Theory of Constraints

How does the Theory of Constraints work with Lean and Six Sigma?

I was recently called by a good friend who told me that he had transitioned into the operational side of his organization as a plant manager. He found out that I was on a Lean and Six Sigma Implementation Project with a financial organization in his area. We had a long dinner together where I learned that his latest mandate was to transform his plant as the benchmark for all other production facilities to emulate.

During lunch, he asked if I would help him and his team with the efforts. He talked in length about what was going on in his facility, and I listened intently. He explained that his plant’s Continuous Improvement teams were doing targeted projects. He said that the projects seem to be successful but their throughput was not increasing. They were not getting any faster although many of their problem machines and processes had been fixed.

Theory of Constraints
At the end of his story, I asked him if he had heard of TLS. I explained that TLS was the integration of Lean, Six Sigma and the Theory of Constraints. I further explained that most companies use Lean and/or Six Sigma to target and execute improvements without really understanding where to truly focus. This results in improvements to processes in the system without improving the speed of the system as a whole.

We talked at length about the power of combining these three powerful methodologies. Our dinner started around 4:30 pm and ended around nine. By the end of our discussion, my friend was excited to get started.

This company, like many other companies, uses a KPI (Key Performance Indicator) that only shows them what their cumulative efforts of the day produced. In this case it was “Yield”: what they initially put into the system and what was produced out of the end. This metric does not give you a systemic view of why you do not perform as expected.

All the steps in between the beginning and the end are hidden. So, like many companies, this one adjusts the system when things do not go as expected, but their adjustments are based on a SWAG (Scientific Wild A!! Guess), not quantifiable data. This had a negative effect on their daily yield.

In order to implement TLS, we need to have a better way to look at our process systemically. We need to have a way of determining where the constraints are located in the system. Following the “Lean” methodology, we decided to use O.E.E. (Overall Equipment Effectiveness) as a primary metric. We had an overall “Plant” OEE, and we had OEE’s per line and per product. OEE would give us visibility to how we were performing with respect to Performance (Ratio of Daily production to effective capacity), Quality (the number of Good vs. Bad product), and Availability (ratio of used production time to available production time).

We also needed to take data on each of the critical steps in each line to understand where our constraint or bottleneck was located in that line. We defined six critical points that were common to each line (because the lines were almost replicates of each other). We defined each critical point in the line in a metric and described throughput (with a common denominator) at each point. Then it was time to track the throughput on a continual basis. We did this for 60 days in order to get a good representative sample.

DropperThis data showed us where the constraint was in each line. The constraint is defined as the “the slowest process in the system.” A majority of companies don’t really understand the concept of constraints. These companies target their Lean and Six Sigma improvements on the most recent “pain.” If the output of the improvement is to increase speed and you are focused on anything other than the constraint, your net gain in speed will be zero. We can never go faster than the constraint.

We discovered that in some lines, the constraint was equal to or better than that line’s demand. These lines were performing well. We found two lines where there was a steady constraint (the constraint did not move) and both lines were only performing around 70% of their demand. On both of these lines, there were multiple Lean and Six Sigma Projects completed but none of the projects were focused on the constraints. The TAKT time of the two lines was set to demand. This resulted in a bottleneck at the constraint in those two lines where inventory accumulated.

Once we understood what was going on, the next step was to set the TAKT time of the two lines to the constraint. Several projects were focused on the two constraints. In the next 3 ½ weeks the facility reached record levels of throughput. By relieving and managing their constraints, they finally felt the results of their other Lean and Six Sigma projects.

Have you seen your organization focusing their Continuous Improvement efforts on processes within a system with the goal of speeding up the system only to fail because they were not focused on the constraint?