priority number (RPN) can be calculated.  If that value exceeds the business threshold value, a mitigation action should be initiated.

Instructions

The goal of the Process FMEA is to reduce risk.  The RPN is a means of quantifying the risk associated with each failure.  If the risk is considered to be excessive, mitigation is appropriate.  In some cases, a business may choose to go into production with a process that has failure modes with high RPN values.  However, that should be a conscious decision by management with a full understanding that they are activating a risky process.  If a mitigation action is taken, the process should be rescored to ensure the mitigation is effective and to determine if there are any unintended consequences.

Risk priority number

The RPN is determined by multiplying the severity, probability of occurrence and detection scores together.  Since the range we are using for each of those scores is 1 to 10, the range for the RPN is 1 to 1,000.  Most organizations will set their RPN threshold level at a value near 100, and that is what we will be using for the exercise in this lesson. In some cases, there is industry or customer guidance for the threshold level. 

The RPN can be used both in its absolute value state to get a sense of the magnitude of the risk and it can be used in a relative sense to determine which risks should be addressed first in a process – the highest.  Some organizations have a threshold requiring mitigation and a zone that ranges slightly above that threshold as a caution.  This is to account for the fact that the probability of occurrence score can change over time as the process yield improves or degrades.  For that reason, it is appropriate to periodically revisit the Process FMEA and determine if the recent process yield information has indicated a change in the risk status of any failure modes.

Mitigation

Mitigation of high risk failure modes in a Process FMEA can be somewhat challenging.  Typically there is very little that can be done about the severity score.  The product design is normally frozen and the primary and secondary functions are well defined.  So there is little than can be done to lower those scores.  

The probability of occurrence score is based upon the actual process quality performance.  For this to improve, there normally needs to be some change to the quality management systems used to control the process or the implementation of problem solving projects to remove sources of defects.  In either case, this will take action on the part of the FMEA analysis team.  During the process development, it is often easier to make changes to the quality management processes that will be used in the process.  Once the process is baselined and in production, often a problem solving project is needed to improve the probability of occurrence score. 

The detection score is closely linked to the quality control plan and approach used in the process.  Adding automation will usually improve the score, however, automation can be expensive to install.  Simple handheld automatic test equipment is a significant benefit and will often be less expensive.  Obviously, the prevention approaches are the best approaches but some aspects of them are best implemented at the time of process design.

Two proven prevention methodologies are Poka Yoke, which means mistake proofing, and Five “S” Disciplines which stand for sort, set, shine, standardize, and sustain.  The Poka Yoke methodology uses product and process design features to control the process for success.  Through product design, fixture or tooling design and sensors, the system self-inspects and ensures that every action is done correctly before the next action can be attempted.  The Five “S” Disciplines focus on the organization and management of the workstation and work-cell.   This approach simplifies the workstation and reduces the opportunity for operator errors.

Example

In the ball point pen example we saw that most of the failure modes exceeded the threshold of 100.  Several mitigation actions were implemented.  The implementation of a bar code system helped to automate the kitting, and a re-design Kanban system improves the initial step of the process.  A cleaning fixture and a prepared cleaning packet with the right amount and right type of cleaner improved the cleaning operation.  A decal fixture checks the alignment and ensures the prep work was done correctly.  The fixtures reduce the risk of damaging the parts and Poka Yoke some of the steps.  In the inspection step, an automated vision system is being installed, but is still operational, so the scoring for that failure mode has not changed, yet.  However, for our purposes, I have shown how the occurrence and detection scores will change once the system is working.  After the vision inspection system is installed, there will just be three failure modes that are at or slightly over 100.  These are all associated with the new fixtures and after some experience is gained with them, the occurrence score may drop further.

Hints & tips

  • Mitigation takes work.  Make sure the change is implemented before giving credit for a lower score.
  • The FMEA team will need to make some judgement calls at times when rescoring the mitigated process with respect to the impact it will have.  The tendency is to be overly generous with the scores.  Make sure every score value is justified.
  • The completed FMEA should be filed and maintained with the other process documentation.
Login to download
  • 00:04 Hi, I'm Ray Sheen.
  • 00:06 Well, we've mapped our process, scored each failure mode,
  • 00:09 now it's time to do something with the results.
  • 00:12 Let's look at risk priority number and mitigation strategies.
  • 00:17 First, a quick refresher on the risk priority number or RPN.
  • 00:21 The RPN is calculated by multiplying the severity, probability of occurrence,
  • 00:26 and detection scores together.
  • 00:27 The RPN will be a value between 1 and 1,000.
  • 00:31 Your organization will establish a threshold for the RPN.
  • 00:35 If the value of a failure mode is above the threshold,
  • 00:38 the failure mode should be mitigated.
  • 00:40 If the value is below the threshold, mitigation is not needed.
  • 00:43 The actual value will vary from industry to industry,
  • 00:46 and sometimes from company to company within an industry.
  • 00:49 Generally, the value is near 100,
  • 00:51 and that is the value we will use in our examples in case study.
  • 00:55 Some organizations will require a mitigation based upon criticality alone,
  • 00:59 which you may recall from an earlier lesson is found using the severity and
  • 01:04 probability of occurrence scores.
  • 01:06 And remember, we're dealing with risk.
  • 01:09 An RPN below the threshold does not mean that the failure mode is impossible.
  • 01:13 It just means that the organizational risk from that failure is so low or
  • 01:17 has already been mitigated, so no further action is needed at this time.
  • 01:21 Now, let's talk about mitigation strategies.
  • 01:24 The risk resolution or failure mode for
  • 01:26 a Process FMEA will usually focus on process design and control.
  • 01:31 The major reason for this is the product design is usually already set, so
  • 01:35 that has become off limits for improvements.
  • 01:38 To mitigate severity, a product design change is normally required.
  • 01:42 Of course, if you're at a nine or
  • 01:43 ten level, you may be able to provide some safety and hazard warnings.
  • 01:47 But otherwise, this will often take more time and
  • 01:50 be more difficult to achieve than one of the other mitigation strategies.
  • 01:54 A high probability of occurrence will normally require an improved process
  • 01:58 control approach or a problem-solving process to improve the process yield.
  • 02:03 It may need better controls of the inputs,
  • 02:06 a better overall quality management approach, or better tools and equipment.
  • 02:10 Change what needs to be changed to reduce operational defects and increase yield.
  • 02:16 Remember, the occurrence score is all about the percentage yield at each step,
  • 02:20 and it's based upon actual performance, not wishful thinking.
  • 02:24 And a high detection score normally means you need to change the process controls
  • 02:29 using different inspection and test approaches,
  • 02:32 which may mean changes to the equipment or the facility.
  • 02:36 Let's consider how the strategy changes depending upon where you are in
  • 02:40 the product and process life cycles.
  • 02:42 I'll start with what to do during the development phase.
  • 02:46 A great time to use a Process FMEA is during process development.
  • 02:50 That way, when you find a failure with a high RPN,
  • 02:52 you will probably have multiple options for how to mitigate that risk.
  • 02:58 You could change facilities or equipment to ones with a higher process yield and
  • 03:02 more inline automated inspection and test points.
  • 03:05 You could design and
  • 03:06 calibrate a process control system such as statistical process control.
  • 03:11 Of course, many times, you can add Poka Yoke features to tooling and
  • 03:14 fixturing to eliminate some failure modes.
  • 03:17 And of course, you wanna document all that good work you did, and
  • 03:20 the Process FMEA is an excellent vehicle for that purpose.
  • 03:24 In some cases, you can design out severity and
  • 03:28 design risk response capability in through inspection and test processes.
  • 03:32 Another reason to do the analysis at this time is that at least in some industries,
  • 03:37 once a process has been validated, it's almost impossible to get regulatory
  • 03:41 approval to change the process, even to improve it.
  • 03:45 So, you wanna start off with an excellent platform on day 1 of the process
  • 03:49 operation.
  • 03:51 Just because it's best to do your mitigation during the process development
  • 03:54 stage, that doesn't mean you can't find improvements
  • 03:57 once the process is in operation.
  • 03:59 It's just that some options may be limited.
  • 04:02 During operations, the focus will be to reduce the defect rates
  • 04:05 which affect the probability occurrence, and to increase the detection capability,
  • 04:10 which normally means improving the inspection and test steps.
  • 04:13 Now, these are usually considered good things, so
  • 04:16 you probably won't have resistance on either point.
  • 04:19 To reduce the probability of occurrence score,
  • 04:21 you need to improve your quality control methods.
  • 04:24 That could mean upgrades to your control plan or
  • 04:26 upgrades to your operator's skill set.
  • 04:29 It could also mean changing some things like adding statistical process control
  • 04:34 and changing your maintenance and calibration approach.
  • 04:36 To reduce the score for detection, you have to change the inspection and
  • 04:40 test methods.
  • 04:41 Use automated in-station testing where possible.
  • 04:44 Avoid manual inspection or attribute gauging and use variable gauging instead.
  • 04:50 Let me take just a few minutes to mention some process disciplines
  • 04:54 that can assist you when mitigating failure modes.
  • 04:57 These two techniques help to reduce both the occurrence score and
  • 05:00 reduce the detection score.
  • 05:02 Both are based upon principles that were pioneered in Japanese manufacturing
  • 05:06 operations.
  • 05:07 The first is Poka Yoke, which means mistake proofing.
  • 05:10 This is designing process controls so that it is very difficult or
  • 05:13 impossible to make a mistake.
  • 05:15 For instance, making all the screws on an assembly the same size and
  • 05:19 type make it impossible to put the wrong screw into a screwhole.
  • 05:23 The second one is the 5S disciplines.
  • 05:27 Sort, set, shine, standardize, and sustain.
  • 05:30 Using these in an operation will reduce errors and
  • 05:33 speed the work, since distraction and confusion is minimized.
  • 05:37 So let's finish with our pen example.
  • 05:39 As you can see, there are a lot of failure modes with an RPN in excess of 100,
  • 05:43 and requiring mitigation.
  • 05:45 Well, I didn't touch the pen design, and
  • 05:48 that means that all the severity scores stayed the same.
  • 05:51 However, I designed and built some fixtures with Poka Yoke elements
  • 05:55 to help with handling the parts, cleaning the barrel, and applying the decals.
  • 06:00 Also, I changed the process to use prepackaged wipes, and
  • 06:03 that removed the problems associated with the wrong cleaner or too much cleaner.
  • 06:07 Most of the changes are now in and ready to go.
  • 06:10 I do have one last automated inspection fixture that is not yet in place, so
  • 06:15 I'm still showing that failure mode as a red.
  • 06:18 However, for training purposes,
  • 06:19 I am showing you what the new scores would be for occurrence and detection.
  • 06:23 But I won't complete that item until the equipment is installed and working.
  • 06:28 Several of the failure modes are just over 100.
  • 06:32 I marked those as yellow.
  • 06:33 The business would need to decide whether or not to accept those risks or
  • 06:37 do a redesign of the product.
  • 06:40 And one interesting point,
  • 06:41 if we do redesign the product to get rid of the decal, and instead,
  • 06:45 laser edge mom's diner on the pen, most of the failure modes on this page go away.
  • 06:51 Since most of them are prepping the decal, or applying the decal, or
  • 06:54 inspecting the decal.
  • 06:56 And remember, if they do go away,
  • 06:58 I will have new failure modes associated with the laser etching process.
  • 07:02 One final comment.
  • 07:04 Remember, the last step of the FMEA process is documentation.
  • 07:08 Once the work is done, document your Process FMEA and
  • 07:12 maintain it with the process documentation.
  • 07:16 Well, at the end of the day, this is what the Process FMEA was all about.
  • 07:21 Deciding what needs to be changed to reduce risk, and
  • 07:24 then implementing those changes.

Lesson notes are only available for subscribers.

PMI, PMP, CAPM and PMBOK are registered marks of the Project Management Institute, Inc.

Gift this course