Waste Occurs When You Expend Resources & Don't Improve Throughput

In my last article, I looked at the Toyota Production System (TPS) from an article in the Toyota Times. I took a simple look at the concept of waste and what it meant after WWII. Whether you agree with this simple logic is less important than thinking about it. In that chaotic time, the definition of waste could be:

“Don’t waste my money when I could reduce the price of a Corella.”

Then I looked at it from a Theory of Constraints (TOC) perspective and put forth this statement:

"Waste occurs when you expend resources that don’t move you toward your Goal or improve your Necessary Conditions".

Money, investment, materials, and time are the basics that float to the front of the mind. One thing that is always true is that Profit is either a Necessary Condition or a Goal. Most companies only make Cost Reduction decisions, not Profit decisions. In TOC, you must make Profit Decisions and know the Profit formula.

The obstacles that prevent you from moving towards your goal unrelated to profits are as varied as the different goals for different people. The measures related to the movement towards those goals may often be less measurable than just what’s in the cash register. Often retail companies can look at star ratings or customer feedback to understand what constraints prevent them from reaching their full potential from reaching their goal.

I developed the Throughput Improvement Process (TIPs based on TOC. Waste for that process could be defined as:

“Waste occurs when you expend resources that didn’t improve Throughput.”

Focus on the bottleneck is the key concept for improving throughput in manufacturing, so Waste here is any action that doesn’t help the throughput of the bottleneck. The bottleneck that is used in the Throughput Improvement Process exists in one of four production states:

1. Running
2. Down
3. Starved
4. Blocked

These states can be represented in time or rate, such as losing 7 parts per hour due to downtime. The chart at the beginning of this article demonstrates a ten-station line with each one of these states measured in JPH. The bottleneck is the machine with the lowest stand-alone throughput, a function of downtime and running time, as in the chart below.

One aspect of the bottleneck that is often unclear is that it has the highest OEE in the line being analyzed, so it will spend most of its time either running or down.

There is little reason to be more detailed than these states since you’ll have to go to bottleneck to see what’s happening. The only improvements that can be made on the bottleneck can be to reduce the cycle time (Running Time) or the Downtime. A quality reject is typically considered a downtime event.

The third state of concern is Starved Time. This means the bottleneck could be running, but no parts are available to load into the workstation. If the problem on the bottleneck can not be resolved quickly, then steps must be taken to remove the starved state from the bottleneck. Some stop gaps may be adding a buffer before the workstation to absorb variation that is caused by the interaction between stations. This is a basic difference between TOC and other CI methods since:

• Starved and Blocked at a workstation are not considered
• Variation, which always exists, is ignored
• Dependency between workstations, a basic concept in systems, is ignored
• Buffers to absorb variation are poorly understood

The last state, Blocked, typically only occurs where Buffers are limited in size. Assembly lines that use Power and Free conveyors, belt conveyors, etc., often fall into this state. Much like the Starved State, buffers may help, or some improvements to pull products away from the bottleneck will help. Buffers, however, are a good idea if this occurs.

Beginning workstations may rarely face a condition when the line runs out of baskets or containers because they are sitting in front of the bottleneck. Being starved for containers really means you are “Blocked.” Adding any more containers just means you will add more full containers in front of the bottleneck. Essentially, this workstation is blocked to the bottleneck. Improving this station will make no difference and is the definition needs to be clarified

This is a confusing point in OEE measures since a station like this will show up as a low OEE, and the bottleneck will have a high OEE. Improving this station will cause its OEE to drop again without impacting Throughput. To finetune Waste in the Throughput Improvement Process.

“Waste occurs when you expend resources that didn’t improve Throughput.”

Kevin Kohls is the world's leading authority on how to apply the Theory of Constraints to the auto industry and manufacturing. His latest book is "Addicted to Hopium" and is available on Amazon.