A Conversation with Forrest Breyfogle

Six Sigma First: A lot of methodologies aimed at improving quality management have been introduced in recent years. We’ve had Companywide Quality Control, Total Quality Management; now we have Six Sigma. What is Six Sigma, first of all, and what makes it so special?

Forrest W. Breyfogle: Let’s reflect on the initiation of Six Sigma and how it has evolved. In the eighties, Motorola had quality problems. To address these issues, they invented Six Sigma and deployed statistical and non-statistical tool-based training through Motorola University. In this training, employees learned the application of quality-improvement tools. Beyond classroom instruction, experienced practitioners mentored students so that they became more proficient at applying the taught quality-concepts. Gains were made at Motorola through the application of these quality tools; however, these newly-trained practitioners were not formally assigned projects that would have quantified financial benefits validated by Finance.

In the mid-nineties, Jack Welch as CEO became interested in, and then implemented, Six Sigma at GE, GE’s Six Sigma deployment was different from Motorola’s. In the GE Six Sigma deployment, trainees underwent extensive, concentrated training in many statistical and non-statistical tools, and the trainees immediately applied the taught tools to assigned projects. These in-class projects were chosen from perceived business importance to bottom-line improvements. Upon satisfactory completion of one or more finance-validated projects, students received Black Belt or Green Belt certification. Within this GE Six Sigma deployment system, Black Belt training was typically 4-weeks over 4 months, where the trained person was a dedicated resource. Green Belt training was typically 2-weeks over 2 months and the trained person was a non-dedicated person who worked on an assigned project in his/her immediate area. 

After training and their first project completion, both Black Belts and Green Belts were to continue working on projects in either their full-time dedicated role or part-time role. In this project-driven system, project execution involved both statistical and non-statistical tools. This GE-deployment system included a support infrastructure where Executives, Champions, Master Black Belts, and others supported project execution, which was primarily accomplished through the Black Belt and Green Belt practitioners working with teams. Within GE, the compensation of executives could even depend upon the financial-validated project benefits. TQM and other past quality and non-quality systems did not have this level of support infrastructure and belt-certifications.

Lean is a system for reducing waste. Eli Whitney’s development of interchangeable parts and Ford’s production-line system were precursors to perhaps the current most benchmarked system for Lean, the Toyota Production System (TPS). However, organizational Lean deployments are typically not close to becoming a TPS.

In the late 1990’s, Lean was integrated with Six Sigma and came to be known as Lean Six Sigma. Many viewed Six Sigma as the methodology of providing quality improvement tools, while Lean provided waste-reduction tools to the practitioner’s tool set. The impetus for Six Sigma and Lean Six Sigma has been to complete financially-beneficial projects. After some period of time, organizations that deploy this Lean Six Sigma system often find that they are hunting for projects so that they can claim financial benefits, even though process owners are not excited about or even encouraging project completion. To illustrate how crazy this can get, we had one trained practitioner who later moved on to another company that used another Lean Six Sigma provider for their deployment. This person later came back to me to say, “Forrest, this company is claiming to have saved a hundred million dollars, but no one can seem to find it”.

Even though the Finance department is to validate Lean Six Sigma projects, often accounting systems do not reflect whether the project is a constraint to the overall enterprise. Many completed project benefits sound good at a localized metric level, but simply do not impact the enterprise metrics as a whole. In fact, these projects could even be detrimental to the business as a whole.  For example, a localized targeted project that improves machine efficiency could yield excess inventory, which would need to be carried by the overall enterprise.

Processes have execution steps.  Processes can have one or more outputs, which can be referenced as “Ys.” Processes also have inputs, which can be referenced as “Xs.”  A process’ output can be significantly impacted by its inputs’ levels. The relationship between the input and an output of a process can be described as Y=f(x); i.e., Y is a function of X.

To establish accountability, organizations often create organizational scorecards with measurement goals.  Often these metrics are established through an organization chart structure so that every organization has assigned metrics and goals.  People and organizations are then tracked as a variance (not to be confused with the square of standard deviation) against scorecard goals.

When an organization establishes subjective goals and then manages against these goals, it is managing to the Ys of the organization.  This approach to management can lead to what I call the Enron problem. At Enron, financial goals were set for each quarter.  At Enron, the message was conveyed that these goals had to be met. Simply managing through the Ys of the processes can lead to the wrong behaviors, and that’s what happened to Enron.

I believe that Enron perhaps started small with a simple legal financial adjustment to meet a quarterly goal.  This simple shifting of funds to meet quarterly goals grew over time to become a major problem that led to Enron’s downfall. So my question is: “Did all the crooked MBAs migrate to Enron or did Enron create the crooked MBAs?” I suggest that Enron’s score cards led to the wrong activities.

A lot of companies are experiencing similar problems to Enron but to a lesser extent. For example, Krispy Krème was shipping donuts that they knew would be returned so that they would meet their short-term financial objectives. This is another example of how people are driven to the wrong activity when they are simply focusing on organizational Ys with no real emphasis on changing the Xs that truly lead to long-lasting Y improvements.

Goals are not bad; however, organizations need to have a system that encourages the improvement of the Xs that in time will improve the Ys of the organization.  A system to accomplish this is Integrated Enterprise Excellence or IEE for short. This system takes Lean Six Sigma and business scorecards to the next level.

Six Sigma First: What makes your approach to Six Sigma different from Six Sigma as practiced in the companies that you described?
Forrest W. Breyfogle: Jack Welch’s GE-Six-Sigma system accomplished some wonderful things; however, the system pushes for the creation of projects. In other words, this traditional-Six-Sigma system focuses on “Let’s go and decide which projects we should work on.”  Why might we want to do this? So that we can pound our chest in victory style when our collective financial Six Sigma project benefits look good.

IEE is more of a business measurement and improvement system than Lean Six Sigma, which, again, focuses on the completion of projects. IEE involves stepping back and strategizing using enterprise-level data to determine where focus efforts would have the most benefits. With this approach, we start assessing what we do at the enterprise value chain level, independent of the organizational chart. We then select a project and determine how we are going to measure activities over time so that we do not react to common-cause variability as though it were special cause. IEE metrics are not bound by calendar quarters or years. With IEE, we are looking at the output of the enterprise as the result of a collection of processes that experience common and special cause variability, where common-cause variability is the result of typical differences between the raw material lots, people, days of the week, work-day shifts, etc.

IEE is more than simple project execution.  IEE strives to have everybody throughout the organization doing the right thing correctly at all times for the common benefit of the overall enterprise. With IEE, an organization becomes a learning organization.

In the book Good to Great, Jim Collins describes a Level Five Leader, where this leader is one that not only leads his/her company to greatness while employed there, but the company remains great using the systems that he/she created even after that leader has left the company. IEE can help organizations create a Level Five Organization that has long-lasting greatness, which is not solely dependent upon on the leadership skills of a few people at the highest organizational levels.

The primary force driving many organizations is their strategic plan. Let’s step back and ask ourselves, “How many of the current strategic objectives are dependent upon who is leading the organization?”  Do we anticipate that there could be a significant company redirection if the senior leadership team changed? If so, the current enterprise system should not be considered one that has the potential of becoming Level Five.

A Level Five enterprise system is one that is robust to the idiosyncrasies of the people in charge. For this to be the case, we need to have a system for creating the strategic plan which lessens dependency on who happens to be leading the organization at any point in time.

Level Five enterprises systems are obtained within IEE by first examining the enterprise goals as a whole. For example, a goal might be to increase monthly profit margins by 2% along with a 5% increase in total revenues in one year.  To accomplish this goal, we would then analyze the organization using the analytical tools of Six Sigma to create a plan for accomplishing these objectives, where focus is given to improving areas of the business that have the most impact on achieving the overall goals.  From this analysis, very specific objectives such as monthly defective rate reduction by 20% and improving sales/marketing metrics by 10% within one year can be created. These metric improvements would be tracked at the 30,000-foot-level, where metric improvements would have specific value-chain ownership and claimed improvements are demonstrated statistically.   

With this approach, we have created an IEE system where improvement and design projects are pulled for creation.  This approach is quite different from traditional Six Sigma deployments where projects are pushed for creation, independent of analyzing the overall enterprise as a whole. If you save a hundred million dollars that no one can find, did you really help the profit margin of the company as a whole?

Six Sigma First: A lot of companies have tried Six Sigma and it did not work. Would you say that it is a methodology that can be applied in any company?
Forrest W. Breyfogle: The answer is “yes,” not only for profit and non-profit companies, but also for government and educational institutions.  The next question is “How is this most effectively accomplished?”

In an August, 2005 ASQ Quality Forum Magazine article, I describe 21 common problems with Six Sigma. One of these problems is the hunt for projects, which I previously discussed. Another problem is that some companies try to start out too fast. It is better to start with a small group and build a solid infrastructure that builds from the enterprise value chain, its metrics, and an enterprise analysis for project selection.  

There are numerous ways of failing to achieve the potential that one might desire through a Lean Six Sigma system. Currently, it seems that everyone wants to offer a Black Belt certificate, including universities. Some of these offerings can be good; however, others can be diluted in the tool sets and methodologies presented during the training. Talk about variability; there is a lot of variability between the quality and tools presented in various training offerings. Often organizations become pennywise and pound-foolish when they select their training provider.  The real cost is people’s time.  Why not get the best you can for people’s training and application time?

Often companies think that it will be a lot cheaper to simply hire a Black Belt or a Master Black Belt. When this is done, often they are disappointed. How can organizations determine in the hiring process that they are bringing on-board someone who will fit well in their organization? How will these new-hires know the organization and people enough to get things done quickly?  Once a character profile for a Black Belt or Master Black belt is described, it is much easier to identify these key employees and then quickly bring them up to speed with the tool set.

 
Six Sigma First: Can you explain to us what your “Satellite-Level” and “30,000-foot-level” metrics are?
Forrest W. Breyfogle: Satellite-level and 30,000-foot-level are IEE metrics terms, where satellite-level metrics are business metrics and 30,000-foot-level metrics are high-level operational and project metrics. A satellite-level metric could describe a company’s profit margin over time independent of calendar boundaries and goals, while a 30,000-foot-level metric could describe the company’s customer delivery timeliness or defective rates. 

Both satellite-level and 30,000-foot-level metrics are assessed using an IEE control chart that utilizes infrequent sub-grouping/sampling so that typical differences between operators, machines, hours-of the days, and shifts are considered sources of common cause variability. If a process is considered predictable, a no-nonsense, easy-to-understand process capability/performance metric is then made. Projects then are pulled for creation if the metric predictions are undesirable relative to what is needed for the organization to achieve its goals. 

An example of a 30,000-foot-level metric is the time a customer waits in the checkout line at a grocery store. As a customer, you do not want to wait a long time no matter when you go to the store.  The grocery store prefers that you have to wait some time since, if you did not wait any time that would indicate that there were excess checkers, which costs the store money.  In addition, when you spend time in the checkout line, you might buy some additional displayed merchandise.

Most grocery checkout demands are quite dependent upon time-of-the-day and day-of-the-week. The grocery store needs to manage this fluctuating demand so that the duration of time for checkout is reasonable.  Within IEE, the grocery store can do this through 50-foot-level metrics. A 50-foot-level metric such as the average number of people in the checkout line requires frequent assessment, where process alterations such as adding or removing a checker from duty are made in a timely fashion so that the 30,000-foot-level metric of overall checkout time remains satisfactory. 

Copyright 1992-2006.
All rights reserved
Smarter Solutions, Inc.

www.sixsigmafirst.com 

Tell us what you think about this article. Send a note to the Editor.