Featured Video
This Week in Quality Digest Live
Quality Insider Features
Kelly Graves
To achieve greatness, one must first think greatness
Sophia Finn
Five essential characteristics to look for
Elizabeth Gasiorowski Denis
The shifting tectonics of modern economy make a case for international standards
Dan Jacob
Fresh thinking for quality in the digital era
Innovating Service With Chip Bell
Find ways to gain real-time customer intelligence that yields insight and action

More Features

Quality Insider News
Chick-fil-A leads; Chipotle Mexican Grill stabilizes
It’s time that healthcare began to learn a long-term strategic approach to transformation
Leading scanner manufacturers involved in development of draft standard
New unit uses wireless and controller-free laser tracker technology
Drip irrigation targets the plant and not the soil
New approach uses light instead of electricity
Features faster measurement speed, onscreen averaging, and auto-rotating display
If you want to understand a system, try and change it

More News

Bob Sproull

Quality Insider

Focus on the Limiting Factor, or Improvement Processes Will Fail

Identify, exploit, and subordinate

Published: Tuesday, October 23, 2012 - 11:13

Much of my career has been spent helping manufacturing companies improve their profitability. Although most improvement experts in the world use lean Six Sigma to initiate improvements, it’s been my experience that the element of focus is missing.

By focus I simply mean where I focus my improvement efforts. In my earlier days, I truly believed that I should be improving the entire process, like trying to solve world hunger, but for the past 10–15 years, I’ve changed my approach. Yes, I’m still an avid user of lean and Six Sigma, but this different approach has made a significant difference in the speed and magnitude of my clients’ bottom-line results. The effectiveness of this approach has enabled the expansion of my client base from mostly manufacturing to service industries like hospitals and maintenance organizations.

I want to start this article by explaining my integrated methodology and then demonstrate, through a brief fictitious case study, how this method works well in a healthcare environment. As I said, I still use lean and Six Sigma to reduce waste and variation, but the difference is where it all takes place. In addition to lean and Six Sigma, I now use the theory of constraints to identify the focal point of my improvements.

The theory of constraints has been around since the mid-1980s, when Eliyahu M. Goldratt introduced this strategy in his book, The Goal: A Process of Ongoing Improvement (North River Press, 1984). Goldratt introduced his Five Focusing Steps as follows:
1. Identify the system’s constraints.
2. Decide how to exploit the system’s constraints.
3. Subordinate everything else to the above decision.
4. Elevate the system’s constraints.
5. If in the previous steps a constraint has been broken, go back to step 1, but don’t allow inertia to create a system constraint.

Although these five steps seem simple enough to understand, for many people the connection to traditional improvement methods was never made. To help people better understand this connection, I found it necessary to use simple graphics to demonstrate what these five steps mean in real life. In other words, I believed if people and companies understood what Goldratt was really saying, then they might better embrace this new process of ongoing improvement. With this in mind, I decided to demonstrate the concept of a physical constraint through the use of a piping diagram with different diameters, as seen in figure 1.

Figure 1. Piping diagram

To make this point, I asked people a very simple question: If you wanted to increase the flow of water through this system, what would you do, and why would you do it? Invariably the answer was that we’d have to increase the diameter of section C because its diameter is limiting flow. So I asked: Would increasing the diameter of any other pipe result in more throughput of water? Again, everyone answered no.


Figure 2. Four-step production process

Next, I introduced a simple four-step process used to produce something… anything, as seen in figure 2. In this process, the individual cycle times are listed at 1, 10, 3, and 2 days respectively for steps 1, 2, 3, and 4. I asked: If you wanted to increase the throughput of product through this process, what would you do and why? About half of the people stated correctly that we would have to reduce the cycle time of step 2 because it takes the longest amount of time.

Then I asked if reducing the cycle time of any other step would increase the process throughput, and all answered no, only step 2. When I asked what would happen if you ran step 1 to full capacity, most people answered correctly that excess inventory would build up in front of step 2. And to my final question: What speed should steps 1, 3, and 4 run at? Bingo, they all understood that it would only make sense to run at the same rate as step 2.

With these two simple drawings, we were able to understand the concept of “identify, exploit, and subordinate.” Once this was understood, then understanding where to focus our improvements was easy: on the system constraint. Let’s look at our hospital example.

When you think about it, being in a hospital atmosphere is really not much different than being in a manufacturing setting. The patients, for example, are what go through the various hospital process steps. The patient enters the hospital, let’s say via the emergency room (ER), and is observed by a doctor for symptoms. The doctor may have tests run to verify what the symptoms might be suggesting, and to decide on treatment and whether to treat the patient in the ER or admit the patient to the hospital for treatment. In this case study, the process is the time being taken to admit the patient into a hospital unit.

If you were to create a high-level process map of what takes place in a hospital scenario, it might look something like figure 3. This scenario isn’t much different than what you might see in a manufacturing setting, so doesn’t it make sense that those same process improvement tools and techniques might work in a hospital?
Figure 3

Let’s add some estimated times to each of the process steps and see if we can make some sense as to why this decision time seems to be taking so long. Figure 4 shows the same process we just observed, but estimated cycle times have been added. We are only interested in the time it takes to admit and not the time it takes to treat and release the patient. With this in mind and remembering the piping diagram and the simple four-step process presented in figures 1 and 2, what would you do?


Figure 4: Patient demand rate is one patient every 50 minutes

Let’s assume that the patient demand rate is such that it surpasses the current capacity of our process. If, for example, the patient demand rate was one patient every 50 minutes, then to satisfy that demand it would be necessary to reduce the processing time to at least that rate. In other words, we are interested in speeding up the process for admitting the patient to the hospital from the ER to at least one patient every 50 minutes. Which process step is the system constraint, or maybe I should ask it this way: Which process step would you focus your improvement effort on to improve throughput through the ER?

The first two steps—patient enters ER, and doctor observes patient’s symptoms—are fine. Each of those steps can handle one patient every 50 minutes. The most significant amount of time in figure 4, the step that will cause patients to back up in the waiting room, is the time waiting for the patient test results to come back so the decision as to whether the patient is treated and released or admitted into the hospital can be made.

Since we have completed Goldratt’s first step, identifying the system constraint, we now have to decide how to exploit the constraint. Here’s where we utilize lean and Six Sigma. We focus on the constraint, the time it takes for further patient testing, and look for waste and variation only there. Maybe the people performing the tests are using batches instead of one-piece flow. Or maybe the request for testing is taking too long to reach the testers. Or maybe even the time required to take a sample of blood is taking too long or maybe a combination of all of these. Each scenario is different, but when we find the sources of delays in the constraint and reduce their impact, voila, we have a shorter time required to admit the patient. It’s important to note that speeding up any other step will not reduce the time it takes to admit a patient.

In this example we see that identifying the system constraint and then focusing our improvement effort there, and only there, yields the greatest opportunity for improvement. One of the reasons that lean Six Sigma initiatives are failing is because maniacal approaches are taken to fix everything. In reality, finding and focusing improvement efforts only on the system constraint will always yield the biggest payback.


About The Author

Bob Sproull’s picture

Bob Sproull

Bob Sproull, a senior business consultant at Novaces, is a lean Six Sigma Master Black Belt and a theory of constraints Jonah. Sproull has served as a vice president of quality, engineering, and continuous improvement for two manufacturing companies and has an extensive consulting background in healthcare, manufacturing, and MRO, and focuses on teaching companies how to maximize profitability through lean, Six Sigma, and constraints management. He is the author of three books, including The Ultimate Improvement Cycle: Maximizing Profits Through the Integration of Lean, Six Sigma, and the Theory of Constraints (Productivity Press, 2009) and with co-author Bruce Nelson, Epiphanized: Integrating Theory of Constraints, Lean and Six Sigma (North River Press, 2012).


Limiting Factors

Bob, I cannot agree with you more. Many Lean/Six Sigma practitioners seem to think that because they have been trained in all these tools – they have to use them. The “KISS” principle still applies …Keep it Simple and Straightforward!”. Marshall Ariza, our North American COE and a Master Black Belt says this about “bottlenecks” .... … In Eli Goldratt’s book “The Goal”, the main character learns about the bottleneck (i.e., the slowest part of a process) by observing his son’s friends on a hike. Each boy hiked at a different speed and consequently were strung out all over the trail. From time to time, they all had to stop and wait for the stragglers to catch up. It wasn’t until the slowest boy, Herbie, was moved to the front of the line that everyone began to march at the same consistent pace. Since they didn’t have to wait and reorganize as much, they made faster overall progress and finished sooner. The learning is clear – the bottleneck controls both the speed and output of the entire process. Yet it never ceases to amaze us to see how many improvement projects are worked everywhere except the bottleneck. In fact, many organizations do not know where their bottleneck is. By quickly identified the bottleneck, projects can be strategically chosen (as opposed to being someone’s pet project). This projects by design will then logically “move the needle” on the organization’s key metrics. (see http://synergisticexcellence.com/reasons-why-lean-six-sigma-fails for full article) Ariza believes that Synergistic’s reduction of Cause Analysis to its rock – bottom basics is an essential part of the “secret sauce” that differentiates Synergistic from other Lean-based change programs. Our underlying philosophy is that project build quality plus maintenance are the two factors crucial to optimum plant performance. We use an inexpensive, yet still cutting edge computer managed maintenance system (CMMS) to each week identify and prioritise the Assets that are consuming the most maintenance hours, and the five most outstanding backlogged work orders. In the manner of Ivy Lee with Charles Schwab at Bethlehem Steel, we prioritize the assets for attention, work on the top listed asset until it is either finished or we cannot progress at that moment, and then and only then, move on to the next asset. The report on the outstanding backlogged work orders allow us to decide whether we can catch up by better utilizing the resources we have, or whether we have to bring in additional resources. There are no other choices! And if we can utilize existing resources, that indicates that planning and scheduling need work, as it is apparent that we are not maximizing use of windows of opportunity. By auctioning the top ten failing assets, we very rapidly get to the point where maintenance is no longer chasing breakdowns, but can actually come up to speed with scheduled work, which also lifts production as reliability increases. Typically we find an ROI on this process somewhere inside 12 months – Often as little as six! Best regards, Denis Backhouse

Limiting Factor

Good reading, Mr. Sproull: though the title may remind some 007's movie, very fashionable in Italy these days, I think you got it. It's years I envisage management systems as flows or streams, as a continuum, however. I'm a chemist, flows are a second nature to me: when I see Logistics to be applied to transportation and warehousing only, I judge it a big waste. I therefore welcome an innovating point of view. Thank you.