posted 19 April 2001 11:47 AM            

For root cause analysis, you need to remember these steps:
1. Define the problem. The problem as seen by the initiator may not be the "Whole" problem, it may not be anything more than a minor symptom.
WHAT went or could go wrong
WHO is affected, try to include every possibility
Is it sporadic or continuous, WHEN does it happen?
HOW did it happen (see step 2, this is the beginning of the identification of root cause)
2. Identify the cause. WHY did it happen, why did it not get caught etc. A lot of systems work using the 5Ys. Ask yourself why after each why you come up with for at least five whys. In other words, until you can no longer come up with any more answers.

A great way to use root cause analysis is to get others involved when you perform it. Get input from all or many departments.

posted 20 April 2001 02:40 PM            

ISO Guy, I've been exactly where you are. And I do mean exactly, we were a hair's breadth from loosing our cert due to commitment issues. Can I help you in any way?

posted 20 April 2001 05:09 PM               

Do like me, try www.monster.com , www.headhunter.com , www.hotjobs.com , www.flipdog.com and others. I saw all kinds of Quality jobs on the ASQ site also.

Good luck

Randy

posted 23 April 2001 03:33 PM               

Every effect has one or more causes and my experience tells me it's usually more than one by the time it manifests itself. Prevention is always the best route and a thorough understanding of cause and effect in your particular process will save a lot of time and effort (MONEY) in mitigating problems before they occur.

Wilson, P.F., Dell, L.D. and Anderson, G.F. 1993. Root Cause Analysis. Milwaukee: ASQC Quality Press is a good book to get info from. I'm not sure it's still available. Kepner Tregoe has some course materials that you could look through. There are several other packages out there that aer basically the same but packaged differently. Sort of like the differences in gurus!

Hope this helps some. Let us know what you end up doing and how you get there!

------------------
See you down life's highway!

posted 27 April 2001 10:02 AM               

Our company had gotten so accustomed to the quick-fix mentality that those band-aids became known as the corrective actions taken with a "logical" root cause derived after the action. This was all neatly documented on a single 8D form and kept on file for audit purposes. It got to the point that sorting operations with fulltime employees dedicated to sorting were included in the process flow. Trouble was the resulting numbers (data) still didnt lie. Rework, scrap, labor was still creating internal ppm's in the low 5 figures. When I studied the MRB reports from the previous year I noticed that the stated non-conformities were consistent and reoccuring. I then did a cost study of what appeared to be the biggest offenders and used that data to choose my battle so to speak. Part 123 was to be my test case. My process followed this route:

1. Identfy the problem or symptom
2. What goals are affected - whats the area of impact
3. Interim actions/Band-aids/Quick Fix = Containment (temporary but necessary)
4. I gather a handful of peers with various process knowledge in a brief one-on-one with each. I couldnt coordinate a meeting so I had to be the runner and piece this together for our "brainstorming". No thought or idea was refused, in fact all ideas were listed as Possible Causes.
5. Going through the list, I assemble available data that would validate/invalidate each possible cause. In other words, if Cause "A" results in 5/1000 discrepancies then I move on, but when Cause "B" impacts 425/1000, I have a contender. So I then isolate my primary offenders which are now backed by data. These then are my "root cause" candidates.
6. Now, to respond to the chosen few, my "team" and I (through another tour of interviews) list alternatives or possible solutions to each "root cause" candidate.
7. Now the deciding function - Will the possible solution work? I give it a rating of 0 thru 5 where 0=None, 1=Very Low, 2=Low, 3=Average, 4=High, 5=Very High
8. Feasible? Can it be done? I rating the possible solution for feasibility just as I did in step 7.
9. The Rating! Out of a possible 25 points I get a score by multiplying #7 x #8. Example: Possible solution - Add more overtime to meet deadlines. Will this solution work? score=4 Is it feasible? score=1 (staff already on too much overtime) Rating for this candidate=4 out of a possible 25. This looks to be a poor solution.
10. Rate all of the possible solutions this way and you'll find that a few will stand out as potentially effective approaches to take. As a "preventive" move, consider all other problems stemming from this primary cause (backed by data) and a highly rated solution. Try to gather cost data on the other potentially affected similar problems. Determine an approximate cost to test and implement the solutions across the board and you will have documented a cost justification for the work that needs to be done. In fact, you should be able to forecast a realistic cost reduction for your company. And you will have also (hopefully) documented a proven process (root cause analysis being the primer) that can sell the need to back your RCA and CAPA efforts with the necessary resources.
I know of few breathing individuals in upper management that can turn their heads away when you hit them with $$$$$$$ backed with proof.

As I said in the begining, I am at the tailend of round one of what appears to be a successful effort and I foresee some positive changes in the way we do business.

I know the method I just explained may be fundamental to the experienced and guru's in our forums but having just learned this from a consultant friend and seeing it working compelled me to share it.