Dear All,

May I post this question on How far is your team going to analyze for the root cause ?

And how do you define your Root Cause ?

Does FMEA, 5 why, Failure Analysis, Pareto, and other techniques are really meant to address the Root Cause ?

My Warm Regards,

This message has been edited. Last edited by: Rolly12,

You have got to reach the bottom in your analysis.The'Root Cause'is the initial of all the causative factors,the one which has initiated the problem.Many times,the analysts blunders by naming a probable cause as a root cause,without performing in-depth analysis.The tools you've mentioned are useful,but the widely used & a perfect tool for root case analysis is 8-D,originated by Ford.I'm enclosing this format which may solve your dillema.

8-D_System.doc (48 Kb, 169 downloads) 8-D system

Root Cause is done in 3 levels most of the analysis stops at the physical and human cause, I know all about your 8d and it is not a perfect RCFA tool.

How much time is provided to your team to complete your RCFA / RCA ?

What do you do to the people that create the fault ?

What do you do if management is involved in the fault ?

Is your Root Cause really meant to stop at the physical cause or to go beyond them ?

Rolly:

As to your original question. It quite depends on the impact of the problem.

For instance, you may use a 5-why system for looking into simpler, less expensive problems because you can expect that it may not be as rigorous as some systems. You may then select a criteria for a more 'advanced' analysis of a system.

For instance, at one of my clients, we utilize 5-why for faults and repetitive failures below a certain cost/production impact level. Once it passes that threshold, there are several self-designed RCFA processes that they use that are more rigorous.

I am attaching a version of the 5-why best practice that is being used (all names have been removed to protect the 'innocent').

Howard

Root_Cause_Analysis_Best_Practice.pdf (72 Kb, 188 downloads)

This is an article that we put out some time ago that went along with the best practice.

Howard

Root_Cause_Analysis_for_the_Rest_of_Us.pdf (33 Kb, 134 downloads)

Dear Howard,

Do you provide amnesty to those people that are involved in the cause of the problem and what do they do with the people involved directly or the culprits of the problem.

I believe that one of the rules that must be applied in a successfull RCFA / RCA initiative is that AMNESTY must be declared even before the start of any RCFA/RCA initiative.

My Warm Regards,

quote:

Originally posted by Rolly12:
...Do you provide amnesty to those people that are involved in the cause of the problem and what do they do with the people involved directly or the culprits of the problem...

If you believe the root cause boils down to what a person did, than you have not reached the REAL root cause. If you ALWAYS believe that a person is not responsible you'll be correct almost all the time.

There are many reasons for human error whether consciously or unconsciously, accidental or knowingly. John Moubray's RCM2 sums it up nicely.

MIke.

Thanks guys,

Its just quite frustrating that most RCA/RCFA techniques does not dwell with the questions I raised. I know dupont have an amnesty program in their RCA initiative. I believe that people commit mistakes and errors and most of the time the mistake committed is beyond their control. If people follow rules and procedures which are flawed then, the person will be in flawed.

I am disgusted to see in one company that part of their initiative to eliminate the error is to humiliate the person who committed the error by putting his picture and his mistake on the wall for everyone to see. These was one from the hard knox industry where I believe that these people who initiated this studid rules should be retired for good.

I have attended one of the leading Root Cause training recently and I thought that I already know what it means since I have been teaching it for a very long time.

I think that we are all part of the problem and its human nature to pick up someone (fall-guy) in order to find the culprit of the problem, this way they seem to be satified in solving the problem.

My Warm Regards,

We have no formal amnesty program, but instances of people being disciplined for errors are small and no-one would ever be humiliated for an honest mistake.

I think some of the most relevant aspects have more to do with culture than with programs or policies. World class companies have a culture that encourages everyone to learn from each other's mistakes in a forward-looking improvement-focused manner, rather than a backward-looking judgemental manner.

"World class companies have a culture that encourages everyone to learn from each other's mistakes in a forward-looking improvement-focused manner, rather than a backward-looking judgmental manner."

What a great "grouping" of words above.

The above statement, or philosophy reflects the Class we all long to belong to.

Does this reflective comment come from a personal observation, or something
that was passed down?

Care to share where this insight is/was derived?

John

[/QUOTE]

If you believe the root cause boils down to what a person did, than you have not reached the REAL root cause. If you ALWAYS believe that a person is not responsible you'll be correct almost all the time.[/QUOTE]

I have heard it said that all errors can be traced back to some form of human error and I agree with this statement. The laws of physics (or chemistry or any other science) remain constant - it is our actions or inactions that waver.

From an engineering perspective you could classify the process as:
Specificy
Design
Procure
Install
Test
Commission
Operate
Maintain
Decommission

At any of these stages errors can occur and my logic tells me they are all ultimately human error. I would be interested to hear of a problem that was not related to human error.

"Do you provide amnesty to those people that are involved in the cause of the problem and what do they do with the people involved directly or the culprits of the problem."

Great question. I haven't heard of this concept but in principle I see some merit. Of course, the person's co-workers may not agree and perhaps there are legal reasons why a person may need to be held accountable under some circumstances. But ultimately, a culture of honest reporting of what happened needs to be in place and that in my view, is not easy to create in isolation. I.e. if the general culture is a blame game one, those charged with RCA have an uphill battle.

What other approaches have folks heard on cultivating honest feedback post-failure? Has anyone experience with amnesty?

Dear All,

Just would like to share this newsletter I just recently wrote about Root Cause Analysis. Hope it shed some value to you. Kindly click on the link below :

http://www.rsareliability.com/newsletterseptember2007issue.htm

My Warm Regards,

Dr Deming taught us the importance of the system. The system includes leadership (that causes care and coordination or not), facilities, equipment, skills, knowledge, processes, documented procedures, undocumented procedures, etc...

When some goes wrong, instead of asking who we should be learning by asking why?

Blaming people for problems caused by an organization's system lets top management off the hook (to do something about the system) and creates fear. The fear stops workers from reporting problems quickly.

So, having to declare an "amnesty" at the beginning of a problem solving session highlights a bigger question of the organization's system. That question is how many other problems are hidden because of fear of retribution?

With respect,

John

John R. Broomfield
Management Systems Consultant

Which is why in this day and age most problems end up at the top eg: BP.

Mike.

We have used many tools to find out the root causes analysis such as QC-7 Tools,K-T Technique,Cause and Effect dia.,5W1H,Tripod etc.The most difficult part is data retrieval and also the chronological events because it did not has tool enough to record them as real time to show us which one come first,then we are blinding state and too much assume phenominal applied then we get difficult phase to find out the bad actors.

The changes and lack of continuous of improvement are also a common manner that make deviations and complexity of problems/failures.Especially time consuming to find the real root cause is quite to long under resources limited.

Random failures of machinery are mostly came from human errors but no one shows his mistake.Because he did not know the result of that actions.We may think about the work process instead of who did mistakes?

I think in my mind-mind that when problem creating in our life why we did solve them easily or let it be as usual.Decision making by ourselves are so fast,correct,optimise and mostly correct based upon data in our hand,skills and knoledges.Can we use us as a model for any root causes?

Best regards,
prasong D.

I have to agree with John R. Broomfield's assessment.

Having recently attend ThinkReliability's 2 day workshop, it is quite clear there are always causeS to failures. The impact to goals determines to what extent we drill down and ask why. TR delivers lots of examples and have a neat post with 115 causes for the Titanic sinking (and this isn't all the causes).

Ultimately we are looking for permanent solutions and by applying appropriate logic to the multiple causes we have many options for solutions.

If your investigation ends with 'human error' or 'procedure not followed' etc. you need to be asking more 'why' questions...Go as far as you need to go to find solutions.

Prasong,

Most industries I've been with suffer the same faith and problem which is lack of data and all I can say is that it is not an excuse for not performing a root cause analysis.

In root cause analysis, what is important is not the data but the evidence itself. Save the part that failed, and have a group of your technical people analyze them or perhaps an outside lab. Most people try to wash away evidences by mostly throwing away the failed part and replacing it with a new one.

Talk to people interview them and have a taperecorder. They will give you clues as to which event started first before the failure occured, was there a rise in temp, a strange noise a big bang etc. The part that failed and the people that you interview will be good grounds for finding evidences.

Again, root cause is not all about data gathering and clicking here and there but rather understanding why the part failed and how we all have contributed to the problem.

Hope this helps,

My Warm Regards,

Rolly et al,

Amnesty?
One should not even begin an RCA program if the intention is to name a person! That is the easy way out - if the purpose is to prevent recurrance, as it should be, then surely the punishment route is the wrong way...

Instead of asking 'Why', which often leads to blaming somebody, ask 'How did this happen'? That is much more likely to find a systemic error.

Sure, human errors are the biggest contributors (some say all of the contributors), but behind these lie a whole heap of latent issues that lead to human errors. Till we suss that out, we are just scratching the surface.