Statisticians and Statistical Organizations How to Be Successful in Today’s World? Ronald D. Snee Snee Associates With Significant Contributions from Roger W. Hoerl, General Electric 2009 Quality and Productivity Research Conference IBM T. J. Watson Research Center Yorktown Heights, NY June 3-5, 2009
The statistics profession is at a critical point in its history and has been for some time. The May 2008 Technometrics article, “Future of Industrial Statistics”, summarized many of the major issues. Two key drivers are global competition and the rapid growth of information technology. The old model for the use of statistical thinking and methods in business and industry, which has been around for at least 50 years, does not work in today’s business environment. This presentation begins with a brief summary of the current state of the profession and then moves quickly to a focus on what statistical organizations and statisticians as individuals need to do to effectively deal with the new environment. The focus is on strategies and approaches that have been found to work. Several case studies will be presented to illustrate the “new model” and the needed changes.
We Need to Change our Thinking
What Should Statisticians be Doing?
Helping Our Organizations Succeed
Focus on Statistical Engineering
“ Embedding” Statistical Tools in Work Processes
Profession appears to be at a crucial point in its history
Recent Technometrics article and blog highlight major issues we must deal with going forward
“ Future of Industrial Statistics: A Panel Discussion”
ASQ Stat Division Newsletter article by Vijay Nair
Disconnect between academic research and practice
We haven’t fundamentally modernized the “model” for applied statistics since the 1950’s
Pure science versus statistics as an engineering discipline?
Leadership is lacking and desperately needed
No evidence that we have critical mass to change
How Should We Respond?
Jump in “fox holes” and wait for the crisis to blow over
Argue against globalization
Understand the fundamental changes in our environment,
Adapt to them
Take advantage of them
Understanding today’s environment will help us understand the future of statisticians and statistical organizations
The Choice is Yours “ Survival Isn’t Mandatory” W. E. Deming
Expanding World of Statistics The Profession Has Responded
Launching of Sputnik by the Soviet Union:
Created the need for design of experiments and other statistical methods in research and development
Food, Drug and Cosmetics Act created the need for statisticians in the pharmaceutical industry
Clean Air Act and the Environmental Protection Agency created the need for environmetrics and the use of statistics in solving environmental problems
Global Competition and Information Technology creates need for improvement
Needs of Employers and Society Define the Roles and Uses of Statistics
Expanding Role of Statisticians
Consult on other people’s projects
Perform routine analyses if needed
Teach statistical tools
Work with technical people
Narrow expertise and accountability
“ Benign neglect”
Lead or collaborate on our own projects
Focus on significant, complex problems
Design training systems
Work with managers and technical people
Broad expertise and accountability
“ In the firing line”
Consultant Collaborator/Leader Computer Scientists Provide an Example of Such a Role
What Should Our Focus Be?
“Anyone can manage for the short term or the long term; real success comes from managing both short term and long term at the same time…
If you don’t manage in the short term, there won’t be a long term” (Jack Welch).
“The complex problems of this world will not be solved at the same level of thinking we were at when we created them.” (Albert Einstein)
We need to
Be bold but not reckless
Helping your Organization Deal with the Global Financial Crisis – Short Term
Cost reduction and short term cash flow
Quick wins essential for sustaining change (John Kotter)
Prudent risk taking
Process understanding is needed
Reducing variation reduces risk
Effective prioritization – working on the right things
Improvement project selection
Customer and employee surveys
Follow the money
Statisticians Can Play a Major Role in Each of These Areas
Reinvigoration of Improvement Bottom Line Improvement Never Goes Out of Style
Some may respond, “been there, done that.”
“ We have already done Lean Six Sigma, and now moved on to bigger and better things”
Improvement is particularly needed now
Lean Six Sigma also helps us make sure that we are working on the right things
The result will be
Immediate, bottom line results
Help with business prioritization
Risk management approaches that balance need for income generation with need to limit risk
What Else Should Statisticians be Doing? A Longer Term View
Greater emphasis on “statistical engineering” relative to “statistical science”
“Embedding” statistical methods and principles into key business process
Making the use of statistical thinking and methods part of how we work
What Does Society Need from Statisticians?
Decades of the 1950s, 60s and 70s
Statistical science needed to be developed to deal with the problems encountered in R&D, Manufacturing and other functions including:
Efficient and effective experimentation
Need for statistical engineering was there, but limitations of available methods created a stronger need to develop statistical science.
21 st Century
Society needs statistics to be primarily an engineering discipline, with a secondary focus on statistical science.
Engineering focuses on how to best utilize known scientific and mathematical principles for the benefit of mankind.
Pure science works to advance our understanding of natural laws and phenomena.
Chemist may attempt to advance understanding of the fundamental science of chemistry
Create a new marketable substance
Chemical engineer would more likely attempt to better utilize the current understanding to greater human advantage.
Determine how to scale up the process to produce this substance commercially,
Engineers Develop Engineering Theory
Engineers do research to develop new theory
Engineers’ theoretical developments:
Tend to be oriented towards the question of how to best utilize known science to benefit society
Rather than on how to advance known science.
Two Examples of Statistical Engineering
Product Quality Management at DuPont
Process and Organizational Improvement Using Lean Six Sigma
PQM – Statistically Based Product Quality Management System
Product Quality Management (PQM)
Framework for managing the quality of a product or service.
Operational system the enables Marketing, R&D, Production and support personnel to work together to meet increasingly stringent customer requirements
“ Within two years product quality had improved to the point of commanding a marketplace advantage and more than $30 million had been gained in operating cost improvements. The statistically based Product Quality Management system developed for “Dacron” was expanded to other products with further contributions in earnings.”
Richard E. Heckert
Chairman and CEO, DuPont Company
ASA Annual Meeting 1986
PQM System – Statistical Techniques Used
Product Release Procedures
CUSUM Process Control
ANOVA and Variance Components
Design of Experiments
Response Surface Methodology
Sense of Urgency Goals Problems Gaps Define Improve Control Results ($$) Measure Analyze
DMAIC Process Improvement Framework II- Data Lean Six Sigma Tools
Six Sigma Uses a Small Set of Tools Capability Analysis Control Plans and SPC Design of Experiments Multi-Vari Studies Failure Modes & Effects Analysis Gage R&R Cause and Effect Matrix Maps Project Charter Control Improve Analyze Measure Define Tool
Six Sigma Tools are Sequenced and Linked Process Process Map Customers Improvement Need FMEA Control Plan C&E Matrix MSA Process Capability Multi-Vari DoE SPC
The Tools Are Part of An Improvement System
Systematic, Focused Approach
Process & Financial ($$)
Recognition and Reward
Improvement Initiative Reviews
Linked to Business Goals
Project Portfolio Management
Sustain the Gains:
Project Tracking and Reporting
Methods and Tools
Facts, Figures, Data
Define, Measure, Analyze, Improve, Control
8 Key Tools:
Sequenced and Linked
Critical Few Variables
Embedding Statistical Thinking in Core Business Processes – Some Examples
Product Quality Management at DuPont
Design and analysis of clinical trials conducted by pharmaceutical and biotech organizations
Driven by FDA
Track safety and injury data – Mandated by OSHA
Managers often study tabular reports and respond to random variation
Plotting safety data over time on a control chart , or even a run chart , can save a lot of time and effort by providing a more insightful view of the process performance.
If the appropriate statistical tools are part of the information system, we would say that tools have been “embedded”.
Whether we like it or not, our environment today is radically different than even 10 - 15 years ago
To prosper in the 21 st century, statisticians need to play broader leadership role
More pro-active and clearly value-adding.
Focus should be on:
Bottom-line improvement – It never goes out of style
Significant, complex problems
Embedding statistical approaches in work processes
A High-Yield Strategy Change Before You Are Forced to Change
Hoerl, R. W. and R. D. Snee (2002) Statistical Thinking – Improving Business Performance , Duxbury Press, Pacific Grove, CA.
Kotter, J. P. (1996) Leading Change, Harvard Business School Press, Boston, MA.
Marquardt, D. W. (1991) ed., PQM: Product Quality Management (Wilmington, DE: E.I. DuPont de Nemours & Co. Inc., Quality Management and Technology Center). A shorter version appears in Juran's Quality Handbook 5 th Edition
Snee, R. D. and R. W. Hoerl (2003) Leading Six Sigma – A Step by Step Guide Based on the Experience With General Electric and Other Six Sigma Companies , FT Prentice Hall, New York, NY,
Snee, R. D. and R. W. Hoerl (2005) Six Sigma Beyond the Factory Floor – Deployment Strategies for Financial Services, Health Care, and the Rest of the Real Economy, Financial Times Prentice Hall, NY, NY.
Technometrics (2008) “Future of Industrial Statistics – A Panel Discussion. Technometrics Blog Link asq.org/discussionBoards/forum.ispa?forumID=77
Cost Reduction and Short Term Cash Flow
Bottom line improvement is needed today more than ever before in, at least in recent history
Productivity = System output / resources used.
You can increase productivity by reducing resources or by increasing system output.
We believe that the statistics profession could be well positioned to identify ways to improve the system
Reinvigoration of Lean Six Sigma can provide the needed improvements
Big Opportunity – Project selection
Prudent Risk Taking – Process Understanding is Needed
Prudent risk taking can be done when we understand our processes;
Critical process drivers
Capability of the processes to meet customer requirements.
Greater use of data and statistical tools can lead to better process understanding.
Statisticians have much to offer regarding quantifying risk and making decisions in the face of this uncertainty
Effective Prioritization – Working on the Right Things
Effective prioritization is always important, but particularly critical in this economy.
Many companies have gone through massive layoffs.
There are simply fewer resources available, both in terms of people and money.
Yet work has to be done if results are to improve.
Careful prioritization of critical needs is required to identify what must be done and what can be dropped or done later
Statisticians can help the organization:
Focus on a few key strategies,
Use data to identify and prioritize improvement opportunities
Use employee and customer surveys to identify opportunities,
Follow the money - large income and expenditures are often opportunities for improvement.
For Further Information, Please Contact: Ronald D. Snee, PhD Snee Associates (610) 213-5595 [email_address]