Which Root Cause Analysis Method is the most Effective?

[Popeos]

Hi guys (and girls)

I have to recommend/request a way to conduct root cause analysis to a large bunch of suppliers.

In my experience, the best tool to do that is the combined use of Ishikawa/5M and then drilling with 5whys on most likely / confirmed causes.

What is the best root cause analysis tool according to you ?
(common sense being great but not an analysis tool with the same value for all users)

Would you object in making the use of a such a tool mandatory to ensure better root cause analysis ?
(so far we just ask suppliers to use the tool they want and we don't get much)

Thanks for your advice

Popeos

 

[qusys]

Re: Which Root Cause Analysis is the more Effective?

Hi guys (and girls)

I have to recommend/request a way to conduct root cause analysis to a large bunch of suppliers.

In my experience, the best tool to do that is the combined use of Ishikawa/5M and then drilling with 5whys on most likely / confirmed causes.

What is the best root cause analysis tool according to you ?
(common sense being great but not an analysis tool with the same value for all users)

Would you object in making the use of a such a tool mandatory to ensure better root cause analysis ?
(so far we just ask suppliers to use the tool they want and we don't get much)

Thanks for your advice

Popeos

I voted for Ishikawa / 5M, but I think they are equivalent.
The key point is that the involved people shall put effort in following a structured approach and methodology, working in multidisciplinary to indentify the most probable cause.
It is also important the role of facilitatorin this case.
Problem method such as 8G allows several method to search forthe most probable root cause.
Together with Ishikawa diagram it is also important the process of decision making to select the most probable cause (s). it is not soeasy but follwing a method can help.
I do not force suppliers for this, because it is a "natural" method. You should look at the result and help in applying a problem solving method for corrective action.

 

[Solinas]

You've left off a few tools. Off the top of my head (Ok, and a quick google):
Affinity Diagrams
Pareto Charts
C&E Matrix
FMEA
Reenactments
Reconstruction
and the list goes on.

I'd vote that there is no best tool. In many cases, it depends. Is/Is-Not may be very helpful for a particular problem, and it may be worthless for others.

If a supplier is trained in Kepner Tregoe, it would likely work best for them.
If a supplier is trained in brainstorming, affinity diagrams and fishboning, that would likely work best for them.
If a supplier is trained .... you get the idea!

As a QA guy, I wouldn't take too kindly having a customer prescribe what tool(s) I should use. I pick tools based on the company culture, knowledge of the skills of the team (improving as necessary), and draw from my experience in which tool is likely to get the best results.

 

[arios]

Kepner-Tregoe includes the is/is-not anaylsis, so you add the results of both

 

[qusys]

Kepner-Tregoe includes the is/is-not anaylsis, so you add the results of both

Agree.
In reality you can use the combination of all the proposed answer.
Besides, my advice that it is also important the process of decision making for the final decision

 

[Jen Kirley]

My answer, though you may not like it, is to not be confined to one or a few tools because different situations can demand different approaches.

It's been said that most problems can be resolved using the seven core quality tools. I tend to agree.

 

[Jim Wynne]

Hi guys (and girls)

I have to recommend/request a way to conduct root cause analysis to a large bunch of suppliers.

In my experience, the best tool to do that is the combined use of Ishikawa/5M and then drilling with 5whys on most likely / confirmed causes.

What is the best root cause analysis tool according to you ?
(common sense being great but not an analysis tool with the same value for all users)

Would you object in making the use of a such a tool mandatory to ensure better root cause analysis ?
(so far we just ask suppliers to use the tool they want and we don't get much)

Thanks for your advice

Popeos

I think apples are MUCH better than oranges.

What tool (of any kind) is "best" depends on the application and the level of skill on the part of the user. Some people are good at problem solving and some aren't. If you have a situation where you feel that you must impose something like five-whys on people, you are saying, perhaps without realizing it, that those people probably aren't good at problem solving. On the other hand, someone who does have acute analytical skills will probably consider five-whys inane and unnecessary.

 

[Bev D]

There is some confusion of tools with methods here.

5 why is a method; a pareto diagram is a tool
Is, Is not is a method, a control chart is a tool

beyond the double dip of the Is, Is Not and Kepner Tregoe, there are missing methods.

Different types of Problems require different types of methods.

I utilize the 'is it plugged in and turned on' (my prior expereince and knowldege with the product and similar processes or products - aka 'look for the obvious') method in the initial stages until I'm sure that the Problem doesn't have a straightforward causal mechanism.

For human error based problems I use the Apollo method

For complex physics Problems I use a version of 5 why on steroids.

I have attached a presentation I gave at the ASQ Lean Six Sigma conference in 2011 that shows my basic approach...

 

[Bev D]

Would you object in making the use of a such a tool mandatory to ensure better root cause analysis ?

I have found that suppliers will not use any method you ask them to use.
I have also found that most suppliers are horrible Problem Solvers when they have a complex Problem. (Many suppliers are horrible when it comes to even basic Problems - they simply 'retrain the operator' and are done)

I have tried numerous approaches for 20+ years adn never been successful in improving the problem solving ability of my suppliers...

I always ask for specific things: containment, remediation, rework, escape cause, root cause, preventive correction action, validation etc. (along the lines of the 8D record)

too often I end up determining containment and root cause and even the corrective action for them... it's faster, easier and much more effective.

 

[Popeos]

too often I end up determining containment and root cause and even the corrective action for them... it's faster, easier and much more effective.

Sadly true.

However we have to keep faith
We might end up teaching some of them to fish instead of being fed with fish

I'll be waiting for further votes to gain from shared experience.


Popeos