Despite the ubiquity of corporate Six Sigma programs and the intensity of their promotion, it is not uncommon for graduates to enter industry with little exposure and less understanding of their administration or purpose. Universities that offer Six Sigma instruction often do so as a separate certificate, unintegrated with any degree program. Students are often unaware of the availability or the value of such a certificate.
Upon entering industry, the tutelage of an invested and effective mentor is far from guaranteed. This can curtail entry-level employees’ ability to contribute to company objectives, or even to understand the conversations taking place around them. Without a structured introduction, these employees may struggle to succeed in their new workplace, while responsibility for failure is misplaced.
This installment of “The War on Error” aims to provide an introduction sufficient to facilitate entry into a Six Sigma environment. May it also serve as a refresher for those seeking reentry after a career change or hiatus.
Six Sigma Defined
“Six Sigma” became a term of art in the 1980s when Motorola used the statistical concept as the foundation for a company-wide quality goal and problem-solving strategy. Popularity of the methodology began to skyrocket in the 1990s, thanks in large part to General Electric’s very public adoption and energetic promotion. Widespread adoption followed, leading to Six Sigma becoming the de facto standard for quality management. It became an actual standard in 2011, with the International Organization for Standardization’s (ISO) release of “ISO 13053: Quantitative Methods in Process Improvement – Six Sigma.”
The foundation of a Six Sigma program, and the origin of its name, is process performance monitoring. Process output is assumed to be normally distributed with the mean (μ, mu) centered between the specification limits. The distance from the mean to either specification limit is given in standard deviations (σ, sigma), a measure of process variation. The goal is to manage processes such that this distance is (at least) six standard deviations, or 6σ, as shown in Exhibit 1. The total width of the distribution between specification limits (USL – LSL) is then 12σ, resulting in approximately two defects per billion opportunities.
Understanding that no process is perfectly stable with its output mean centered between its specification limits, the developers of the Six Sigma methodology accounted for the difference between short- and long-term performance. Empirical evidence indicated that process means could be expected to shift up to 1.5σ from center when long-term operational data is analyzed. This results in a 4.5σ distance from the mean to one of the specification limits and a higher reject rate. The shift essentially eliminates defects from the further specification limit, yielding an anticipated reject rate one-half that of a “4.5σ process,” or 3.4 defects per million opportunities (DPMO). This is the target reject rate quoted by Six Sigma practitioners. Exhibit 2 presents the mean shift and resultant reject rate graphically.
The true goal of any process manager is to achieve zero defects, however unrealistic this may be. Six Sigma process control seeks to come as close to this target as is economically and technologically feasible. It engenders vastly more aggressive objectives than “traditional” process control that typically employs μ ± 3σ specification limits.
The term “Six Sigma” is an umbrella covering two key methodologies, each with a unique application and purpose. The DMAIC (“duh-MAY-ik”) methodology is used for process improvement; the most frequently employed methodology, users often use “Six Sigma” and “DMAIC” interchangeably. The DMADV (“duh-MAD-vee”) methodology is used for product or process design. A description of each, including the five-step process that forms its acronym and some example tools used during each, follows.
DMAIC – for Process Improvement
Existing processes that underperform can be improved using the most common Six Sigma methodology, DMAIC. The acronym is derived from the five steps that comprise this problem-solving process: Define, Measure, Analyze, Improve, and Control. Exhibit 3 presents a basic flowchart of the DMAIC process. Note that the first three steps are iterative; measurement and analysis may reveal the need for redefinition of the problem situation.
For brevity, each phase of DMAIC is presented in a summary table. The top cell of each table contains a description of the phase activities and purpose. The lower left cell contains typical outputs, or deliverables, of that phase. These items will be reviewed during a “phase-gate” or similar style review and must be approved to obtain authorization to proceed to the next phase. The lower right cell lists examples of tools commonly used during this phase. Full descriptions of the tools will not be provided, however. Readers should consult the references cited in this post, or other sources, for detailed information on the use of the tools mentioned, as well as other available tools.
As shown in Exhibit 3, a review should be conducted at this point in the process to verify that the problem definition remains accurate and sufficient. If adjustments are needed, return to the Define phase to restate the problem situation. Modify, append, or repeat the Measure phase, as necessary, and analyze any new data collected. Repeat this cycle until measurement and analysis support the problem definition.
Though lessons-learned activity and replication are focused on processes other than that which was the subject of the Six Sigma project, they are included in the Control phase discussion for two key reasons:
1) They are vitally important to maximizing the return on investment by limiting the amount of redundant work required for additional processes to capitalize on the knowledge gained during the project.
2) They take place at the conclusion of a project; therefore, the Control phase discussion is the most appropriate to append with these activities.
Some descriptions of the DMAIC methodology will include lessons learned and replication as additional steps or follow-up; others make no mention of these valuable activities. To minimize confusion and encourage standardization of best practices, they are considered elements of the Control phase for purposes of our discussion.
DMADV – for Product or Process Design
It will be evident in the following discussion that there are many parallels between DMAIC and DMADV. The five steps that comprise DMADV are Define, Measure, Analyze, Design, and Validate. The first three steps, it may be noted, have the same names as those in DMAIC, but their execution differs because each process has its own purpose and objectives. Overall execution of DMADV, however, closely parallels DMAIC, as can be seen by comparing the flowchart of DMADV, presented in Exhibit 4, with that of DMAIC in Exhibit 3.
The fundamental difference between DMAIC and DMADV is that DMADV is proactive while DMAIC is reactive. Another way to think of this distinction is that DMAIC is concerned with problems, while DMADV is focused on opportunities. Though other “acronymed” approaches to proactive analysis exist, DMADV is the predominant methodology. For this reason, it is frequently used interchangeably with the umbrella term Design for Six Sigma (DFSS), as will be done here [DMADV doesn’t roll off the tongue quite so eloquently as DMAIC or DFSS (“dee-eff-ess-ess”)].
The phases of DMADV are presented in summary tables below. Like the DMAIC summaries, the lists are not exhaustive; additional information can be found in the references cited or other sources.
Review results of analyses with respect to the opportunity definition. If revisions are needed, return to the Define phase and iterate as necessary.
Though an organization’s efforts are most effective when the inclination is toward proactive behavior, or preventive measures, DFSS is in much less common use than DMAIC. The lingering bias toward reactive solutions is reflected in the greater quantity and quality of resources discussing DMAIC; DFSS is often treated as an afterthought, if it is mentioned at all. This provides a significant opportunity for any organization willing to expend the effort to execute a more thorough development process prior to launch. A proactive organization can ramp up and innovate, satisfying customers’ evolving demands, while reactive competitors struggle with problems that were designed into their operations.
Belts and Other Roles
Perhaps the most visible aspect of Six Sigma programs is the use of a martial arts-inspired “belt” system. Each color of belt is intended to signify a corresponding level of expertise in the use of Six Sigma tools for process improvement. The four main belts in a Six Sigma program are Yellow, Green, Black, and Master Black. Other colors are sometimes referenced, but their significance is not universally accepted; therefore, they are excluded from this discussion. Responsibilities of the belted and other important roles are described below.
The belts in an organization are directly responsible for executing Six Sigma improvement projects. To be successful, they need the support of other essential roles in the program. These are described below.
There are three important caveats offered here. The first is common in many contexts – launching a Six Sigma program does not ensure success. Put another way, desire does not guarantee ability. A successful program requires the development of various disparate skills. “Expert-level” skills in statistical analysis, for example, provides no indication of the ability to develop and implement creative and innovative solutions.
Second, achieving six sigma performance has the potential to be a Pyrrhic victory. That is, a misguided effort can be worse than no effort at all. Analysis failures that lead to poorly-chosen objectives can divert resources from the most useful projects, causing financial performance to continue to decline while reports indicate improving process performance. Many organizations have abandoned their Six Sigma programs as administration costs exceed the gains achieved.
The third caveat is the “opposite side of the coin” from the first. Any individual interested in improving process performance or product design need not delay for lack of a “belt.” Certification does not guarantee success (caveat #1) and lack of certification does not suggest imminent failure. This introductory post, other installments of “The Third Degree,” past and future, and various other resources can guide your improvement efforts and development journey. No specialized, status-signaling attire is required.
This installment of “The War on Error” series was written with two basic goals:
1) provide an introduction that will allow those without experience or formal training to understand and participate in conversations that take place in Six Sigma environments, and
2) provide a list of tools accessible to beginners to be used as an informal development plan.
Readers for which the first goal was achieved are encouraged to take full advantage of the second. Your development is your responsibility; do not wait to be invited to the “belt club.”
JayWink Solutions is available for training plan development and delivery, project selection and execution assistance, and general problem-solving. Contact us for an assessment of how we can help your organization reach its goals.
For a directory of “The War on Error” volumes on “The Third Degree,” see “Vol. I: Welcome to the Army.”
[Link] “ISO 13053: Quantitative Methods in Process Improvement – Six Sigma – Part 1: DMAIC Methodology.” ISO, 2011.
[Link] “ISO 13053: Quantitative Methods in Process Improvement – Six Sigma – Part 2: Tools and Techniques.” ISO, 2011.
[Link] “Six Sigma,” “DMAIC,” and “Design for Six Sigma.” Wikipedia.
[Link] “Integrating the Many Facets of Six Sigma.” Jeroen de Mast; Quality Engineering, 2007.
[Link] “The Role of Statistical Design of Experiments in Six Sigma: Perspectives of a Practitioner.” T. N. Goh; Quality Engineering, 2002.
[Link] “Six Sigma Fundamentals: DMAIC vs. DMADV.” Six Sigma Daily, June 17, 2014.
[Link] “DMADV – Another SIX SIGMA Methodology.” What is Six Sigma?
[Link] “Six Sigma Belts.” Jesse Allred; Lean Challenge, February 18, 2019.
[Link] “Six Sigma Belts and Their Meaning.” Tony Ferraro; 5S Today, August 22, 2013.
[Link] The Six Sigma Memory Jogger II. Michael Brassard, Lynda Finn, Dana Ginn, Diane Ritter; GOAL/QPC, 2002.
[Link] The New Lean pocket Guide XL. Don Tapping; MCS Media, Inc., 2006.
[Link] Creating Quality. William J. Kolarik; McGraw-Hill, Inc., 1995.
Jody W. Phelps, MSc, PMP®, MBA
JayWink Solutions, LLC
If you'd like to contribute to this blog, please email firstname.lastname@example.org with your suggestions.
© JayWink Solutions, LLC