2012

STATISTICS ROUNDTABLE

Turning Shewhart’s Challenge Into Opportunity

by Ronald D. Snee

Nearly 70 years ago, quality pioneer Walter Shewhart threw down the gauntlet: “The long-range contribution of statistics depends not so much on getting a lot of highly trained statisticians into industry as it does in creating a statistically minded generation of physicists, chemists, engineers and others who will in any way have a hand in developing and directing the productive processes of tomorrow.”1

To that list of technically oriented people, he should have added leaders and managers. Although statistics has made some headway in technical functions, we still have not created a generation of statistically minded leaders. After all, leaders are the ones who have the ultimate responsibility for “directing the productive processes.” As it becomes ever clearer that leaders are not going to become statistically minded on their own, it falls to statisticians to help show them the way.

The Opportunity

On the surface, it looks like an easy sell. Under the pressures of global competition and relentless technological advances accelerating the speed of business, the need for significant, measurable and continuous improvement has never been more obvious. Companies across industries face imperatives to:

  • Get to market faster with new products.
  • Achieve more compliant processes.
  • Provide on-time and in-full customer delivery.
  • Improve throughput, cost per unit and margins.
  • Improve yields with fewer defects, rework and scrap.
  • Increase equipment uptime and plant and capacity usage.

Statistically based improvement methods like statistical quality control (SQC), statistical process control (SPC), Six Sigma and lean have been demonstrably effective in improving performance.2,3 Statistical thinking and methods also address variation, which has been found to affect all measures of process performance—quality, cost, delivery and customer satisfaction.4,5 Moreover, widely available, easy-to-use software puts the ability to do sophisticated calculations at virtually anyone’s fingertips.

And, as if all those factors aren’t persuasive enough, a number of companies have made statistical thinking a requirement for their leaders and an integral part of their operations. These companies include Motorola, Allied-Signal/Honeywell, General Electric, Home Depot, DuPont, Bank of America, 3M and W.R. Grace.

Despite these notable examples and the availability of methods and software, legions of today’s leaders remain blissfully unaware that by empowering nonstatisticians with statistical thinking and tools, they could dramatically improve performance and make far better business decisions.

Meanwhile, many statisticians and quality professionals have remained stuck in their now rapidly diminishing consultative roles of analyzing data, designing experiments, teaching statistical tools and consulting on other people’s projects while having little leadership responsibility and less accountability.6

This impasse offers statisticians and quality professionals a golden opportunity to step forward and make the case to leadership for creating the generation of statistically minded leaders that could take their organizations to the next level of continuous, sustainable improvement. As Table 1 shows, throughout the past century we’ve seen steadily increasing importance placed on statistics.

In addition, many statisticians and quality professionals have migrated from traditional consultative roles to new leadership roles in improvement efforts, as shown in Table 2. Today, those statisticians and quality professionals are in position to complete the changeover.

What would success look like? In statistically minded organizations, you would find employees of all types working in teams, using statistical thinking and methods and other problem solving tools in combination with process knowledge and understanding to solve problems and improve processes. George Box has pointed out that the ingredients for achieving this success are ready in hand:7

  • Simple, but powerful, problem solving tools
  • Human resources to apply them

However, Box also points out that breakthroughs cannot happen until top management realizes the enormous potential of these ingredients and resolves to empower the workforce to apply improvement tools in all operations on projects likely to save money.

Starting the Revolution

We’ve learned repeatedly that management must lead any culture change if it is to be successful in the long term. Statisticians can begin this revolution by showing leadership the power and potential power of a statistically minded organization—that only results from a carefully designed program based on these eight principles:

1. Keep a focus on the bottom line. Too often, statisticians and quality professionals take the attitude that “show me the data” is their paramount responsibility. In fact, it should be “show me the money.” In proposing pilot projects in Six Sigma or some other statistically based improvement method, make it clear to leadership that the principle of project selection will be bottom-line impact.

No project will be approved unless that bottom-line benefit has been identified and validated by the finance department. In my experience, such projects typically yield annual savings of $50,000 to $250,000 and even more.

2. Provide a small number of tools linked and sequenced in an overall improvement framework. Exposing people to a wide variety of tools “cafeteria style” and expecting them to select the right ones and figure out how to use them is a low-yield strategy. Instead, provide a small set of tools integrated with a problem solving framework that is sequenced and linked together. The user will know how the output of each tool is used as the input for one or more other tools.

The five-phase define, measure, analyze, improve and control (DMAIC) improvement process has proven to be a highly effective improvement framework. The broad use of DMAIC for improving existing processes adds predictability, discipline and repeatability to projects and provides a straightforward vehicle for promoting the widespread use of statistical and quality techniques.

Through DMAIC you can link and sequence the required tools regardless of their source, including lean (which seeks to eliminate various forms of waste through such approaches as just-in-time manufacturing), total quality management, which seeks to integrate all organizational functions to meet customer needs and organizational objectives) and other improvement tool sets.

The key, however, is to confine the set to a few tools and make sure each tool generates outputs that become inputs for the next tool in the sequence.

3. Provide easy-to-use software. Widely available, easy-to-use software has let the statistical genie out of the bottle. Statisticians no longer have the market cornered for calculating statistics and applying statistical tools. Instead of resisting such software, make it available to people throughout the organization and teach them how to use it effectively.

Don’t worry about the software replacing you. It will make you obsolete only if you cling to your old role as a reactive statistical services provider responding to other people’s projects instead of a proactive leader of improvement.

4. Provide how to, project based training. Leaders often hesitate to embark on full-scale improvement programs based on unfamiliar methods because of the great investment of time in training and the long lag time between that training and concrete results. You can overcome this objection by combining training with live projects. Project based training not only produces immediate financial business results, but it also makes for far better training because participants will take real-world projects more seriously than classroom exercises.8

5. Use a comprehensive training success model. All too often, training evaluation forms only ask whether the participants liked the training and learned the material. Project based training demands a more robust model of training success. Certainly find out if participants liked the training, using session surveys and end-of-day feedback forms. Use mentoring, project reviews and certification exams to determine whether they learned the method.

But add two more dimensions to the evaluation. First, did they use the methodology? For example, how many of the tools, on how many completed projects and in what period of time? Second, did they get results, measured by improvements in process performance and in bottom-line results?

6. Create supporting infrastructure. As management begins to understand the power of statistically based improvement, and projects begin to proliferate, stress the need for dedicated personnel and management systems and the need for a permanent cadre of improvement leaders. Management systems should be created to embody the new way of thinking in the entire organization and sustained over time—creating, in short, a new culture.

Those management systems include: a project selection process, career planning for improvement leaders, training capability, a project review process, a project reporting and tracking system, an audit system for previously closed projects, a reward and recognition plan and a communications plan.

Also, stress to management that an infrastructure of quality conscious people—permanent change agents who lead, deploy and implement improvement projects—provides the organization with a significant advantage over competitors with a less rigorous and systematic approach to improvement and organizational change.

7. Use top talent. Significant improvement is too important and difficult to be left to anyone other than top talent. Help leadership understand that the clearest message it can send about how seriously it regards an improvement initiative comes when it announces the names of people taking key roles. Rank and file employees know their colleagues well. They will judge immediately whether the effort has been staffed with top talent or people simply available or unable to do anything else.

8. Harvest the low hanging fruit. The more quickly you can produce tangible results from statistically based improvement projects, the easier it will be to convince management of their value and to win the allegiance of participants. In the early stages of improvement, project teams should look for quick wins, such as:9

  • Correcting obvious problems with a process
  • Fixing broken measurement systems
  • Ensuring the consistency of process inputs, whether raw materials in manufacturing or data in nonmanufacturing processes

Work should also be streamlined through the reduction of complexity, waste and nonvalue added work.

The later stages of improvement can focus on optimizing and controlling processes by improving value added work steps, shifting the process average and reducing variation around it, improving process flow and reducing cycle time. But first, gain some traction for the program. Remember, one of the strongest spurs to maintaining momentum and sustaining gains comes from the effect that achieving of significant, measurable benefits has on the culture.

People like to succeed. When they see tangible results, they are eager to repeat the process. So is management.

Expanding Roles

Now more than ever, statisticians and quality professionals have opportunities to influence how organizations run their operations. To be successful in this new role, statisticians and quality professionals must recognize that they are in the culture change business, which requires that they first help change leadership’s thinking through the kind of carefully structured initiatives and principles described here.

As the world of statisticians and quality professionals expands from problem solving to process improvement to organizational improvement, the ultimate culture change—changing how people think—will follow. This is illustrated in Figure 1.


Becoming a leader is not as daunting a task as it might seem. The essential principles of leadership are simple. As a leader, you should:10

  • Provide direction: Leaders show the way.
  • Communicate: Leaders develop understanding and hope.
  • Enable, coach, counsel and provide resources: Leaders set people up for success.
  • Recognize results and reinforce desired behavior: Leaders catch people doing things right.

For statisticians and quality professionals, precipitating change in management thinking will be the first test of leadership. Lead the leaders and you become one yourself. © Ronald D. Snee, 2007

REFERENCES

  1. W.A. Shewhart, Statistical Methods From the Viewpoint of Quality Control, Graduate School of the U.S. Department of Agriculture, 1939.
  2. R.D. Snee and R.W. Hoerl, Leading Six Sigma—A Step-by-Step Guide Based on Experience With GE and Other Six Sigma Companies, Financial Times Prentice Hall, 2003.
  3. R.D. Snee and R.W. Hoerl, Six Sigma Beyond the Factory Floor—Deployment Strategies for Financial Services, Health Care—and the Rest of the Real Economy, Financial Times Prentice Hall, 2005.
  4. R.W. Hoerl and R.D. Snee, Statistical Thinking—Improving Business Performance, Duxbury Press, 2002.
  5. R.D. Snee, “Making Another World: A Holistic Approach to Performance Improvement,” Deming Lecture, Joint Statistical Meetings, Seattle, Aug. 8, 2006.
  6. R.D. Snee and R. W. Hoerl, “Statistical Leadership—As Traditional Workplace Roles Change, Learn to Transition From Consultant to Leader,” Quality Progress, October 2004, pp. 83-85.
  7. G.E.P. Box, “The Missing Ingredient,” Six Sigma Forum Magazine, February 2006, pp. 20-21.
  8. R.D. Snee, “Make the View Worth the Climb—Focus Training on Delivering Better Business Results,” Quality Progress, November 2001, pp. 58-61.
  9. John P. Kotter, Leading Change, Harvard Business School Press, 1997.
  10. Snee and Hoerl, “Statistical Leadership—As Traditional Workplace Roles Change, Learn to Transition From Consultant to Leader,” see reference 6.

RONALD D. SNEE is principal of performance excellence and lean Six Sigma initiative leader at Tunnell Consulting in King of Prussia, PA. He has a doctorate in applied and mathematical statistics from Rutgers Uni-versity in New Brunswick, NJ. Snee has received the ASQ Shewhart and Grant Medals and is an ASQ fellow.

Sign Up for Newsletters