Theory Of Constraints (TOC) Blog #3: Subordinate to the Constraint

In our first two blogs:

•We provided a summary of TOC, Theory of Constraints

•Dove into Step #1: Identify the Constraints – Bottleneck analysis

•Continued into Step #2: Exploit the Constraint – Creativity over Capital

Please take a look at those blogs to catch up, now we will move onto Step 3. As a plant manager, whenever I take over a new plant, I ask my team to remember my 4 rules to manufacturing. Rule #2 is: Subordinate to the Bottleneck. This aligns directly with Step 3 of TOC, to subordinate and synchronize to the constraint.

Step 3: Subordinate / Synchronize. In this step, we have already analyzed and optimized the constraint itself. As they say “you cannot get blood out of a stone”, and we move onto step 3 with either:

1)There are no more identifiable projects to improve the constraint itself or

2)The cost to improve the constraint becomes cost prohibitive (poor ROI to justify for example).

Now that we have completed this analysis and agree to move onto step 3, we begin to look for ways to improve the constraint using strategies OUTSIDE of the bottleneck itself. There are 2 main ways that we want to focus our efforts:

Approach #1 for Step 3: Take away the excuses: When coaching the team on Approach #1, I always tell the team “Don’t allow for ANY excuse for the constraint not to run”. What does this mean? It can mean several things and here are some examples from manufacturing. Depending on your application, not all of these may be relevant but hopefully it gives you an idea where to start:

A) Ensure that there is always product ready to enter the constraint. NEVER starve the constraint. This may require OT or buffer support at times, but ensure there is always product ready for the constraint.

B) If the constraint is driven by manpower, ensure that is ever an excuse, how?

i. Can we do tag relief (run through breaks)? Shift patterns?

ii. Do we have backup personnel in case of absenteeism?

iii. Ensure training is always up to date, including backfills?

iv. Once you have backfills, do you have backfills to the backfills? I’m not kidding, if this really is stopping your business from running, being overly cautious is never a bad idea.

C) Ensure all Materials, PM, etc. are in place / completed to ensure the constraint is ready to run at all times as well.

D) Have you ever seen an obvious sign like “Coffee Hot!” and think, how can this be needed, but you know those signs are there because someone tried to chug the coffee, someone jumped over the rail at the zoo, and so on? This might be one of those but again, there is a reason for saying it. Once the constraint is fully stocked, PM’d, manned and ready to roll, please ensure there is always somewhere for this product to go. A bottleneck is RARELY always the bottleneck, there will be times when downstream there is an issue and the bottleneck gets blocked down. Prepare racks, buffer stations, etc. to ensure this product not only has enough product to run through it, but that it also always has somewhere to go.

Approach #2 for Step 3: Offload Content: Please note that both Approach #1 and #2 can be used…its not an “either / or”, but usually you can separate your teams to attach both simultaneously. In Approach #2, we look for ways to offload work. In the famous example from the book “The Goal”, the Plant Manager takes his son and a group of scouts hiking and the slowest kid had the largest and heaviest bag. They each take a piece of the load and the hike speed improves drastically. Well, we are all on a journey, or a hike, so anything we can do to lighten the load of the largest issue will help. 

Step 3: Real life example for Approach #2: At a recent client, the constraint was a welding station (not uncommon) that had multiple robotic units working in unison. Although we had optimized the robots operations in Step 2, we had not yet achieved the results we wanted. We identified that 2 of the robots had more content than the others, and therefore were the last to exit. We investigated and found that some of those welds could be reached by a downstream non-constraint station. We started moving the welds when the quality team came running down and stopped up. They stated that this “will” affect quality. “will” is a strong statement when there is no evidence, and a scary one normally, but this not being my first rodeo I replied that we would run sample parts and if they could show a mean shift in quality in a 5 / 25 / 125 piece sample run, we would find another way. We made the change and I did follow up with the quality team as promised, but there was no evidence of this proclaimed quality shift. This successfully shifted the bottleneck to another area and improved overall performance above our target goal.

Remember, there will always be a bottleneck! Now, onto Blog #4!!!