Wait a minute. I reread your post. You have confused me. An 8-D is a reaction to a problem.
I don't think you can show the effectiveness of preventive action.
> ...8.5.3 PREVENTIVE ACTION
> The organization shall identify preventive action to eliminate the causes of
> potential nonconformities to prevent occurrence. Preventive actions taken
> shall be appropriate to the impact of the potential problems.
> The documented procedure for preventive action shall define requirements for
> (a) identifying potential nonconformities and their causes; (b) determining
> and ensuring the implementation of preventive action needed; (c) recording
> results of action taken; (b) reviewing of preventive action taken.
It says 'Reviewing of the action taken' but if you didn't have an existing problem you have no data to compare it to. Nope - I can't see how you can show a preventive action is effective.
I prefer to use the following definitions:
- Corrective action: is focused on the effect. It avoids that the non-conforming product arrives to the customer. For instance: doing a 100 % check to sort out deffective product.
- Preventive action: is focused on the cause. Avoids recurrence of the problem.
With these definitions the 8D can be considered a valid method for implementing preventive actions. The effectiveness can also be evaluated.
All you are saying is you can verify the preventive action does, in fact, prevent or segregate the predicted nonconformance. I guess you can call that verifying the effectiveness. I guess my way of thinking is that to verify effectiveness you have to have before and after data.
I agree within your definitions.
[This message has been edited by Marc Smith (edited 18 November 2000).]
I think the 8D is but only one tool to be used in what can be called "preventive action". My 8D form uses the terminology on the 7th "D" as "Action to prevent recurrance". This is reactionary to a nonconformance condition when applied solely to the nonconformance itself specifically. By identifying the "root cause" and searching other potentially effected areas or products or operations, etc., where the correction to the nonconformance can now become an action to prevent a nonconformance where none had actually happened. I feel that by calculating the risk (in a qualitative sense only) that resided in the other potentially effected areas and applying the determined action to prevent the nonconformance, based upon the correction to the root cause, the effectiveness of the preventive element would be in proportion to the effectiveness of the corrective element. This effectiveness would still be theoretical based on the one verifiable piece of evidence.
Of course in the purest sense of the term, preventive action would be seeking potential for nonconformities and eliminating them ahead of time. In that case an 8D would be useless as there would be nothing to use as a reference for effectiveness.
I learned corrective actions a long time ago. Well before Ford started calling it an 8-D. Preventive action in those days was what you did (what we now call poke-yoke) to ensure that if your 'corrective' action failed the defect would be identified. To me, preventive action as it related to a corrective action is typically a poke yoke. Most of the 'hot words' are little more than variations on a theme, if you will.
Evolution of the definition of preventive action has reached the point where the expectation is that you predict failure modes - as I understand it. In automotive the DFMEA and PFMEA address the predictive aspect for the most part. In other sectors the ISO folks are asking that a company look at data and predict from that - as I understand it.
In so far as using the 8-D to 'track' the potential problem, after some thought and reading I now agree. This is embarrassing as I have never been wrong before...
Some musings and some extracts from some listserves - with respect to ISO 9001:2000:
Element 8.5.3 of the new standard states that your organization must identify preventive action to eliminate the causes of potential nonconformities to prevent them from becoming real nonconformities. The preventive action taken must be appropriate to the impact of the potential problems.
The ‘old farts’ out there like me may have noticed an evolution of what a preventive action is. In days of old (well, the 1980’s) a preventive action was part of the result of a corrective action. It used to be what you did to ensure a problem which had already occurred would not again or would be caught through a Poke-Yoke or other mistake-proofing methodology.
Problem: Part not machined.
Reason: Tool broke
Root Cause: Feed Speed set too high – Operator error
Corrective Action: Revise tool change instruction, train operators.
Preventive Action: Sensor to check shaft OD post-operation to ensure cut was made. A failure will stop machine.
Things have changed to predictive (potential). The standard requires a documented procedure for preventive action that defines the requirements for:
a) identifying potential nonconformities and their causes;
b) determining and ensuring the implementation of preventive action needed;
c) recording results of action taken;
d) reviewing of preventive action taken.
Note that clauses 8.5.2 Corrective Action and 8.5.3 Preventive Action are almost identical, hence the processes that are developed for them will be very similar. It is strongly recommended that your organization use the processes developed for corrective actions to address preventive action to standardize the approach and make it easier to implement. The personnel with overall responsibility for preventive action are likely the same personnel with overall responsibility for corrective action.
Preventive action is action taken to eliminate the cause of a potential problem that has not occurred. Your organization has to ensure that the action taken is appropriate to the potential problem. The personnel who ensure it is appropriate preventive action should be the same people that ensure appropriate corrective action.
Preventive action focuses on studying your system and looking for where a problem might occur. Think data analysis, including looking for trends. Management then takes actions to ensure that the problem doesn't occur. Section 8.4.3 - Improvement processes - states that continuous improvement will be part of your management system. In other words, you look for both potential problems and opportunities for improvement at the same time. Thus, Preventive action and Continuous improvement can be easily combined.
To satisfy the requirements for preventive action you must develop a process for identifying the potential nonconformances and their causes, for determining and ensuring the implementation of the preventive action needed, then recording the results of the action taken and finally reviewing the preventive action taken. See corrective actions (8.5.2) herein for more background on this process and element.
The major difference between corrective and preventive action is that the nonconformances don't find you, you find the [potential] nonconformances. Some organizations find it quite a bit easier to be reactive and not proactive - preventive actions are not as common as corrective actions. The secret here is the ability to identify potential nonconformances and then correct them before they become actual nonconformances. This obviously is not easy, but invested interest in identifying nonconformances before they occur is well worth it - ask anyone who has had a product recall (I am sure we are all familiar with the recent news of product recalls).
Potential Audit Questions
1. How well integrated are your preventive action and continuous improvement actions?
2. Do you feel your organization seeks preventive actions enough? How could you do more to satisfy this requirement? How could your organization do more to excel at this requirement?
3. How does your organization identify the need for preventive actions? Do all workers identify preventive actions, or do you have a team or someone responsible for identifying preventive actions?
4. How does your organization assess whether or not the preventive action is appropriate? Do many preventive actions arise without ever being implemented? Why?
--Patti inquired (in part);
--> We don't understand just how
--> far we need to go with preventive actions: do they need to be
--> documented & approved by someone, & then followed-up on
--> for effectiveness (similar to corrective actions)? Or is it
--> sufficient just to document what "preventive action" situations
--> have taken place & file them for later management review?"
> I have also encountered external auditors that come in looking for an entirely
> separate system for the handling of "preventive actions". What I have done to
> steer them in the right direction (toward proper interpretation of the
> 'intent' of the ISO 900X requirement) is to show the auditor where in our
> system preventive actions are incorporated. The topmost place in any QMS
> system to illustrate the presence of preventive actions is the existence of
> the system itself. The fact that you have a working QMS that requires
> management to periodically analyze the effectiveness of the system (internal
> audit results, customer complaints, internal nonconformities, etc.) with an
> eye toward undesirable trends is in and of itself a preventive measure
> (management is after all reviewing this information in order to improve the
> system where needed and thereby PREVENTING future problems). This is just one
> area of your existing system that you can use to demonstrate "preventative
> actions". I am sure that there are other areas within your existing system
> (supplier evaluations, training, etc.) that you can use as well.
> Hope this is helpful to get the 'ole grey matter' going.
> Ethan Andrews
I can't re-lookup and change every old link to posts in the old forums. Most of them *should* still be there. If you want to find the post in the New forums, if the link to the thread in the old forums works (most of them should...), look at the thread (topic) title and what forum it is in. Then back here in the New forums - go to that forum and look for the thread topic title - OR - do a Search for the key words from Title (Note - you can search entire threads or just the 'subject' or 'title' - if you look in the Forums search page you'll see the options.
This is an area we have had considerable discussions internally and with our QS-9000 registrar. There was never any real argument over corrective action...I think everyone can agree to that. When it comes to preventive action, the debate erupts. I view it this way...Say that there was an issue on product X. The 8-D process was followed and as part of step 7, Prevent Recurrence, actions were taken to make sure that the problem "never, ever, ever" happened again on product X. This would be reactionary to the issue and would be considered as part of corrective action. However, if I then took the same fixes and applied them to products A, B, and C where this problem has not occurred yet, this is clearly preventive action. This approach will "...eliminate potential causes of conformities" for products A, B, and C by using "...appropriate sources of information." Unfortunately the Big 3 in there infinite wisdom and omniscience further clouded the picture and added section 22.214.171.124 Corrective Action Impact which says we need to apply actions to similar processes and products.
We have also taken the approach of assigning preventive actions when our auditor (external and internal) identify "opportunities for improvement". As the issue identified is compliant but not very robust against future nonfonformances, we view this as a great opportunity for some preventive action. Of course there's always one in the crowd that will argue that this is more like continuous improvement.