Lean Six Sigma

Getting Good by Becoming a Little Less Ungood Each Day

Define, Measure, Analyze, Improve, Control ... Six Sigma problem solving for continuous improvement.

The Lean Six Sigma methodology of problem solving is a systematic, team-oriented approach to solving the various problems in manufacturing processes, with the aim of the elimination of the eight wastes (muda) in the DOWNTIME acronymn:

Defects
Over-Production
Waiting
Non-Utilized Talent
Transportation
Inventory
Motion
Excess Processing

A reduction in any of the 8 wastes carries the potential for increased safety, quality, and productivity. So where do we begin?

The use of accurate data-keeping methods in any process is absolutely critical to the Six Sigma problem solving process as it relies heavily on statistics in determining the top areas of opportunity. In other words, without good data, we don't know what our greatest waste is, and we don't know what our greatest opportunity for improvement is. A good data system is the foundation we build our projects on and the way we can track and measure our results. A proper data system automatically generates charts, graphs, and statistics including top causes of waste for the facility, department, and workstation. Once such a system is in place, we simply look to the data to show us our next area of focus. It is important that we use this data, rather than feeling, to guide our problem solving efforts.

Using our data system to guide us, we can now use the DMAIC problem solving process to reduce, and hopefully eliminate waste. The DMAIC process is defined below:

Define: As the quote, often (falsely) attributed to Albert Einstein, goes, "If I had only one hour to save the world, I would spend fifty-five minutes defining the problem, and only five minutes finding the solution." This is the most critical part of the process, and without a proper problem definition, our efforts from this point forward could be directed incorrectly. We need to understand what the problem is,  where our opportunity for improvement lies, how we are falling short of customer demands, and what our goals are. 

Measure: Once we have successfully defined the problem, we can begin measuring the process and the shortfall. We should look to historical data to establish baselines and trends.

Analyze: This is where we determine the root causes of variation and defects in the process. During this step there is no substitute for going and seeing the problem firsthand (gemba). To determine root causes, it can be helpful to see the problem at the end of the process and trace it backwards in the process until the problem is no longer apparent, locating the point of origin. Often, studying the point of origin will lead to the discovery of a root cause.

Improve: To improve our current state we need to create and implement countermeasures. It is critical that our countermeasures address a root cause, otherwise we risk using  work-arrounds which only complicate the problem.

Control: In order to see lasting results from our problem-solving efforts, we must sustain our process design improvements, by controlling the process and it's performance.

Though this system of problem-solving is usually reserved for manufacturing, I've found the skills I've learned in implementing these techniques useful in other instances, including data-systems and programming. After all, programming is essentially logical problem solving. Furthermore, the more problems I solve as a programmer, the better I am at creating data collection systems, and the better data collection systems my team and I have at our disposal, the better we are out solving real-world manufacturing problems.

---

Add Comment

Comments:

C. WILLIAMS Dec. 14, 2017, 12:05 p.m. Dude, this is sweet!

EMontoy Dec. 14, 2017, 2:35 p.m. You need a logo!

J Perez Dec. 14, 2017, 8:23 p.m. I’m with Eric. A logo and graphics would make it more apealing. Nice job.

Tabre Dec. 14, 2017, 8:41 p.m. Thanks for the feedback, guys. I'm not much of a logo designer but I'm sure I can come up with something. :)

Black Belt Aug. 13, 2018, 8:04 a.m. Six Sigma is awesome!