I’ve learned recently that about any search term that includes “Six Sigma” does well. Regular readers know that I haven’t talked much about Six Sigma over the many months I’ve been involved with this blog. I’m not a Six Sigma “any-kind-of-belt” but I do know a bit about the statistical tools that Six Sigma advocates. (Design of Experiments is a bit above my pay grade but I’m betting the number of actual Design of Experiments projects run in any given year by manufacturers is relatively low.) All to say, it’s not because I don’t like Six Sigma that I don’t talk about it much. It’s more that Six Sigma and Lean and their relationship are misunderstood by most managers and I’ve focused on trying to clarify the “lean” part of it.
Here’s the thing…Lean and Six Sigma have a common goal: the identification and subsequent reduction or elimination of process variability. Let me pause here and say a few words about this definition. If you websearch “Lean vs Six Sigma”, you’ll come across a number of descriptions that don’t get this part right. One comparison that I just found gets it all wrong: it posits that the focus of Lean is “Cost Reduction”, while the focus of Six Sigma is “Quality and Culture”. The focus of Lean is not “Cost Reduction”. Further, try telling any experienced practitioner of Lean that “Quality and Culture” are not its primary focus and see what kind of response you get. Other comparisons will try to tell us that Lean focuses on removing waste, while Six Sigma focuses on reducing variation. (Aren’t waste and variation pretty much the same thing?) I’ve also seen comparisons that attribute enhancements in speed to Lean Tools and improvements to Accuracy/Precision to Six Sigma. All these definitions are working too hard to come up with some sort of dichotomy of the two approaches….”Lean does this, but Six Sigma does that.” It’s better think of both approaches as having the same goal.
The difference is in the methods they use to get there. The Lean tools are more straightforward, therefore, more useful and relevant, at least initially. The Six Sigma tools are more sophisticated (if that’s the right word). They comprise a variety of statistical tools that dig deep into the sources of variability. Something called Design of Experiments is a resident of the Six Sigma toolbox.
Let me give an example that, I hope, will illustrate the difference between the two approaches. Let’s imagine a manufacturer of PVC pipe wants to implement Lean and Six Sigma tools. How does the company decide what tools to apply to what conditions and situations?
Well, let’s start with the most obvious, most likely sources of variation. In my experience, those are:
- Disorganization of the workplace
- Lack of standardization of leader and work practices
- Poor flow of materials and information
- Poor communications about daily operations
- Lack of engagement of operators, leads, supervisors in problem solving and decision making.
These issues are most effectively addressed with traditional Lean tools like:
- 5S
- Visual Factory
- Value Stream Mapping
- Operator Work Instructions and Leader Standard Work
- Team Kaizens
The same company has issues with scrap. In this case, the structured Six Sigma approach to identifying and analyzing causes and the accompanying analytical tools are going to be useful.
- Pareto Chart
- Histograms
- Control Charts
- Trend Charts
- Design of Experiments (or DOE, which gets very sophisticated. It’s a way of nailing down root causes when there are a variety of sources of variation and the “weight” of each source isn’t well known. For example, DOE could be used to determine the best mix of heat, production rate, and PVC “recipe” under varying conditions of humidity.)
Can the Lean and Six Sigma tools be implemented simultaneously or should an organization use one set of tools before the other? They can be implemented together, of course, but my recommendation is that more straightforward tools be used to attack the more visible sources of variation and that means Lean Tools first. Don’t get me wrong, by “straightforward”, I don’t mean “easier to implement”. I’ve seen too many companies do poorly at the “get this mess organized” stage to believe that. It’s just that an organization needs to go after the variability that’s kicking its butt day in and out before it goes after the more challenging varieties. Going back to our PVC scrap example above, there’s little value in trying to get just the right mix of “recipe”, temperature, and rate to produce the best product before you’ve implemented good operator practices and maintenance procedures.