Monday, November 2, 2009

Six Sigma-Its no substitute for knowledge

I recently was having a conversation with a senior manager who was lamenting that "six sigma analysis" did not find the root cause of a longstanding problem but our scientific knowledge did. I think this may have been meant as a dig at me since I'm a Master Black Belt but I think he makes a great point, slightly misguided but still a great point. Six Sigma, or any other problem solving process is only as good as the information analyzed. We've all heard the adage "Garbage In-Garbage Out". I've thought alot recently about the perspective of this senior manager and it reminds me of something that I think is worth reminding ourselves from time to time. Six Sigma is no substitute for knowledge. In the problem I was discussing with this senior manager, no routine data was collected on the process step that produced the problem.

Some people seem to think that Six Sigma is an excellent method for discovery of new knowledge. I understand why this misconception occurs. One of the strongest tools in the Black Belt's toolbox is Multivariate Regression. This toolset is used to uncover hidden relationships in data and thereby help uncover the root causes of problems that can't be solved in other ways. This, I believe, leads people to assume that Six Sigma properly applied can find new knowledge, and it can, but only when the data is present. The strength of six sigma is in the disciplined, logical problem solving approach that is inherent within the DMAIC method. But, Six Sigma will not find the root cause of a problem if there is no data on that problem or that particular root cause. The DMAIC method teaches us to follow the data, and trust the statistics. We still have to use our brains though and when the data and statistics do not present an answer to the problem, we have to expand our circle of thought to include other information that didn't get included in the first go around.

In all of the six sigma classes that I have taught and all of the mentoring I have done with project leaders I have emphasized that six sigma is not a substitute for your brain. Use the DMAIC method, but follow the data and trust the statistics and if they tell you that you have not found root cause, go look somewhere else for the answer. One of the best places to go is to the guru. Sometimes, scientific experimentation has already produced the answer to our problem but we may not have ready access to that knowledge because it resides in someones head, rather than in a library somewhere.

3 comments:

Brad said...

Hey Jim,

Great post! I see this problem all the time as well. Leaders expect answer to pop out of the tools and processes without them having to provide any insight.

Your example of multivariate regression is a good one. You can super crunch giant data sets to find relationships, but someone with business knowledge needs to figure out if those relationships make sense and if they are actionable.

I often tell leaders that the data itself is not the story. The story is what is happening in the business, the data simply supports it.

Thanks for the post!

Unknown said...

It's Very Informative Blog... Thanks for Posting...
https://goo.gl/1uvc7H

Unknown said...


I just want to say thank you for sharing this post, it was really awesome and very informative. Thank you.
If somebody looking for Six Sigma training certification in India. Join us