The Demise of Six Sigma: The Right-Sizing of a Problem-Solving Methodology
After roughly 30 years (1979–2009), Six Sigma-DMAIC, when examined as a structured, scientific–based, problem-solving methodology (and its record of providing tangible, bottom-line benefits) stands up to the test of time. It has however, not lived up to the remainder of its wide claims as a stand-alone program for strategy, change management, leadership development, and as a quality and continuous improvement strategy, these weaknesses primarily being traced to Six Sigma’s minimal/poor consideration of the human, behavioral, and team-participative aspects of creating and driving sustainable change.
In this article, David Joecken, a Six Sigma Master-Blackbelt at a privately-owned tier-one automotive supplier, examines Six Sigma’s historical legacy and projects its future role. From Six Sigma’s initial creation and conception thirty years ago, to its widespread appeal, a potential exaggeration and overextension is postulated. A subsequent exploration and critique of the potential strengths and over-extensions is conducted in the following areas:
- Problem Solving Methodology (DMAIC)
- Bottom-Line Benefits Provider
- Overall Business and Change Management Strategy
- Leadership and Management Development Program
- Quality and Continuous Improvement Strategy (replacing TQM and other programs)
A final observation using the (character witness of) North American-specific history of quality management and improvement initiatives, postulates that Six Sigma, although a credible methodology, was oversold as part of an ongoing cultural and economic convention.
Introduction:
In the last 50 years, many continuous improvement and quality management methods and initiatives have been developed and adopted by the United States and global industry. Some of these initiatives — TWI (Training within Industry), Deming Cycle/SPC, Zero Defects, TQM (Total Quality Management), Lean, and TOC (Theory of Constraints) — appear to have been introduced as an all-encompassing panacea to an entire organization’s problems — that is, until the next imitative was introduced (Schoen, 2007). In retrospect, each of these initiatives has had something unique and specific to add towards the advancement of continuous improvement and quality management, and deserve their portion of a positive historical legacy and continued use. Six Sigma is no different. It was rolled out as a scientific/determinist approach to problem solving by Motorola in the early 1990s and quickly expanded outwards to be a solve-all philosophy of strategic and change management, continuous improvement, leadership and product development and the best approach to bottom-line savings. Recent history has shown that these ambitious claims have not been borne-out universally and rarely in particular. This article will attempt to summarize the industry’s recent experiences and issues with Six Sigma, most specifically with its shortcomings on the human side of improvement and change. This article also attempts to create a vision of what Six Sigma’s legacy will be, and which long(er) lasting benefits will be fondly remembered and practiced a decade from now.
Historical Background:
The creation of Six Sigma is generally dated to 1979 and credited to Bill Smith, a senior engineer at Motorola. Smith, in response to a management proclamation that “the real problem at Motorola is that our quality stinks!”, developed and formalized a deterministic approach to identifying and preventing defects in production using a scripted method, statistics and traditional quality tools (Schoen, 2007). It was essentially a scientific problem solving method originally using a fourstep-MAIC (Measure, Analyze, Improve, Control) approach. It became practiced derigueur in manufacturing at Motorola by the mid-to-late 1980s. Fast forward ten years, Motorola’s quality levels and more probably bottom line savings (using unique accounting practices of additive, forward extrapolation) attracted the attention of Fortune 500 CEOs like Jack Welch at GE and Larry Bossidy at Allied Signal. Welch saw Six Sigma as something more than just a production-problem-solving methodology used to improve quality, but as the core of GE’s business strategy and something to be infused into the hearts and minds of every GE employee: “This is not the program of the month. This is a discipline. This will be forever” (Slater, 1999). Allied Signal’s Bossidy “…applied it to every business, from inventing and commercializing a new product to billing and collections…” (Harry, 2006). Both CEOs also utilized Six Sigma training as a proving ground and management development program for organizational elites. During this period, the traditional (or preexisting) quality and continuous improvement initiatives, primarily TQM, were largely scrapped or simply dissolved, as was the case at AlliedSignal.
Since going mainstream, legions of Six Sigma consultants and pundits continued to stretch the basic MAIC into DMAIC (Define was added to address initial setbacks), and subsequently into an all encompassing organizational business and change strategy. The problem solving methodology was also expanded beyond improving existing process and products. Design for Six Sigma was added on as a Six Sigma approach to identifying and developing new products. Additionally, a structured Six Sigma method for creating organizational strategies and change was also tacked-on (Breyfogle, 2003). Based on this chain of events between 1979 and 2009, Six Sigma’s role expanded to: the following:
- Problem Solving Methodology (DMAIC)
- Bottom-Line Benefits Provider
- Overall Business and Change Management Strategy
- Leadership and Management Development Program
- Quality and Continuous Improvement Strategy (replacing TQM and other programs)
In the following pages, we will briefly critique Six Sigma’s performance in each of these areas, and explore the underlying reasons behind the potential successes and setbacks.
Six Sigma as a Problem Solving Methodology:
The DMAIC methodology has been generally accepted as a credible and effective approach for Problem Solving since the mid-1990s. Although the deterministic nature of y=f(x) focus and structured approach of hypothesis generation and data-based testing, are basically the core of the scientific method, the DMAIC method is underpinned by an expansive toolbox of statistical methods and very accessible improvement techniques. Essentially, all traditional statistical, quality engineering and basic TQ tools are evident in this toolbox. In isolation, the scientific method and these tools mentioned are not unique, nor invented by Six Sigma (deMast, 2007). But when both are woven into the DMAIC methodology, a distinction from previous problem solving techniques (such as TQM, Basic Quality Tools) occurs.
After the initial criticism for the looseness with the 1.5 sigma shift of long-term data, the percent defective predictions referenced in the long-term sigma tables and other statistical extrapolations (Wheeler, 2005) subsided, the scientific, deterministic and data/fact driven basis of Six Sigma led to widespread approval in academic and industry circles. Six Sigma and the DMAIC methodology are present in the curriculum of most major academic institutions; the American Society of Quality adopted and began promoting the methodology in the late 1990s. In general the DMAIC methodology has held its own, although there are lingering criticisms that the full use of methodology and its subsequent complexity, thoroughness and lack of speed, may only be necessary to solve only the smallest percent of complex problems; quality traditionalists stating that 90 percent of problems can be solved using basic quality tools (Kondo, 1994).
Six Sigma as a Bottom Line Benefits Provider:
Six Sigma’s initial savings benefits as reported by Motorola, GE and AlliedSignal were initially categorized as quality related, and because accurately assessing costs in “quality is not a concept commonly used in economic theories” (deMast, 2007), their exactness may come into question. However exact or inexact, this financially blessed quantification of benefits distinguished Six Sigma as being much more appealing than TQM for example, as a tangible benefits provider (Klefsoe,Bergquist, Edgeman, 2006.). Introductions to Six Sigma’s bottom line benefits to new potential clients displayed an endless, linear, almost exponential increase in quality savings. These based on the initial reports of these large (perhaps poorly run prior to adoption), technically diversified companies with a very high cost base (reference the slide below; a typical Six Sigma Intro Slide (Winkler, 2004); the GE portion of this slide has been subsequently removed). Because these types of potential quality cost benefits should be initially very real, management would have been very eager to adopt the approach and to include the summary savings into its regular financial reviews. Thus, the real driver of adoption of Six Sigma in many examples was this type of business case. Once the initial savings began to wane, management’s support may have waned as well, which is detrimental for the success of any long-term improvement plan. Thus, the propensity was that the subsequent savings due to Six Sigma projects in areas outside of quality were lumped together; per Jeroem deMast: “such conceptual erosion is better avoided…economists have good reason to distinguish between product quality, process efficiency, cost and capital and not labeling it all quality” (deMast, 2007). The best example for Six Sigma’s (DMAIC) wide adoption being based on the bottom-line was the less remarkable level of Design for Six Sigma’s (DFSS) adoption. DFSS is an ancillary Six Sigma method (note the addition in the slide above) that is predicated on defect prevention through proper design. Subsequently the quality benefits of defect prevention are even more difficult to quantify (Klefsoe et. al., 2006).
Overall, Six Sigma’s contributions as a benefits provider are perhaps very real but not at the level, and with the duration, promised by consultants. If the organization’s primary motive for adoption is based on these potential contributions, the effectiveness and of course longevity of the program may be compromised.
Six Sigma as an Overall Business and Change Management Strategy:
Announcing that a company was “Six Sigma” as its business strategy appeared to have been commonplace for many companies in the mid-to-late 1990s. Absent a company having a clear and documented business strategy, the Six Sigma strategy in the least gave a company the appearance of strategic direction. This “Six Sigma Strategy” is typically summarized as improving financial results by improving all aspects of all business processes (Antony, 2004). Adopting this approach to the improvement and routinization of all business processes may have been initially innovative and appealing, but an organization that is predicating its entire business strategy on superior operations is in the least short-sighted, as all or most competitors are doing the same. De Mast once again admonishes: “Six Sigma should not be [used] as a substitute for a good strategy…it should be used to implement, execute and leverage the strategy” (deMast, 2007).
Using Six Sigma as a method to identify and eliminate barriers and as tactical support to strategy attainment appears to be the strategy direction most of the “Six Sigma” companies such as 3M, Home Depot, Raytheon and GE(!) have gravitated towards, since the cost savings promises mentioned in the previous section were more or less unfulfilled, and growth expectations underperformed (Hindo, 2007). The 3M story of unbridled devotion to Six Sigma as an all-permeating strategy is the most well documented example of attempting to routinize and potentially stifle areas that create true competitive advantage and are traditionally left alone such as innovation and R&D. Although they had recorded the GE-like-savings from their Six Sigma program initially, creativity and innovation were thwarted and subsequently earnings per share and growth were affected for a string of consecutive quarters. The necessary shift in business strategy that 3M is undergoing; “from profitability and process discipline to growth and innovation” (Hindo, 2007), is perhaps the by-product of using Six Sigma as a stand-alone business strategy.
Although no specific dismantling of Six Sigma’s performance as a change management strategy occurred in this section, its weaknesses relative to addressing the human aspects of change, motivation and behavior will be addressed in the following two sections.
Six Sigma as a Leadership and Management Development Program:
As companies adopted the Six Sigma approach to problem solving, driving bottom line benefits, and as an overarching strategy, they invariably gravitated towards the practice of labeling their Six Sigma Champions, Master Black Belts, and Black Belts as their future leaders. Thus, Six Sigma also became a development program and proving ground for an organization’s managers and future leaders. During my employment at AlliedSignal and Honeywell, it was observed that Six Sigma Black Belt and Champion training was the official leadership development program for a time. This was not altogether incorrect, as Deming stated in his System of Profound Knowledge in The New Economics that the purpose was to “transform the prevailing style of management” and three of the four main concepts in this system: “knowledge about variation, appreciation for a system, and theory of knowledge” (Deming, 1994), were directly and indirectly addressed with the Six Sigma training on variation reduction, Input-Transformation-Output, and DMAIC as a substitution for Plan-Do-Check-Act .
The remaining concept: understanding the psychology of individuals (Ishikawa described this as respect for humanity) appears to be sorely lacking in a typical Black Belt training program. There is, from personal experience with three different Six Sigma programs, very little understanding and dissemination of team building and motivational theories shared in the training. At Honeywell, Black Belts were generally instructed that they had carte-blanche to recommend any necessary changes based on their advanced training and elevated status; that was perhaps all that was needed to understand about team dynamics, collaboration and individual motivation. This is usually sufficient to affect many improvements at the process-level. As a matter of course, Six Sigma-style leaders may attempt to apply the deterministic principles of dependent and independent variables, and finding the critical Xs to the behavior of individuals and the organizational culture at large, and, as David Lengacher, resident Master Black Belt at Achieving Business Excellence magazine, warns us that this: “is one area of continuous improvement where it has, thus far, failed miserably. The failure occurs when organizations identify culture, or workforce behavior, as one of the critical Xs that drive performance [they are] response variables …and almost always impossible to measure” (Lengacher, 2008). This lack of appreciation, or perhaps utter disregard, for the complexity of human behavior and organizational culture may have its foundation in the Welch-style-edict that since participation and support are mandatory and non-negotiable, individual motivation and cultural transformation are essentially guaranteed; this is speculative at this point. We will explore this point further in the next section.
In summary, Six Sigma does provide future leaders with a solid understanding of the concepts of systems thinking, variation, and the accumulation of knowledge, as well as stretching them with a difficult project and preparing them for fact-based leadership. But, based on the negligible appreciation for human and cultural aspects, Six Sigma training as currently designed should not be considered as a comprehensive leadership development program.
Six Sigma as a Quality and Continuous Improvement Strategy (and as a replacement for its predecessors; TQM, and other programs):
Many of the North American organizations that adopted Six Sigma for the reasons listed above, already had adopted a concerted (or a multitude of ) quality and continuous improvement program(s). These programs were instituted in the 1980s and early 1990s as response to the “economic growth and manufacturing dominance of Japanese industries” (Klefsoe, et. al., 2006). Most of these programs, such as “Just-in-Time (JIT), Total Quality Management (TQM), Statistical Process Control, Target Costing, Benchmarking, Process Re-engineering, Theory of Constraints (TOC), Activity-based Costing/Management (ABC/M), Balanced Scorecard (BSC)” (Albright, Lam, 2006) were based in part on the tenets and results of the Toyota Production System (TPS) that was developed in an evolutionary manner since the 1950s through the work of W. Edwards Deming and Taichi Ohno. Because of the slow, evolutionary basis of TPS development, there was no single, unequivocal codification of what it was, and how it could best be transferred to North American industry; there was just the clear desire to be as efficient and quality-minded as the Japanese. Thus began a proliferation of consultants and piecemeal adoption of the many initiatives listed above, some attempting to address specific issues in management and manufacturing (SPC, JIT), others attempting to address the system more holistically (TOC, TQM, Lean); all or most of these initiatives expounded the concept of “Kaizen” or continuous improvement (Albright, Lam, 2006,). The adoption of TQM as a management system and continuous improvement program appears to have been the most common, perhaps based on the system-wide approach or the superiority of its consultants (sic). However, after a decade or so of unabated rollouts of new initiatives, there is some evidence that US industry’s desire for quality improvement began to wane (Schoen, 2007) in the mid-to-late 1980s. Enter Six Sigma as a quality and continuous improvement management system.
Although TQM shortcomings such as its vagueness and difficulty with assigning financial benefits (Schoen; 2007) were easily replaced by Six Sigma’s clear DMAIC approach and bottom-line promises, its strengths were the focus on continuous improvement, teamwork and involving every individual within the organization. Many critics of Six Sigma feel that although the project-focused nature of Green Belt and Black Belt improvements are a key component in its success, typically, these type of activities do not iteratively improve the same process. They also see “a risk that the belt-based infrastructure has an unavoidable tendency to glorify some people and, hence, not sufficiently support the TQM value of ‘everybody’s commitment’” (Klefso, et. al.; 2006). This glorification of the few, and project-based nature of Six Sigma is represented and evidenced by the typical requirements and deliverables of its Control Phase, where a work-instruction and/or a Control Plan may be created or modified, but no clear and overt requirement to ensure that a formal communication and buy-in from the process owners and participants (the non-Belts), as well as an admonition for an ongoing cycle of improvement is specified (Breyfogle, 2003). The Belts simply hand off the project improvements to the owner and move onto the next Pareto item, the next project. Additionally, Six Sigma teams are typically formed situationally around the completion of the Belt projects, thus the perceived benefits of team dynamics may also be temporary. This differs vastly from the Toyota/TPS concept of continuous improvement in evidence by the Toyota/GM joint venture NUMMI where there is no “Belt” hierarchy, all employees are strongly encourages to participate, and the control mechanism, Standardized Work,” is the basis for iterative, continuous improvement (Adler, 1993).
Despite a potential counter-argument from the Six Sigma camp that they would iteratively use the data to revisit and improve necessary processes and problems again and again (using the same team members) once the initial “low hanging fruit” projects are completed, this is also not formally communicated in Six Sigma literature. Therefore, it appears to be a diffuse and exclusionary approach to continuous improvement; one that does not fully utilize the participation of those best suited to sustain and continue the improvements.
As stated similarly in the previous section on leadership, Six Sigma’s faint recognition of the importance of human aspects and behavior in continuous improvement may be a major factor of consideration and drawback when applying it as a Continuous Improvement program. That said, the Toyota/TPS approach to “Kaizen” was permitted a solid twenty-plus years to develop prior to US industry taking notice of its benefits. Whether Six Sigma adjusts its structure to more thoroughly address the human side of change and continuous improvement remains to be seen.
Conclusion and the potential legacy and future role of Six Sigma:
Based on the examinations of Six Sigma’s purported and actual benefits in the sections above, it appears clear that Six Sigma’s time as a complete management system and improvement methodology has already come to an end. . There is now an unusual level of agreement between the staunchest critics that “it should only be used in product manufacturing where the idea of reducing defects really makes sense,” to its oldest advocates like Jack Welch, who now believes [Six Sigma] should “not be used everywhere, that Six Sigma’s role will be minimized in the future” (Davenport, 2008). Another indication of this potential minimization and adjustment is the recent marriage of Lean and Six Sigma — Lean Sigma — an attempt to address Lean’s lack of structure and data and Six Sigma’s shortcomings with waste reduction (outside of defects). These are compromises that may lead to more TQM-like dilution and lead to reiteration of Six Sigma’s shortcomings with the human aspects, “although [LeanSigma] certainly extends the power of Six Sigma, its not Lean…these methods tend to neglect the people and the cultural, the accessibility and the inclusiveness” (Sayer,Willams, 2007).
These negatives aside, a clear legacy and future benefits will remain for Six Sigma, best summarized by Thomas Cutler in Business Excellence Magazine: “…after thousands of six sigma deployments over the past decade or so…organizations are focusing on becoming data driven, paying attention to real-time data and turning that into knowledge. When they do this, six sigma becomes a tool kit to solve certain types of problems, but not the definitive methodology that runs a business.” (Cutler, 2008). Reading between these lines and in summary, the following observations and recommendations can be proposed:
- Six Sigma’s strengths and legacy should be summarized as having provided an accessible, structured, data driven, project-based, results oriented problem solving methodology (DMAIC) with a powerful toolset; it is a proven provider of tangible results. Therefore:
- Six Sigma’s future role in the (data-driven) organization should be the appropriate utilization of DMAIC (all measureable problems should be approached with this method) and its complimentary array of tools within the framework of a larger quality management and continuous improvement strategy. Six Sigma should not provide the strategic or management directions (overall business, quality, HR, finance, etc.); it should support the goal attainment of these higher level directions as a tactical methodology. The larger framework would ideally use DMAIC to monitor the success of Six Sigma, Quality, Lean, TOC and simple, TQ-style improvement projects using an iterative, PDCA-like approach.
Final note and recommendation for future research:
During the writing of this article it became rather evident that the furor behind the initial widespread appeal and subsequent overextension of Six Sigma’s purpose and role in the organization appears to be part of an ongoing tradition within US quality management; reference the amount of US initiatives and their literature vs. Japanese-based initiatives. The speed and the breadth of this recent overextension may be explained culturally (consumerism), economically (shortsightedness), or simply blamed on their side-effects: the legions of US quality consultants and pundits. Whatever the reasons (these should be further explored), without a solid, credible treatise, comparing and contrasting the shelf lives of US quality initiatives with slowly evolving TPS-like initiatives, it is quite probable that the “Next Six Sigma” will be here very soon. And it won’t have a complicated name.
We believe that future is Tallyfy.
