Quality Progress | June 2013 P Putting Best Practices to Work www.qualityprogress.com | June 2013 Minimizing Impact of Supplier Inconsistency p. 26 QUALITY PROGRESS Within Supply Chain Reach Bolstered supply chain ensures drugs get to HIV/AIDS patients in Kenya p. 18 Plus: Could automated trading use a dose of quality? p. 12 Library service quality in the digital age p. 40 Volume 46/Number 6 increase Your effectiveness With valuable lean Six Sigma Training Stand out From the Crowd With lean Six Sigma Training and Certification From villanova University – 100% online Gain advanced expertise that can distinguish you and increase your earning potential with a Master Certificate in Six Sigma. Renowned Villanova University, ranked by U.S. News & World Report as the #1 Regional University in the North for nearly two decades, provides resumeboosting Lean Six Sigma training – 100% online. Attain Prestigious lean Six Sigma Credentials online As a leader in Lean Six Sigma education online, Villanova enables you to earn must-have credentials through innovative video-based e-learning technology. Enjoy maximum mobile convenience with streaming video, and downloadable MP3 and MP4 lectures as you complete your Six Sigma master certificate online. Then round out your credentials with certification in one of these in-demand areas: Six Sigma Green Belt Lean Sensei Lean Black Belt Lean Master Black Belt Act Now to Save $945 on a 3-Course Master Certificate! 800-341-7106 | www.VillanovaU.com/Efficiency © 2011 All Rights Reserved. Made Available by University Alliance® – The Nation’s Leading Universities Online. SC: 198042zv1 | MCID: 14408 TrAdiTioN. ACAdeMiC exCelleNCe. oNliNe CoNveNieNCe. ^ĞůĞĐƚƚŚĞnterprise Quality & PDM NCMR QMS Software ISO/TS FMEA ISO Nonconforming Quality ISO 9000 PDM Quality Software ISO 13485 NCM QMS Nonconforming Materials TL 9001 CAPA ISO Quality Compliance Software Manufacturing Quality Assurance Product Data Management Manufacturing Supplier & Materials Quality Compliance Process Nonconformance Discrete Corrective Actions Quality Discrete ERP Manufacturing Quality Compliance QMS Software Discrete Risk Assessment Nonconforming Materials ISO 9000 FMEA ERP ISO/TS MRB QMS Software NCM ISO 9000 Discrete Calibration ISO/TS 16949 ISO/TS ISO 13485 FMEA Supplier & Materials AS9100 Compliance Corrective Actions AS9100 TL 9001 Nonconforming FMEA Actions Inspections Corrective Manufacturing Quality ISO ISO 9000 Quality Assurance Risk Assessment Calibration Process ISO/TS ISO Product Data Management ISO/TS 16949 ERP ISO 13485 Receiving Inspections CAPA Calibration Quality Quality Software Process Supplier Quality Assurance MRB CAPA PDM ISO/TS Supplier TL 9001 Product Data Management Materials Inspections Nonconformance FMEA Quality Systems Software Receiving Quality Management Software Supplier Rating MES CAPA QMS Software Manufacturing Product Data Management Calibration Quality QMS Process Supplier & Materials QMS QMS Software Calibration ISO/TS QMS Software Quality FMEA Manufacturing ISO 13485 Rating CAPA ISO/TS QMS Calibration ISO 13485 PDM AS9100 Quality Calibration Quality Compliance QMS PDM Manufacturing Receiving Nonconformance Quality Assurance Materials Compliance ERP CAPA ISO Process ISO 13485 ISO/TS MRB TL 9001 Process ...ǁŝƚŚƚŚĞŵŽst VALUE : Integrated modules for Quality and FDA Compliance Management: CAPAͻŚange ManagementͻRisk Assessment ...and more! Risk Management in tŚĞQuality System events Flexible: Leading edgĞŇexible workŇow adapts to all business processes, witŚŽƵt programming : Integrates witŚ3rd party business systems Scalable: Readily adapts to enterprise environments, and deployments Supplier Management: Collaborates witŚ^ƵƉƉůŝĞrs tŚrougŚ^ƵƉƉůŝer Business Intelligence in decision-making witŚŚƵŶĚreds of conĮgurable cŚarts and reports 800-354-4476 ͻ info@etq.com www.etq.com/quality What’s Online in the ASQ Knowledge Center? CASE STUDY Shortening Throughput Time of a Hospital’s Billing Process See how a Dutch hospital applied Lean Six Sigma to shorten its billing process by one month. Key tools used for the project included the Gantt chart and critical path analysis. BASICS Learn About Quality—Root Cause Analysis Brush up on the basics of root cause analysis and see recommended resources for more information, including a two-part introductory webcast, and case studies and articles about successful applications, and books and courses to continue your learning. WEBcast Patient Safety—Eliminating Harm System-wide Learn how Henry Ford Healthcare System achieved a 26 percent reduction in harm events and a 12 percent reduction in mortality over a three-year period through its No-Harm Campaign. The approach involves every business unit and all members of the workforce, and is integrated into the organization’s strategy, culture, daily operations, and educational mission. Access this month’s featured content and more Web exclusives in the ASQ Knowledge Center at asq.org/knowledge-center/featured.html. training CERTIFICATION CONFERENCEs MEMBERSHIP PUBLICATIONS Contents Putting Best Practices to Work | June 2013 | www.qualityprogress.com FEATURES 18 CASE STUDY To All Corners of Kenya Delivering medicine to Kenyans living with HIV/AIDS was much improved through a project that used an ISO 9001 approach to solidify the supply chain that handled the life-saving drugs. by Tiffany Darabi 26 SUPPLY CHAIN The Power of Prediction Use quality and statistical methods to predict parts demand and adjust to inconsistent supplier deliveries to stay one step ahead in servicing customers. by Gregory A. Kruger 34 STATISTICS 18 Partial View Repeatability and reproducibility studies can help evaluate the parts of a measurement system that may need improving. by Víctor M. Aguirre-Torres and Maria Teresa López-Alvarez 40 SERVICE QUALITY Getting a Good Read Implementing quality management helped libraries learn to better serve their patrons in the internet age. by Todd Bruns and Rendong Bai Only @ www.qualityprogress.com • Keen on Kenya Hear from Tiffany Darabi, author of “To All Corners of Kenya,” p. 18, discuss the project that strengthened supply chain activities and helped deliver HIV/AIDS drugs more efficiently to patients throughout the nation. • Volviendo a los Fundamentos Back to Basics translated in Spanish. • Ready Reference Brush up on quality fundamentals, including your vocabulary with our handy glossary. 26 DEPARTMENTS 6 LogOn 8 Expert Answers 14 47 • Extolling the value of statistics. • When an audit finding is refuted. • Banking on customer satisfaction. Quality Progress/ASQ 600 N. Plankinton Ave. Milwaukee, WI 53203 Telephone Fax 800-248-1946414-272-1734 414-272-8575 Keeping Current Mr. Pareto Head 66 QP Toolbox 68 QP Reviews Email Follow protocol of first initial and full last name followed by @asq.org (for example, vellifson@asq.org). Article Submissions Quality Progress is a peer-reviewed publication with 85% of its feature articles written by quality professionals. For information about submitting an article, call Valerie Ellifson at 800-248-1946 x7373, or email manuscripts@asq.org. COLUMNS 5 12 QUALITY PROGRESS Mail • A new plan for Toyota. • ASQ World Conference wrap-up. 14 QP Up Front Motivating factors. Perspectives Fortifying automated trading with ethics and quality. 50 Author Guidelines 47 Standards Outlook 50 Measure for Measure 52 Quality in the First Person 54 Career Corner 72 Back to Basics Clarifying some confusion over ISO 9001 conformance. How environmental conditions can impact tests and equipment. Dealing with diabetes. Understanding stakeholder management. Finding different perspectives through DMAIC. 2013 Software showcase & directory p. 56 To learn more about the manuscript review process, helpful hints before submitting a manuscript and QP’s 2013 editorial planner, click on “Author Guidelines” at www. qualityprogress.com under “Tools and Resources." Photocopying Authorization Authorization to photocopy items for internal or personal use or the internal or personal use of specific clients is granted by Quality Progress provided the fee of $1 per copy is paid to ASQ or the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. Copying for other purposes requires the express permission of Quality Progress. For permission, write Quality Progress, PO Box 3005, Milwaukee, WI 53201-3005, call 414-272-8575 x7406, fax 414-272-1734 or email reprints@asq.org. Photocopies, Reprints And Microform Article photocopies are available from ASQ at 800-248-1946. To purchase bulk reprints (more than 100), contact Barbara Mitrovic at ASQ, 800-248-1946. For microform, contact ProQuest Information and Learning, 300 N. Zeeb Road, Ann Arbor, MI 48106, 800-5210600 x2888, international 734-761-4700, www.il.proquest.com. Membership and Subscriptions NEXT MONTH - performance review The plan-do-check-act method to improve 360˚ surveys. - Reinventing excellence How Corning stays on the cutting edge. ASQ’s Vision: By making quality a global priority, an organizational imperative and a personal ethic, the American Society for Quality becomes the community for everyone who seeks quality technology, concepts or tools to improve themselves and their world. Quality Progress (ISSN 0033-524X) is published monthly by the American Society for Quality, 600 N. Plankinton Ave., Milwaukee, WI 53203. Editorial and advertising offices: 414-272-8575. Periodicals postage paid at Milwaukee, WI, and at additional mailing offices. Institutional subscriptions are held in the name of a company, corporation, government agency or library. Requests for back issues must be prepaid and are based on availability: ASQ members $15 per copy; nonmembers $23 per copy. Canadian GST #128717618, Canadian Publications Mail Agreement #40030175. Canada Post: Return undeliverables to 2835 Kew Drive, Windsor, ON N8T 3B7. Prices are subject to change without prior notification. © 2013 by ASQ. No claim for missing issues will be accepted after three months following the month of publication of the issue for domestic addresses and six months for Canadian and international addresses. Postmaster: Please send address changes to the American Society for Quality, PO Box 3005, Milwaukee, WI 53201-3005. Printed in USA. 4 QP • www.qualityprogress.com For more than 60 years, ASQ has been the worldwide provider of information and learning opportunities related to quality. In addition, ASQ membership offers information, networking, certification and educational opportunities to help quality professionals obtain practical solutions to the many problems they face each day. Subscriptions to Quality Progress are one of the many benefits of ASQ membership. To join, call 800-248-1946 or see information and an application on p. 33. List Rentals Orders for ASQ’s member and nonmember buyer lists can be purchased by contacting Michael Costantino at the Infogroup/Edith Roman List Management Co., 845-731-2748 or fax 845-620-9035. upfront QP QUALITY PROGRESS Means to Motivate Contribution proves to be key component Executive Editor and Associate Publisher Seiche Sanders Associate Editor Mark Edmund Assistant Editor WHAt motivates people most in their work lives? Would it surprise you to learn Amanda Hankel it’s not money? Recent research reveals that people are motivated most by their ability manuscript Coordinator to make steady progress on meaningful work. Daniel Pink, a keynote speaker at ASQ’s World Conference on Quality and Improvement, talked about this concept at the May event. (For a brief recap on the conference turn to page 17 or visit http://asq.org/blog.) By nature, quality professionals are eliciting positive changes within their organizations all the time. Immediate gratification, however, is what’s sometimes lacking. That is not the case for the organization featured in this month’s cover story, “To All Corners of Kenya,” p. 18. The organization, Kenya Pharma, delivers much-needed, affordable HIV/AIDS medication to a network across the African country of Kenya. The work, as is stated in the article, is truly a matter of life and death. The ISO 9001-certified Kenya Pharma project is funded by the U.S. Agency for International Development and implemented by Chemonics International. Its quality management system (QMS) spans the entire business cycle. The movement toward a QMS and subsequent certification has allowed the company to make great strides in the number of customers they serve and how well. “On a daily basis, ISO 9001 helps ensure problems that do arise in the supply chain are systematically documented and addressed not only to correct an immediate issue, but also to gather data on larger trends impacting the project’s ability to serve its customers through the corrective and preventive action process,” according to the article. Read about how quality really is making a difference in this insightful case study. Valerie Ellifson contributing EDITOR Megan Schmidt COPY EDITOR Susan E. Daniels Art Director Mary Uttech Graphic Designer Sandy Wyss Production Cathy Milquet Advertising production Barbara Mitrovic Digital Production specialisTS Laura Franceschi Julie Schweitzer Media sales Naylor LLC Lou Brandow Krys D’Antonio Norbert Musial Rob Shafer Donnie Tuttle (ASQ TV) Media sales Administrator Kathy Thomas Also in this issue, help with handling inconsistency with parts suppliers. If you’ve Marketing Administrator ever struggled to get what you want, when you want it, be sure to read “The Power of Matt Meinholz Prediction,” p. 26. Finally, in “Getting a Good Read,” p. 40, read about how libraries Editorial offices reengineered their processes to better serve customers in the digital age. QP Phone: 414-272-8575 Fax: 414-272-1734 Advertising offices Phone: 866-277-5666 ASQ administration CEO Paul E. Borawski Seiche Sanders Editor Don’t forget to watch the new episode Managing Directors Julie Gabelmann Brian J. LeHouillier Michelle Mason Laurel Nelson-Rowe To promote discussion of issues in the field of quality and ensure coverage of all responsible points of view, Quality Progress publishes articles representing conflicting and minority views. Opinions expressed are those of the authors and not necessarily of ASQ or Quality Progress. Use of the ASQ logo in advertisements does not necessarily constitute endorsement of that particular product or service by ASQ. June 2013 • QP 5 logon +Seen&Heard Statistical ignorance StayConnected Find the latest news, quips and targeted content from QP and its staff One additional caution to add to The March 2013 Statistics Roundtable Anderson-Cook’s article is to avoid the column by Christine M. Anderson-Cook, tendency to repeat past plans without “Let’s Be Realistic,” (pp. 52-54) makes a adjustment for the current situation. I number of important points, but misses specifically mean using past test plans, QUAlITY PRogRESS | MARCH 2013 P Putting Best Practices to Work www.qualityprogress.com | March 2013 5 Takeaways From the Revised iSo 19011 p. 38 ISO 22000 QualitY PRoGR the Right FooD SAFETY ingredients Methods for making food safer Quality culture Risk management FDA Good manufacturing practices Plus: Closing down the hidden factory p. 50 HACCP Food Safety Modernization Act Creating the perfect form p. 64 VolUME 46/nUMBER 3 one. I fully agree sample sizes and measures because that decision “that is what we did before.” Each making is the key product, production line and situation is to the value of different and should have assumptions, statistical tools. conditions and analysis checked to be The intangible certain the analysis and recommenda- of slot machine tions are valid. Sure, it is quicker to payouts is similar follow the same steps and repeat past to the question analysis rather than think through the Publisher Seiche Sanders: @ASQ_Seiche Associate Editor Mark Edmund: @ASQ_Mark Assistant Editor Amanda Hankel: @ASQ_Amanda Contributing Editor Megan Schmidt: @ASQ_Megan editor@asq.org www.facebook.com/ groups/43461176682 www.linkedin.com/groups/qualityprogress-magazine-asq-1878386 we hear in business: “What is keeping problem again. Many would rather avoid you up at night?“ Many decisions we opening their statistics books and think- must make in business have intangible ing hard about the underlying assump- qualities. Brand image is one example. tions. Yet, it is where the value of statisti- Sorting out what is and is not part of the cal analysis lies. We owe it to decision statistical analysis story often helps to makers to do a proper analysis, check frame the issue under discussion. assumptions, apply the right techniques, QP popular state the limitations and encourage an (All URLs case sensitive) understanding of statistics. 1.Crunching the Numbers: How the 2012 Salary Survey data were compiled. (http://bit.ly/ crunchnumbers) One of the key issues I see facing those of us active with statistical thinking is the ignorance of most of The current climate is prime for our colleagues. While ASQ provides expanding the use and understanding a lot of information, the content may of statistical methods. We (statisti- need to focus more on the value of cally thinking folks) have the skills that proper statistical thinking to encourage are in demand. We can add value with application. The Wall Street Journal ran our knowledge. I encourage all to read an article titled, “Data Crunchers Now Anderson-Cook’s article again, then look the Cool Kids on Campus,” extolling the for opportunities to apply your statistical value businesses are finding in statisti- knowledge to improve decision making. Fred Schenkelberg Los Gatos, CA cians. The American Statistical Asso1 ciation and others have declared 2013 as the International Year of Statistics. Consider an online basic statistics class I’m currently taking though edX that has 47,000 students enrolled.2 It seems that we (the business world at large) are finding statistics of value. 6 Executive Editor & Associate QP • www.qualityprogress.com Reference and note 1.Carl Bialik, “Data Crunchers Now the Cool Kids on Campus,” Wall Street Journal, March 1, 2013, http://bit. ly/datacrunchers. 2.edX is a nonprofit enterprise of its founding partners Harvard University and the Massachusetts Institute of Technology that features learning designed specifically for interactive study via the web. ’s Most articles 2.Corrective vs. Preventive Action: Do you know the difference between corrective and preventive action? (http://bit. ly/correctvsprevent) 3.Learning to Fish: The fishbone diagram can help you achieve career growth. (http://bit.ly/ learningtofish) 4.Paving the Way: Seven data collection strategies to enhance your quality analyses. (http:// bit.ly/pavingtheway) QP QP Online Extras@ QUALITY PROGRESS www.qualityprogress.com PAST CHAIR • Life-saving improvements Listen to Tiffany Darabi, author of “To All Corners of Kenya,” p. 18, talk more about the project that helped deliver HIV/AIDS drugs in Kenya more efficiently by improving supply chain activities in this month’s Author Audio interview. James J. Rooney, ABS Consulting, Global Government Division CHAIR John C. Timmerman, Gallup Inc. CHAIR-ELECT • Volviendo a los Fundamentos Stephen K. Hacker, Transformation Systems International Read this month’s Back to Basics column translated in Spanish. TREASURER Chava Scher, RAFAEL—Advanced Defense Systems (retired) Quick Poll RESULTS Each month at www.qualityprogress.com, visitors can take an informal survey. Here are the numbers from last month’s Quick Poll: What poses the greatest risk to today’s organizations? • Market uncertainty. • Data management. • Technology. PARLIAMENTARIAN Karla Riesinger, ASQ DIRECTORS 44.4% 36.1% 19.4% Visit www.qualityprogress.com for the latest question: Are you on Twitter? • Yes, I actively tweet. •No, but I’d be open to trying it. • I plan to set up an account soon. • I’ll never tweet. QualityNewsToday Recent headlines from ASQ’s global news service (All URLs case sensitive) Behind the Scenes: Boeing’s Dreamliner Battery Fix It was a logistical effort Boeing had never faced before: simultaneously moving a small army of technicians to 13 international locations, transporting 15 tons of tools per repair kit and installing newly designed equipment in the field, taking five days per airplane and working around the clock in two 12-hour shifts. (http://bit.ly/boeingbehind scenes) London to Create Airport of the Future London City Airport wants to eradicate the nuisances associated with flying by creating a streamlined experience with the help of a technology known as the Internet of Things. Essentially, it is an overarching platform that allows a variety of devices or machines to communicate with one another. (http://bit.ly/airportoffuture) Heather L. Crawford, Apollo Endosurgery Raymond R. Crawford, Parsons Brinckerhoff Ha Dao, Emerson Climate Technologies Inc. Gary N. Gehring, Saskatchewan Ministry of Governmental Relations Kathleen Jennison Goonan, M.D., Goonan Performance Strategies Eric A. Hayler, BMW Manufacturing Co. James M. Loseke, Sargento Foods Inc. Joanne D. Mayo Elías Monréal, Industrial Tool Die and Engineering Richard A. Perlman, Bayer HealthCare Steven J. Schuelka, Calumet College Daniel E. Sniezek, Lockheed Martin (retired) G. Geoffrey Vining, Virginia Tech Department of Statistics Alejandra Vicenttin, Alejandra Vicenttin Advisors Bharat Wakhlu, Tata Services Ltd., a division of Tata Sons J. Eric Whichard, JE Whichard and Associates QP EDITORIAL REVIEW BOARD Randy Brull, chair Administrative Committee Brady Boggs, Randy Brull, Jane Campanizzi, Larry Haugh, Jim Jaquess, Gary MacLean, R. Dan Reid, Richard Stump Technical reviewers Andy Barnett, David Bonyuet, David Burger, Bernie Carpenter, L.N. Prabhu Chandrasekaran, Ken Cogan, Linda Cubalchini-Travis, Ahmad Elshennawy, Mark Gavoor, Kunita Gear, Daniel Gold, T. Gourishankar, Roberto Guzman, Ellen Hardy, Lynne Hare, Ray Klotz, Tom Kubiak, William LaFollette, Pradip Mehta, Larry Picciano, Gene Placzkowski, Tony Polito, Peter Pylipow, John Richards, James Rooney, Brian Scullin, Amitava Sengupta, A.V. Srinivas, Joe Tunner, Manu Vora, Keith Wagoner, Jack Westfall, Doron Zilbershtein Want the latest quality-related news and analysis? The QNT Weekly e-newsletter delivers it every Friday. Subscribe now at http://email.asq.org/subscribe/qntwk. June 2013 • QP 7 expertanswe Doing the right thing Q: During an audit closeout meeting, what ing a finding on behalf of their function. • The finding is wordy, poorly written or is the right thing to do if the auditee refuses unclear about what to address or im- of those involved in the audit. However, with to accept a correct nonconformance report prove, so the auditee cannot understand awareness of the most common root causes, (NCR) by the lead auditor? Should the lead what the real gap is. an auditor will be equipped to avoid most auditor submit his or her report with or without the disputed NCR regardless? What if the auditee—such as a director or CEO—interprets the case according to his or her opinion and sees there isn’t an NCR? Ayad Mosab Al-Ani Salalah Free Zone, Oman • NC statements appear to be opinionbased rather than fact, failing to provide sufficient evidence and references. disagreements over NCs. Obviously, an auditor’s soft skills, including tact, reasoning and persuasion, are important. • An opportunity for improvement is mis- Having management commitment from the classified as an NC. Inconsistent use of auditing and auditee organizations, following audit terminology and definitions is often a auditing protocol, adequately planning and rookie mistake. preparing for the audits, offering solid value- • The same finding is unnecessarily added feedback, promptly verifying corrective A: If your nonconformance (NC) is factual, has repeated. An auditor who identifies a actions, linking audit action effectiveness to business impact and appropriately references document control issue and counts sepa- the organizational bottom line, and presenting a standard or specification, you should not rate NCs for the multiple areas in which it the audit’s effectiveness to management will drop the finding. appears, for instance, will be perceived as also help ensure a healthy audit management It appears from the question that the being interested in generating a number process. auditee may not have been aware of the NC of findings rather than providing valuable until the closing meeting. Don’t keep an NC a surprise until the closing meeting. If the audit feedback. • Issues unrelated to the audit are Even if the auditee is a high-profile individual, such as a director or CEO, the auditor should stick to the task. During my audits, if I is a multi-day event, you should discuss any bundled with the findings. More often am convinced of the importance of a finding, I NCs with the auditee as you find them, ideally than you would expect, auditors incorpo- explain it to the auditee, outlining the benefits during daily caucus meetings with the auditee. rate known issues for the audit area into of addressing the gaps and the risks of not Discussing NCs before the closing meeting the current audit report, using the audit as addressing the gaps, and provide references will provide the auditee an opportunity to leverage to get things done. There is noth- to support the NC. I also give the auditee the clarify the situation or produce evidence. ing wrong with verifying through samples opportunity to explain any disagreement. If Why would an auditee refuse to accept an whether past issues are still potent and the reason for objection is not supported by NCR? The most common root causes are: reporting them with evidence. Bundling facts and evidence, I will let the finding remain, • The audit finding is outside the scope past issues without verification, however, and the auditee will have to respond to it. may cause auditee refusal. Management should be counted on to respect of the contract or audit. In the case of supplier audits, the finding goes beyond • There is no sign of continued or future the agreed-on purchasing requirements.1 In business. A particularly difficult supplier the case of internal audits, the finding may audit scenario can occur when an existing be beyond the responsibility of the auditee contract is near expiration and there has function. been no word on extension while the • Addressing the NC would not add value, such as recreating missing paperwork that has no value for the business. • The auditee is not empowered to ac- 8 Other reasons will likely be as varied as the personalities, character traits and capabilities customer is auditing and issuing findings. • Hostility between the auditor and auditee creates a strained, “gotcha” mentality. • The auditor bullied the auditee into ac- cept the findings. Auditees who are too cepting a finding rather than convincing junior in the organization’s hierarchy may the auditee through careful and thorough worry about the consequences for accept- reasoning. QP • www.qualityprogress.com the integrity and value of the audit process, upholding findings that are genuinely offered. Govind Ramu Director, quality assurance SunPower Corp., San Jose, CA reference 1.Govind Ramu, “Expert Answers: Information Collection,” Quality Progress, March 2013, pp. 8-9. Bibliography Russell, J.P., “Exit Meeting,” chapter in The ASQ Auditing Handbook, third edition, ASQ Quality Press, 2005. Russell, J.P. and Terry Regel, “Chapter 1: Reporting Problems,” After the Quality Audit: Closing the Loop on the Audit Process, second edition, ASQ Quality Press, 2000. rs Keeping customers happy determined, they can be translated—using by changes, running pilots, selecting team Q: I’m developing a customer-oriented tools such as critical-to-quality analysis—into members who will participate, ensuring the system for updating a service operation that appropriate customer satisfaction metrics for project has a sponsor and communicating to allows managers to gather information, make each category or segment. all stakeholders. changes in service delivery and assess the More than likely, your customer satisfac- Your organization may have a preferred effects of these changes. How do I imple- tion metrics will involve performance around implementation approach that is already set ment that in banking? What is the best way on-time delivery, quality and subjective mea- up with software and a database. It would be to measure customer satisfaction based on sures of performance, such as friendliness, wise to make a few inquiries at the outset. this approach? Which quality system should product knowledge and helpfulness. Which quality system? I use? How do I develop a system framework for updating these services? Mahmoud Madgy Dubai, United Arab Emirates A: You have asked quite a few questions that Achieving satisfaction levels Again, without knowledge of your organiza- Next, you must determine if your internal tion, it is difficult to recommend a specific processes help you meet all of the require- quality system. Actually, a more basic ques- ments and how capable they are of achieving tion is, “Why do you feel the need to operate the required levels of customer satisfaction. within a quality system?” will be difficult to answer in the short space Where there are deficiencies in meet- If your organization has already imple- allotted. Perhaps I can, at least, point you in ing requirements, processes must be either mented a quality system, perhaps there is a the right direction. modified or new processes put in place. natural fit for your service delivery system Where processes are not capable, improve- within that quality system. If so, I suggest you ments must be made. explore that avenue. The framing of your questions indicates the focus of your thinking is on the customer and you are intent on building a closed- Don’t be surprised to find gaps in require- Otherwise, if no such system exists, you loop system. This approach will provide a ments and processes not capable. In my might want to complete your service delivery solid foundation for whatever method you experience, many organizations never origi- project before addressing a quality system develop. nally designed and built their service delivery project. That way, you’ll be able to complete processes with the intent of supporting cus- your project in the event the organization Measuring satisfaction tomer satisfaction. Now that service delivery does not have an appetite for taking on a Several important considerations should be is recognized as an important component of quality system project. More specifically, if addressed before you can begin to measure customer satisfaction and a key differentia- you must expand the scope of your service customer satisfaction. These include: tor—particularly in high transaction environ- delivery project to include the building of a • Categorizing and segmenting your cus- ments—these processes are now receiving quality system, you may never complete the the attention they are due. service delivery project. Keep the project tomer base to better identify and align customer requirements. • Establishing a set of independent listening scopes distinct and manageable. Implementation approach T. M. Kubiak posts for each customer category or seg- Without knowledge of your organization, it is President ment. difficult to recommend a specific implemen- Performance Improvement Systems • Analyzing customer data obtained through the listening posts to determine tation approach. However, there are basic elements of customer requirements. most implementations that I suggest you After the customer requirements are consider, such as: involving those affected Asked and Answered Submit your question at www.qualityprogress.com, or send it to editor@asq.org, and our subject matter experts will help you find a solution. Weddington, NC Bibliography Kubiak, T. M., The Certified Six Sigma Master Black Belt Handbook, ASQ Quality Press, 2012, pp. 148-156. Kubiak, T. M. and Donald W. Benbow, The Certified Six Sigma Black Belt Handbook, second edition, ASQ Quality Press, 2009, pp. 66-65. Shipley, David, Carl W. Keller, James Bossert, Steven S. Prevette, Duke Okes, Dale Crownover and T. M. Kubiak, “Multiple Choice—What’s the Best Quality System?” Quality Progress, July 2003, pp. 25-45. June 2013 • QP 9 QUALITY COUNCIL OF INDIANA CQIA PRIMER C CSSBB PRIMER C CQE PRIMER C CQT PRIMER C PRIMERS CSQE PRIMER C CMQ PRIMER C CRE PRIMER C LSS PRIMER C CQI PRIMER CCT PRIMER C C CQA PRIMER CSSGB PRIMER C CQPA PRIMER C C Our Primers contain study material for the current ASQ bodies of knowledge plus sample questions and answers. The Primers may be taken into the exam. The completeness of our materials makes them the most widely used texts for Certification Training. SOLUTION TEXTS Quality Council of Indiana offers detailed solutions to all questions presented in the corresponding Primer. CD-ROMS QCI offers user-friendly interactive software to assist students preparing for ASQ examinations. Each CD contains 1000 total questions. Examinations are timed and summarized graphically. A help file provides explanations and references. The CDs run on Windows XP and newer. CQE CSSGB CQA Mail Orders Information Fax Orders Quality Council 812-533-4215 812-533-4216 of Indiana Telephone Orders Order Department 800-660-4215 602 W. Paris Ave. Internet Orders W. Terre Haute, IN 47885-1124 www.qualitycouncil.com Juran’s Quality Handbook Implementing Six Sigma The Quality Technician’s Handbook 6th Edition 2nd Edition 6th Edition The essential quality reference for most ASQ exams A great CSSBB reference Great for CQT and CQI exams. by Forrest W. Breyfogle, III by Juran & De Feo En Español CSSGB PRIMER by Gary K. Griffith Spanish Green Belt Quality Dictionary The Spanish version of the CSSGB Primer. by Tracy Omdahl LSS Primer RAM Dictionary More than 2500 definitions. Great for any ASQ certification. The Lean Six Sigma Primer is written to a QCI BoK. There are more case studies and lean content than in any other QCI products. 400 questions are included. A solution text is also available. by Tracy Omdahl Contains 2800 definitions. Helpful for Reliability and Quality Engineers. ISO 9001 Internal Auditing Primer ISO Primer by Bensley & Wortman by Greg Wies & Bert Scali A convenient book for training internal auditors to the ISO 9001 expectations. An instructor CD is available. Reliability & Maintenance Analyst CD by Bryan Dodson Solve your Weibull, reliability, warranty, Bayesian & Maintenance, prediction & estimation problems. Measurement Analyst CD Performs all measurements required in the AIAG manual. Contains ANOVA methods and excellent graphs. Site and global license available!!! Used by Chrysler, ITT, FedEx, Ford, TRW, GM, HP, U.S. Postal Service ISO Presents a thorough treatment of the ISO implementation and documentation process. There are generic manuals on the CD. Quality System Handbook by Edenborough QSH Details the selection, organization, and writing of quality documents. The disk contains procedures and work instructions. PERSPECTIVES BY Andrew Kumiega, Michael Davis and Ben Van Vliet Bank on It Ethics and quality will strengthen automated trading and finance AUTOMATED TRADING, including high- ers, mathematicians (usually called quants) ness implies irresponsibility or imprudence. frequency trading, now accounts for more and computer engineers. Each specialty But what constitutes prudence, or good than 70% of daily trading volume.1 Though has well-defined ethical standards. How- faith, in automated trading system research not everyone agrees, most academic studies ever, like most new industries, the ethics at and development? This is where ordinary support the claim that automation benefits the intersections or interfaces among these regulation ends and the creation of new ethi- the overall functioning of society and mar- groups is still evolving. The intersection of cal codes for automated trading begins. kets.2-5 Automated trading tends to mitigate the three professions gives rise to new ethi- volatility, add liquidity and reduce transac- cal issues at the organizational level. Overregulation may stifle competition, forcing a one-size-fits-all solution to such problems. Because self-regulation is gener- tion costs for all market participants. A matter of ethics ally thought to be more cost-effective than occur every day without mistakes. The In the age of automation, financial innova- governmental regulation, the financial in- proportion of incorrect trades is small. tions—new strategies, methods of analysis, dustry long ago established self-regulatory However, automated trading systems oc- execution venues, instruments, technolo- organizations, such as exchanges and the casionally do have quality problems due to gies and data sources—are the ultimate Financial Industry Regulatory Authority, as various types of errors. These occasions source of competitive advantage. How is proxy regulators of their markets and mem- have forced the industry to take stock of its this innovation to be kept as socially useful bers. Yet, even exchange regulations cannot responsibilities to its shareholders, other as possible? Given innovation’s rapid pace, cross national borders without significant market participants and society. neither governmental nor exchange regula- international coordination. Only ethics can Consider what happened to Knight Capi- tion can keep up. One of the problems—as provide general principles and specific rules tal last August. The firm lost $440 million in John Boatright, professor at Loyola Univer- that cross national borders, market cultures less than 45 minutes when its new trading sity in Chicago, points out— is that “people and even professional disciplines. software flooded the market with orders, (in finance) believe that the task of creating almost bankrupting the firm.6 Even such and enforcing ethical rules and standards exchanges (for example, NASDAQ, New rare events may reduce investors’ appetite is the job of legislatures and regulators, York Stock Exchange and CME Group Millions of computer-generated trades As enablers of automated trading, the not themselves.”7 Inc.) and even the off-exchange execution No code, public or venues, consider their ethical responsibili- private, addresses ties to include constructing fair markets, overall ethical helping to sustain confidence in them. responsibilities in As a starting point for fixing this problem, automated trading many authors point to the basic premise of of the professional Isaac Asimov’s Three Laws of Robotics: “a groups that create robot may not harm humanity, or, by inac- and operate these tion, allow humanity to come to harm.”8 automated systems. Regulators are beginning to address One problem for ethics in automated trading and finance is that market outcomes (as with all outcomes) are uncertain for risk and thereby may cause the average these ethical issues. For example, while the and ambiguous. Yet, the best way to pro- investor to not invest in the stock or futures Securities and Exchange Commission and duce the social benefits of free markets is market. These rare events also invite the the Commodity Futures Trading Commission to foster market agents with a robust ethi- public to view financial markets as casinos, have long prohibited market manipulation, cal framework that combines the ethics of not capital allocation mechanisms. they have recently lowered the bar to prov- the various groups. An industrywide ethical ing it from intent to recklessness. Reckless- climate can determine how organizations Automated finance brings together trad- 12 QP • www.qualityprogress.com and professionals perceive problems, has been prudent and that its automated obligations to external stakeholders, and trading systems function properly, are op- techniques can provide a structure for un- formal standards of right and wrong. erationally safe, and are robust enough to derstanding and solving the interdisciplin- behave acceptably during extreme events. ary ethical problems. Quality techniques Quality and standards Of course, the ability of automated As a starting point, quality management then can be employed to find solutions To improve the ethical interface between trading and finance firms to prove the for root causes rather than (the industry’s professions, other industries have turned stability of their systems depends on the current mode) containment of individual to quality management standards. Quality availability of execution-venue-simulation outcomes. is not only a software testing or statistical facilities to fully test those systems. Such framework, but also an ethical one (inso- simulation facilities should provide testing tures for organizational responsibility, far as the standards in question are ones against all manner of extreme market and regulators can assess the rigor of research, everyone in the relevant disciplines—at infrastructure events. development, operations and control with- their rational best—want everyone else Following ISO 9000-style standards tes- Given formal decision-making struc- out infringing on the secrecy of propri- in those disciplines to follow, even if that tifies to an automated trading firm having etary trading strategies and their sources would mean having to do the same). satisfied its obligations to prove and docu- of competitive advantage. QP To be sure, many trading firms, execution venues and third-party vendors of trading system components perform ment trading strategies and technologies that operate safely, reliably and profitably. Such standards would mean the finan- rigorous quality control over their pro- cial industry is taking the lead in improv- cesses—not strictly out of ethics, but from ing its own management systems. The fed- an understanding that a quality approach eral government then should be less likely to innovation is, in the end, the source of to issue prescriptive regulations, allowing competitive advantage. However, even it to concentrate its surveillance efforts on these systems can fail at the interfaces the integrity of the trading organization as because methodology has not been coordi- opposed to the details of individual trad- nated across professions. ing systems. (That is how regulators have Quality processes operationalize (in responded to the aerospace industry’s part, at least) an organization’s ethics. In AS9100 standard.) Likewise, global indus- the automated trading and finance indus- try standards should present significant try today, discussion of quality typically opportunities for regulatory harmoniza- focuses on specific guidance for improve- tion across national boundaries. ment of trading systems within the professional silos. But beyond simple software Framing ethical problems testing, quality in finance can include To mitigate the possibility of a breakdown all professionals—computer engineers, in the global trading mechanism, the quants and traders alike. automated finance-trading industry needs The ISO 9000 family of standards—and a global ethical framework. Properly fram- its industry-specific derivatives—are the ing the ethical problems will enable root most widely recognized standards for a cause analysis and long-term fixes, rather quality management system (QMS). In than short-term regulatory containments other industries (where public safety must of undesirable outcomes. A QMS standard, be ensured), ISO standards define how such as the (ISO 9000-style) AT 9000 proj- firms ought to do business in, for example, ect we are working on, should enable the aerospace, chemicals, medical devices and industry to address professional, organi- food safety. zational and industrywide responsibilities. A similar standard for automated trad- A broad ethical framework must ensure ing and finance should include software prudence without infringing on the neces- and hardware testing that proves the firm sary independence of trading firms. REFERENCES 1.Jonathan A. Brogaard, “High Frequency Trading and Its Impact on Market Quality,” Northwestern University, 2010, www.futuresindustry.org/ptg/downloads/hft_trading.pdf. 2.Boyan Jovanovic and Albert Menkveld, “Middlemen in Limit-Order Markets,” New York University, 2011, http:// papers.ssrn.com/sol3/papers.cfm?abstract_id=1624329. 3.Terrence Hendershott, Charles Jones and Albert Menkveld, “Does Algorithmic Trading Improve Liquidity?” Journal of Finance, 2011, Vol. 66, No. 1, pp. 1-33. 4.Terrence Hendershott and Ryan Riordan, “Algorithmic Trading and Information,” NET Institute Working Paper No. 09-08, 2011, www.im.uni-karlsruhe.de/upload/ publications/218bfa37-798e-44df-8ad3-8289af8a0920.pdf. 5.Frank Fabozzi, Sergio M.Focardi and Caroline Jonas, “HighFrequency Trading: Methodologies and Market Impact,” Review of Futures Markets, special issue, Vol. 9, 2011, pp. 7-38, www.conatum.com/presscites/hftmmi.pdf. 6.Jenny Strasburg and Jacob Bunge, “Loss Swamps Trading Firm,” Wall Street Journal, Aug. 2, 2012, http://online.wsj. com/article/SB100008723963904438664045775647720839 61412.html. 7.John Boatright, ed., “Ethics in Finance,” Finance Ethics, John Wiley & Sons, 2010. 8. Isaac Asimov, “Runaround,” Astounding Science Fiction, March 1942. ANDREW KUMIEGA is the director at a proprietary trading firm. He is also an adjunct professor at the Illinois Institute of Technology (IIT) in Chicago. He holds a doctorate in industrial engineering from the University of Illinois in Chicago. An ASQ member, Kumiega is a certified quality engineer, a quality auditor and a software quality engineer. He is the co-author of Quality Money Management (Academic Press, 2008). MICHAEL DAVIS is senior fellow at the Center for the Study of Ethics in the Professions and professor of philosophy at IIT. He earned a doctorate in philosophy from the University of Michigan in Ann Arbor. Davis has published more than 200 articles and chapters and authored seven books. His most recent book is Code Writing: How Software Engineering Became a Profession (Center for the Study of Ethics in the Professions, 2009). BEN VAN VLIET is an assistant professor at IIT’s Stuart School of Business in Chicago. He holds a doctorate in management science from IIT. He is an author or co-author of Quality Money Management (Academic Press, 2008), Building Automated Trading Systems (Academic Press, 2006) and Modeling Financial Markets (McGraw-Hill, 2004), and has published several articles in the areas of finance and technology. June 2013 • QP 13 keepingcurre automotive A New Path Forward A Toyota details strategy to improve design and production approach After battling four years of setbacks—including global financial crisis, worldwide recalls, an earthquake and a tsunami—Toyota Motor Corp. in Japan, the company that brought us the Toyota Production System and the concept of lean manufacturing, is once again introducing a new approach. In late April, the organization’s president, Akio Toyoda, provided details to Automotive News about an overhauled R&D strategy he hopes will help lift the automaker to sustained growth and improve product development efficiencies by 20 to 30%. Since taking over as CEO in 2009, Toyoda’s focus has been to make the organization less susceptible to the external risks that have plagued the carmaker during the past few years. The new method is called the Toyota New Global Architecture (TNGA), and it focuses on simultaneous development of multiple models and massive use of common parts. The ultimate goal is to boost global sales to 10 million vehicles per year—a feat no carmaker has accomplished to date. “If you compare it to soccer, it’s like changing the field formation,” Toyoda said. “As the head coach, I have changed us from a defen- Mr. Pareto Head By Mike Crossen 14 QP • www.qualityprogress.com sive formation to an offensive one.” While Toyota announced it was undertaking a new product development strategy last year, this was the first time details had been discussed. The first cars built using TNGA will debut in 2015. By increasing product development efficiency, time and money savings will be reinvested into new technology and designs, better manufacturing processes and enhanced quality, said Mitsuhisa Kato, executive vice president and global R&D chief. Toyota will use more components made to global standards rather than to proprietary specifications, which should help in worldwide procurement from large suppliers and reduce costs. Initially, cars under the same platform will share 20 to 30% of their parts, but Kato said the eventual goal is to commonize 70 to 80% of parts. “Commonization is not new,” Kato said. “But we are trying to push it to a higher level.” Additional efforts to maximize efficiencies include grouping future vehicles by launch timing and engineering them simultaneously to take advantage of commonalities. Before, according to the Automotive News article, planning was based on individual model without considering the big picture. To speed R&D processes, in April the company reorganized into four parts—Lexus, Toyota-brand vehicles for mature markets, Toyota-brand vehicles for emerging markets and the Unit Center, which conducts R&D on shared core technologies, such as engines and transmissions. Also in April, Toyota established new R&D facilities at its Toyota City headquarters. The facilities are designed to enhance collaboration among powertrain, prototyping and production engineers. The facilities are also equipped with a new aerodynamics lab and a wind tunnel with a maximum speed of 155 mph. The wind tunnel uses a moving belt under the cars that better simulates airflow around a moving vehicle than if it were sitting on a stationary platform. With these improvements underway, Toyoda is confident his organization can get back on its feet, achieve its goals and sustain its tradition of continuous improvement. BIBLIOGRAPHY Automotive News, “Toyota’s 10 Million Plan,” May 3, 2013, http://asq.org/qualitynews/qnt/execute/ displaySetup?newsID=15926. nt manufacturing asq joins coalition focused on manufacturing ASQ has joined the Smart Manufacturing Leadership Coalition (SMLC), a public-private partnership developing platforms for experimentation of large development projects, or testbeds, and training for nextgeneration manufacturing technology platforms. The SMLC was launched last July to drive improved manufacturing efficiencies, reduce waste and integrate manufacturing intelligence in real time across an entire production operation. ASQ, which has been indirectly involved in SMLC initiatives for months, recognized that SMLC efforts aligned well with ASQ’s longstanding role in manufacturing. ASQ said it hopes its subject matter experts can help the coalition develop workforce training, tools and innovations to implement breakthrough smart manufacturing technologies. “The Smart Manufacturing Leadership Coalition has its goals rooted in quality, which is helping companies perform more efficiently with higher quality outcomes,” said ASQ CEO Paul Borawski. “We are excited for the opportunity to share our resources with the SMLC, accentuating quality through the development and growth of the organization.” “ASQ’s extensive knowledge in helping organizations increase efficiencies will be an asset to the Smart Manufacturing Leadership Coalition as we continue to improve manufacturing productivity,” said SMLC Chairman Dean Bartles. “We are pleased to have ASQ as part of the SMLC, as both organizations share the passion for improvement.” As a member of the SMLC, ASQ joins companies such as Alcoa, General Motors, Honeywell, Pfizer and Rockwell Automation. asqnews CALL FOR PRESENTATIONS ASQ has issued a call for presentations for the 2014 World Conference on Quality and Improvement, which will be held May 6-7 in Dallas. The submission deadline is Aug. 2. For more information, visit http://wcqi. asq.org/2014/call-for-presentations/ index.html. TEAM AWARD APPLICATIONS ASQ is looking for teams to participate in next year’s International Team Excellence Award process. The deadline for intent to submit forms is July 1. For more information, visit http:// wcqi.asq.org/team-competition/ timeline.html. ASQ TO HOST TC 69 MEETING ASQ will host the International Organization for Standardization (ISO) annual meeting for members of Technical Committee (TC) 69 on applications of statistical methods June 3-7 in Mil- waukee. Created in 1948, TC 69 has representatives from 48 countries and focuses on standardization in the application of statistical methods, including generation, collection (planning and design), analysis, and the presentation and interpretation of data. For information about ASQ standards involvement, visit www. asq.org/standards. ASQ QUALITY INSTITUTE ASQ is offering a series of introductory quality courses July 15-17 at its headquarters in Milwaukee. The ASQ Quality Institute will feature training not normally available in classroom settings. Courses include quality basics, integrated quality management, auditing for improvement, failure mode and effects analysis, and corrective and preventive action. Visit www.asq.org/ promotional/milwaukee for details on the courses, instructors and special registration offers. ASQ Journal spotlight Every month, QP highlights an openaccess article from one of ASQ’s seven other journals. This month, make sure you read “Quantile-based Optimization of Noisy Computer Experiments With Tunable Precision,” which appeared in February’s edition of Technometrics. ASQ co-publishes Technometrics four times a year with the American Statistical Association. The article, written by Victor Picheny, David Ginsbourger, Yann Richet and Gregory Caplin, addresses kriging-based optimization of stochastic simulators. To access the article, click on the “Current Issue” link on Technometrics’ website: http://asq.org/pub/techno. From there, you also can find a link to information about subscribing to the quarterly publication. June 2013 • QP 15 keepingcurrent Q Who’s Who in NAME: Timothy J. Robinson. RESIDENCE: Laramie, WY. EDUCATION: Doctorate in statistics from Virginia Tech in Blacksburg. CURRENT JOB: Professor of statistics at the University of Wyoming (UW) in Laramie. INTRODUCTION TO QUALITY: A response surface methods course taught by Raymond H. Myers at Virginia Tech and assisting Myers with a short course on experimental design that Robinson delivered at the Ethyl Corp. PREVIOUS EXPERIENCE: Robinson worked as a high school mathematics teacher in vastly different settings—inner city and private school environments—before he decided to attend graduate school. ASQ ACTIVITIES: Robinson is the Statistics Division’s chair for student and early career grants for the Statistics Division’s Fall Technical Conference. OTHER ACTIVITIES: Robinson and his wife, Dawn, operate a small statistical consulting company with contracts with the U.S. Fish and Wildlife Service and the National Park Service. Previous clients have included Merck, Southwest Airlines, Ethyl Corp., the New Zealand Department of Conservation, and the American Samoa Government’s Department of Marine and Wildlife Resources. PUBLICATIONS: Robinson has had more than 40 peer-reviewed manuscripts published. He is also the co-author of Generalized Linear Models With Applications in Engineering and the Sciences on a Generalized Linear Models, second edition (Wiley, 2010). RECENT AWARDS: Robinson was included in the 2012 class of ASQ fellows. He and his colleagues who presented “Multiple Objective Pareto Frontier Optimization for Split-Plot Experiments” received the Shewell Award for the best presentation at last year’s Statistics Division Fall Technical Conference. Last spring, he also won a UW Promoting Intellectual Engagement Award for Classroom Teaching. In 2007, he received the John P. Ellbogen Award for Meritorious Classroom Teaching, UW’s highest award for classroom teaching. PERSONAL: He and Dawn have four children, ages 2 to 12. FAVORITE WAYS TO RELAX: Spending time with family, hunting big game in Wyoming and hiking in the wilderness areas outside of Laramie. 16 QP • www.qualityprogress.com dateinqualityhistory QP occasionally looks back on a person or event that made a difference in the history of quality. June 18, 1926 Philip B. “Phil” Crosby was born in Wheeling, WV. Crosby was a businessman and author who contributed to management theory and quality management practices. He is widely recognized for promoting the concept of zero defects and for defining quality as conformance to requirements. Crosby is credited with reinforcing the message to do things right the first time. He also became well-known for acquainting quality professionals with the following theories: The price of nonconformance is a measure of poor quality; prevention is a means to eliminate quality problems; leadership is a requirement to make progress, teamwork is the principle for work; and customer requirements define the standard of quality performance. Crosby’s first book, Quality is Free, has been credited in part with igniting the quality revolution in the United States and Europe. He wrote 13 books, including The Absolutes of Leadership in 1996 and Quality and Me, an autobiography published in 1999. Crosby was diagnosed with and treated for esophageal cancer in early 2001. A few months later, he developed adult respiratory distress syndrome. His condition worsened, and he died in August 2001. WordtotheWise To educate newcomers and refresh practitioners and professionals, QP occasionally features a quality term and definition: Supermarket A lean term meaning the storage locations of parts before they go on to the next operation. Supermarkets are managed by predetermined maximum and minimum inventory levels. Each item in the plant is at a designated location. SOURCE: “Quality Glossary,” Quality Progress, June 2007, p. 57. ASQ World conference speaker: money isn’t always the only motivator for workers Indianapolis served as the backdrop to ASQ’s 2013 World Conference on Quality and Improvement, held in May. Thousands of quality professionals from nearly 50 countries flocked to the Indiana Convention Center for a jam-packed schedule of educational and networking opportunities, keynote speakers and a bustling trade show floor. Highlights included opening keynote, author Daniel Pink, who explained what motivates people to act, and to do their best work. Guess what? It’s not money. He discussed the findings of a study where results showed that tasks requiring mechanical skills got better results where there was greater monetary incentive offered, but surprisingly, the results were the opposite for tasks requiring “even rudimentary cognitive skill.” Results, in fact, were poorer the more reward that was offered. Pink referenced W. Edwards Deming’s early teachings which addressed this subject: Deming said reward as a motivator squeezes out a person’s intrinsic motivation over the course of their lifetime. Money still motivates, of course. But it’s not the core motivator. “Pay people enough to take the issue of money off the table,” Pink said. “Management is a technology from the 1850s that was designed to get compliance,” he said. “We need some measure of that in the workplace today, but what we need is engagement. If you really want engagement, you can’t manage into it.” Examples of organizations finding ways to create engagement included Facebook, which allows new engineers to choose which team they want to work with, and FedEx, where employees are given a short period of time each week where they can work on whatever they want. It has led to great innovation. Finally, Pink discussed what the great- est motivator is: each day, making progress on meaningful work. Karen Martin, Jamais Cascio, James Melton and Sally Hogshead also gave keynote or featured speaker presentations during the three-day conference. Awards and events Other highlights of the conference included the release of ASQ’s Global State of Quality research and an accompanying panel discussion, with representatives from India, Finland, Germany and the United States. “The ASQ Global State of Quality Research: Discoveries 2013” is the first of three reports that will be released this year. “Discoveries 2013” provides a first view of the data, collected from nearly 2,000 organizations in 22 countries. Future reports—due out in July and in November during World Quality Month— will provide deeper insights and greater context to the findings. Key findings? “There are significant gaps in the governance and management models of quality, and the availability and use of metrics in manufacturing and servicebased organizations worldwide,” according to a press release. “The ASQ Global State of Quality Research: Discoveries 2013” report also shows disparities between manufacturing and services in quality management frameworks and training. Also of note at the conference was the always-engaging ASQ International Team Excellence Awards. Argentina’s Telefonica de Argentina and Thailand’s Pruksa Real Estate Public Company Limited-QCI Team were awarded silver-level status for increasing quality and financial savings. In its 28th year, 32 teams from 10 countries competed for gold, silver and bronze status. No teams in this year’s awards process earned gold-level status. shortruns THE QUALITY FUNCTION DEPLOYMENT (QFD) Institute has issued a call for papers for its 19th International Symposium on QFD, which will be held Sept. 6-7 in Santa Fe, NM. For more information, visit www.qfdi.org/call_for_papers.html. THE INTERNATIONAL ASSOCIATION for Food Protection will hold its 2013 conference July 28-31 in Charlotte, NC. Registration and housing information is now available at: www.foodprotection.org. NOMINATIONS ARE BEING sought for the 2013 National Institute of Statistical Sciences’ Jerome Sacks Award for Outstanding Cross-Disciplinary Research. For more information about the award, visit www.niss.org/news/awards/ jerome-sacks-award-outstanding-crossdisciplinary-research. WORLD ACCREDITATION DAY 2013 will be celebrated around the globe on June 9. This year’s theme—Accreditation: Facilitating World Trade—was selected because it is relevant in developed and developing economies. For more information about the day and ways to join the celebration activities, visit www. ilac.org. CORRECTIONs In an article about his death, the state in which Donald Feigenbaum resided was incorrectly reported in the April 2013 edition of QP (“Donald Feigenbaum Passes Away,” p. 17). Feigenbaum resided in Pittsfield, MA, until he died of heart failure in March. QP regrets the error. In the May 2013 edition of QP, Huawei Technologies Co. Ltd. was inadvertently omitted from the special section devoted to ASQ Enterprise and Site Members (pp. 50-52). Huawei Technologies Co. Ltd. is currently one of ASQ’s Site Members. ASQ apologizes for this oversight. June 2013 • QP 17 To All Corners of Kenya Project ensures a reliable supply chain provides medicines to people with HIV/AIDS In 50 Words Or Less • Kenya Pharma is a $550 million USAID-funded project implemented by Chemonics International, which delivers drugs to people with HIV/AIDS throughout the African country of Kenya. • The project uses an ISO 9001 approach to improve and solidify supply chain activities and ensure medicines are provided to Kenyans affected by the disease. by Tiffany Darabi case study EFFECTIVE SUPPLY CHAINS deliver products on schedule and within budget to customers. Those customers shouldn’t have to think about the behindthe-scenes process or worry about whether the goods will be delivered on time. The ISO 9001 requirements for building a quality management system (QMS) provide a framework for business operations that capture the intricacies of supply chain processes while simultaneously providing a method for improving operations. A prime example of using an ISO 9001 approach to supply chain management can be seen in the ISO 9001-certified Kenya Pharma project, funded by the U.S. Agency for International Development (USAID) and implemented by Chemonics International, a development company based in Washington, D.C. THE SKIPPER OF a taxi boat, along with his young helper, navigate the waters of Lake Victoria to reach Mfangano Island to deliver supplies to the Sena Health Center. (All photos courtesy of Chemonics International) June 2013 • QP 19 Kenya Pharma ensures a reliable supply of high-qual- Setting the stage ity, low-priced pharmaceutical commodities for people Kenya Pharma operates a demand-driven (pull) distribu- with HIV/AIDS in Kenya. The project’s customers are tion system with 169 facilities submitting orders to the children and adults, including pregnant women, from central warehouse. In partnership with DHL-Kenya for some of the poorest and most remote regions in the distribution and with Phillips Healthcare Services Ltd. country. The ability of the project to operate a success- for warehousing, Kenya Pharma effectively engages the ful supply chain is fundamentally about serving people— private sector in the developing world to deliver results. much more than just being a business proposition—because it is quite literally a matter of life or death. Through a hub-and-spoke distribution network, the 169 ordering points supply 626 antiretroviral therapy The decision to implement an ISO 9001-certified (ART) dispensing points, which generally provide a full QMS was driven by the desire to create a systematic ap- range of HIV/AIDS prevention, treatment and care ser- proach for supply chain practices to ensure continual vices. The ordering points also supply an additional 654 improvement of services to the Kenyans served across dispensing points that provide only prevention of moth- the 225,000-square-mile country. er-to-child transmission (PMTCT) services. ISO 9001 certification process / figure 1 Chief of party Participates in meetings with the registrar and ensures that project staff are also available to meet with registrar as needed. Launch (April 5, 2012) Get ISO 9001 certified (July 2012) Management representative Manages schedule of auditor and staff. Process leaders Speak knowledgeably about role as reflected in quality management system (QMS). All staff Speak knowledgeably about role as reflected in QMS. Internal audits (May 21-24, 2012 and at least annually after that) Follow processes. Keep records. Suggest improvements. Chief of party Oversees selection of internal auditors. Oversees audits to ensure they are done according to ISO 9001 requirements. Management representative Ensures training of internal auditors. Schedules and oversees internal audits. Process leaders Make available at least one member of each team to participate in internal audits. Internal auditors Conduct and report audit results and identify improvements. All staff Prepare for and participate in audits. 20 QP • www.qualityprogress.com Management review (June 11-14, 2012, and at least annually after that) Certification assessment/ periodic reviews (Certification assessment occurs every three years; periodic review occurs annually) Chief of party and senior staff Meet at least annually to review data, quality policy and quality objectives. Identify corrective and preventive actions to ensure effectiveness of QMS. Management representative Aggregates continual improvement and internal audit data for analysis and facilitates management reviews. Process leaders Respond to outputs from the management review as needed. All staff Respond to outputs from the management review as needed. case study BOXES CONTAINING HIV/AIDS medicines are loaded onto a taxi boat to ferry them from Mbita Point on Kenya’s mainland to Sena Health Center on Mfangano Island. By the end of 2012, the supply chain was serving more Implementation approach than 349,000 ART patients representing about 58% of the The basics of how to build a QMS and obtain ISO 9001 total treatment population in Kenya. The remaining popu- certification are known across the quality industry: lation is supported mainly by a supply chain run by the • Document what you do—document control. Kenyan government. • Do it—product realization. Kenya Pharma began the process of building its QMS in 2011, obtaining ISO 9001 certification in the summer of 2012 • Keep the proof that you’ve followed what you’ve documented—record control. (see Figure 1). The journey actually began in 2010, howev- • Check to make sure you’re doing what you documented er, when Chemonics International achieved ISO 9001:2008 and examine your work to make improvements and doc- certification for its home office QMS, a unique distinction ument them—measurement, analysis and improvement. amongst international development companies. 1 Chemonics’ QMS spans proposal development to project •Ensure your management team reviews progress, drives improvements and champions the system— management practices—the entire business cycle. Having management commitment.2 an ISO 9001 approach in place at the company’s home of- Chemonics’ home office experiences also helped to fice meant the Kenya Pharma project could build on an es- provide a clear roadmap for implementing an ISO 9001-cer- tablished culture of entrepreneurship, employee ownership, tified QMS. Often, however, the success of an implemen- systemization and continual improvement of best practices. tation will hinge not on technical knowledge, but on the Chemonics currently implements more than 80 inter- softer side of the equation. Effective national development projects in 60 developing countries, change management, team owner- Check out the May episode working with donors such as USAID, the U.S. Department ship and executive commitment are of ASQ TV, which focused on of Defense and the U.K.’s Department for International equally—if not more—important. supply chain management and Development. In the company’s $550 million USAID-fund- Ensuring the effort was grassroots the Kenya Pharma project. Visit ed Kenya Pharma project, Chemonics and its client, US- and designed by and for the project http://videos.asq.org/home AID, saw a unique opportunity. staff was at the forefront of the ap- to see the proach. entire May Using ISO 9001 logically built on the approach for imple- episode or menting the Kenya Pharma supply chain, which was based On my first trip to Nairobi in early on employing private-sector, internationally recognized 2011, the Kenya Pharma team spent the individual management practices that support supply chain opera- some time discussing what having segment on tions to fight the spread of the HIV/AIDS epidemic in Kenya. a management system would mean the project. June 2013 • QP 21 JOSEPH ACHIENG, who lives in Kibera, a large low-income informal settlement in Nairobi, walks through a maze of footpaths and trenches to get to the Tabitha Medical Center to pick up his medication. for the project. The quality policy the team developed— For example, a notable innovation of the Kenya with its emphasis on the customer-focused result of an Pharma project from its onset in 2009 is its field service optimized, continually improving supply chain—says it representative (FSR) model. FSRs operate throughout all and carried the team through the entire process: the country, serving a similar function to that of pharmaceutical sales reps: acting as customer service liaisons, Kenya Pharma is committed to using its ISO 9001 but without a sales function. Their role is to regularly quality management system (QMS) to optimize a sus- engage with each of the 169 ordering sites that the proj- tainable supply chain that provides commodities for ect serves to monitor the ordering and receipt process, prevention, care and treatment of persons with HIV/ receive feedback on expired or nonconforming product AIDS in Kenya. The goal of the Kenya Pharma QMS is and provide advice on appropriate storage. to continually improve our ability to provide effective The percentage of orders received on time has in- service and high-quality supplies to persons in need. creased from 61 to 78% due in part to the FSRs’ work. We will do our job ethically and follow all statutory In this way, the project improved on its already-effective and regulatory requirements seeking to exceed our services by using ISO 9001—paired with a strengths- stakeholders’ expectations. We shall always keep the based approach—to have the greatest impact for ben- welfare of our patients in mind, striving to add life to eficiaries. Although the project was already strong, ISO their days and days to their life. 9001 brought the following advantages: • The record control guidelines pushed the staff to In particular, three themes played a key role in the imple- reach the next level to institute records naming con- mentation process, making the project’s ISO 9001-certified ventions and centrally store its records. This saved QMS a true operational foundation for the supply chain. staff time in the long run by making these documents 1. Play to your strengths. The Kenya Pharma project was operating with zero stock-outs—a rare feat in easier to find and making the staff’s internal operations more efficient. the developing world for a pharmaceutical supply chain. • It brought a level of formality to process management The project sought to take that quality and reliability of and improvement with a set vocabulary and rubric to delivery, accuracy and cost-effectiveness to the next lev- put structure around what was already working well. el. With clearly documented practices that staff adhered • It allowed staff members working at different points in to repeatedly and were held accountable for, the project the supply chain to understand each other’s roles better. was able to get more from its margins. 2. Make it collaborative. The process of document- 22 QP • www.qualityprogress.com case study ics, including our Washington, D.C.-based quality man- Pharmaceutical shipment: time 32% to clear customs (in days) / figure 2 agement unit (QMU) and Kenya Pharma project manage- 25 ing supply chain practices was intensely collaborative. It was a cohesive effort among multiple teams at Chemon- ment unit (PMU), the Nairobi-based Kenya Pharma field office staff and two QMS specialists from our Washing- 15 with project staff in Nairobi. 10 Pharma field office staff to lead the process, supported by the QMU and QMS specialists with oversight and co- 21.8 20 ton, D.C., office, who spent months at a time working A smaller team was formed from within the Kenya 47% After QMS 68% Before QMS 5 5.7 After QMS 2.5 QMS 0 = quality management system Sea shipments ordination from the PMU. All worked together to develop the project’s quality policy, and supply chain process Before QMS 53% 1.8 Air shipments QMS = quality management system leads were assigned from among the project staff to take charge of designing their process documentation using tools, templates and trackers adapted from Chemonics’ home office ISO 9001 implementation process. Results of health facility ­supply chain data quality ­audits / figure 3 Many interactive process mapping and process review sessions were held involving anyone who was part of a particular process—no matter how small his or her 32% 47% role might be. This provided some key benefits: 68% 53% • Diverse perspectives were included, enhancing the accuracy and utility of the final product. • Players from different parts of the supply chain—from Before QMS quality assurance to quantification to procurement— were able to understand parts of the supply chain they had not necessarily been previously exposed to, fostering a deeper appreciation for each other’s work Pass After QMS Fail QMS = quality management system along with empathy for the types of issues co-workers experienced. collection, processing and reporting, ensuring consistent • This collective learning opened the door for innova- monitoring and evaluation over time, and also for facili- tive suggestions for how to streamline supply chain tating continual improvement of supply chain manage- operations. ment and performance. Indicator data include definitions, Through this process, all were learning together. For units of measure, disaggregation types, management util- example, those who knew ISO 9001 well were learning ity, data collection methods, data sources, cost to collect supply chain operations and vice versa. Everybody had data, data acquisition schedules, limitations and formats a role to play. for data presentation. 3. Results, results, results. As project staff docu- The project also adheres to World Health Organiza- mented processes, it was critical to keep the end in sight: tion (WHO) treatment guidelines and a series of indus- the client, USAID, and the project’s beneficiaries: the pa- trywide technical standards in procurement, distribution tients. The project’s performance management plan in- and warehousing. These also were purposefully woven cludes 43 detailed output and outcome indicators across into the documented processes to ensure the supply the supply chain, including customer satisfaction, qual- chain and all of its industry and technical requirements ity assurance and transportation. These were purposely were captured and reinforced by the QMS. woven into each process and documented at the appro- Using ISO 9001 requirements as an operational founda- priate steps so staff has results and targets in mind as it tion for Kenya Pharma’s supply chain was a strategic invest- executes processes. ment to holistically capture all aspects of the supply chain Each indicator includes detailed instructions for data and serve as a playbook for staff and blueprint for success- June 2013 • QP 23 How to use our QMS / figure 4 Four steps to success ful supply chain operations in the developing world. This method, in turn, can be duplicated worldwide and potentially scaled up across diversified health commodities and pharmaceuticals that address issues such as malaria, tuberculosis, reproductive health, family planning and emerg- 1 Follow processes Make sure you are familiar with our process maps so we’re all on the same page and clear about roles and responsibilities. Process maps embody best practices and proven methods for providing maximum value to our clients: USAID, SDPs, and ultimately, the Kenyan population touched by the Kenya Pharma supply chain. ing pandemic threats, as well as equipment and medical supplies supporting lab services. Reaping the benefits In July 2012, DNV Business Assurance conducted the Kenya Pharma project certification audit, identifying no nonconformities. Richard Dreiman, Chemonics’ CEO at the time, addressed staff, friends and partners at the certification celebration, emphasizing the real value of the process. “The project was already operating with high standards 2 Keep records Each process map describes what records should be developed and kept. Records prove that our quality processes have been followed and provide historical documentation of our work on Kenya Pharma. of quality management even before seeking certification ... now we have an official, internationally recognized stamp of approval,” Dreiman said. He also noted the importance of Kenya Pharma as “one of the first USAID projects to become ISO 9001:2008 certified.” The Kenya Pharma ISO 9001-certified QMS has created a robust foundation for supply chain operations—a home for the 3 Participate in audits There are two types of ISO 9001 audits: internal and external. Internal audits are conducted by a trained group of Kenya Pharma staff and are designed to identify process improvements and discover areas where we might not be following our processes or keeping records as our QMS specifies. External audits are conducted by our registrar, DNV. Our first audit to obtain our initial ISO 9001 certification will be this summer. The project will have periodic audits in the summer of 2013 and 2014. During both types of audits, you may be asked about the work that you do and your role on the project. project’s top-notch supply chain process documentation— which can be improved continually with the plan-do-checkact framework as part of the ISO 9001 approach. The focus of ISO 9001 is on the customer—reinforcing and systematizing the Kenya Pharma quality policy—“…adding life to their [patients’] days and days to their life.” On a daily basis, ISO 9001 helps ensure problems that do arise in the supply chain are systematically documented and addressed not only to correct an immediate issue, but also to gather data on larger trends impacting the project’s ability to serve its customers through the corrective and preventive action process. 4 Suggest improvements With our QMS, we now have a built-in mechanism to capture, assess and implement suggestions. Corrective actions are submitted when you identify a problem with a business process or when you discover a trend that could be tied to one or more business processes. Preventive actions are submitted when a business process is already functioning well, but you identify ways that Kenya Pharma can improve it or prevent potential problems. More specifically, integrating an ISO 9001 approach has delivered several key benefits for Kenya Pharma that will help lead to long-term successes: Flexibility. Having used ISO 9001, it is possible to change supply chain operations more smoothly and quickly while maintaining consistency across the board. With ISO 9001 document control, and employee awareness and training requirements, the Kenya Pharma QMS provides a central system in which all work processes and change notifications are effectively captured. Process improvement becomes seamless, flowing through all aspects of the supply chain without the QMS = quality management system SDP = service delivery point USAID = U.S. Agency for International Development risk of being siloed. This has been especially significant to Kenya Pharma because the project must be sensitive to policy shifts and regula- 24 QP • www.qualityprogress.com case study tory requirements from external sources: WHO, USAID, handout used to educate team members about the QMS. the U.S. government and the Kenyan government. It can This effective teamwork could be seen from the begin- now push out information and updates faster to remote ning and in the launch week for the QMS. Launch week FSRs who work around the country and, in some cases, was constructed with each day having a theme—docu- in very remote locations. Additionally, the Kenya Pharma ment control, record control, internal and certification au- supply chain has improved measurably in a number of oth- dits and continual improvement. Every session was taught er areas since the QMS implementation: The average time by a team of two people—one from our Washington, D.C. for sea and air shipments to clear customs decreased from office and one from our Nairobi-based operations. 21.8 days to 5.7 days, and 2.5 days to 1.8 days respectively By the end of the week, it was important that the in- (see Figure 2, p. 23). The percentage of health facilities country staff members were the go-to people and viewed passing supply chain data quality audits increased from 53 that way by their counterparts. Every session opened to 68% (see Figure 3, p. 23), and significant improvement and closed with Q&A contest on material previously cov- was seen in the project’s performance in forecasting fu- ered, with prizes for the winners. ture costs to manage and operate the supply chain. And—most importantly—everyone from the recep- Sustainability. “We are focusing on ensuring that tionist to the project drivers to the project director this operation is sustainable,” explained Kenya Pharma participated. This strengthened the project’s culture of Chief of Party Steve Hawkins. “Chemonics’ model is inclusiveness—the softer side of business operations, ‘build, operate and transfer.’ ISO certification helps us which goes a long way in the end to achieving results. achieve that model.” Kenya Pharma’s management representative, Jackson The resulting benefits from this process were not only Kariithi, said, “The process of rolling out and putting the to strengthen the supply chain against staff turnover and QMS in place has created significant cohesion within the be malleable in the face of client shifts or changes at the entire team. We now put quality at the forefront of our global health policy level, but also to transfer supply every activity.” chain operations to local Kenyan counterparts at project For supply chains around the world, there’s no ques- end in 2014. In effect, Chemonics was hired to work itself tion ISO 9001 will provide a solid backbone for capturing out of a job—a different context from what most busi- operational nuances. Operational efficiency is a chal- nesses face, but one that we, as an international develop- lenge for any supply chain to effectively serve its custom- ment company, encounter on every project. ers and meet financial expectations. For Kenya Pharma, This challenge has valuable lessons for any context. its quality anthem developed by project staff provides a Design your system with these questions in mind: “Could testament to how interlinked quality is with the project’s I turn our supply chain over to someone else to run? Can mission of delivering HIV/AIDS commodities to Kenyans: I teach them how to do what I do?” If the answers are yes, you can be confident in the Well, well, well, we’re Kenya Pharma sophistication of your QMS and knowledge that you have Well, well, well, we’re here to serve reaped the benefits of an ISO 9001 approach. Asking Well, well, well, we’re Kenya Pharma these questions and examining supply chain operations Quality is our job. QP from this angle puts a fresh perspective on your business operations and ensures you develop a robust system that is attuned to the micro and macro features of your supply chain upstream and downstream. REFERENCE AND NOTE 1.Tiffany Darabi, “Getting a Boost From Quality Tools,” case study, ASQ, January 2012, http://rube.asq.org/2012/01/social-responsibility/quality-tools.pdf. 2.For a step-by-step guide to implementing ISO 9001, see John Orthaber’s “Get Your Ducks in a Row,” Quality Progress, October 2010, pp. 40-46. Project culture. An anecdote shared by project staff in Nairobi—a team of about 40 people, all but two of whom are Kenyans—is that this process brought staff members together as a team and made them more invested in their work on Kenya Pharma. The continual improvement process created a culture in which everyone had a voice, was learning and had a role in innovating and improving. See Figure 4 for a condensed version of a TIFFANY DARABI is the director of the quality management unit at Chemonics International Inc. She has worked for Chemonics since the fall of 2003, spearheaded the development of Chemonics’ ISO 9001-certified quality management system, and oversaw the Kenya Pharma ISO 9001 certification process. An ASQ member, Darabi holds a bachelor’s degree in international relations from John Hopkins University in Baltimore. June 2013 • QP 25 The Power of Prediction by Gregory A. Kruger In 50 Words Or Less • Inconsistent supplier delivery further complicates already difficult-to-predict service parts supply needs and can negatively impact an organization’s ability to serve its customers. • Organizations can use quality and statistical methods—each with benefits and drawbacks— to predict service part demand and minimize the impact of supplier delivery inconsistency. SUPPLY CHAIN Minimize the impact of inconsistent supplier delivery by using statistical quality models service parts inventory management can be daunting. You can all relate to needing your car repaired. Say you need a new water pump. You really don’t care that demand for your particular model of water pump is low volume, intermittent and difficult to forecast. You also don’t care that the supplier of this water pump does not necessarily deliver on time to the regional service parts warehouse. You just want the car fixed—and fast. Quality in service parts management means availability without excess inventory. In the service parts business, low-volume demand can dry up, leaving you holding the bag with excess and obsolete inventory write-offs. While my service parts inventory management experience is in the electronic test business, this situation is directly analogous to the all-too-familiar need for a new water pump in a car. June 2013 • QP 27 Applying statistical models can derive a statistical model to determine safety stock Considerable work has been done in applying statistical given both demand and supply uncertainty. inventory stocking models to the service parts situation.1 While the original safety stock model that looked only For example, the Poisson probability model frequently at demand uncertainty may have recommended a safety works well in modeling the variation seen in low-demand stock of X for a given supply lead time and demand situa- service parts. An alternative to the Poisson is a hybrid tion, the new model will recommend a safety stock of X + model that handles the more intermittent demand situa- Y where the incremental quantity Y covers the uncertainty tion by separating the probability of no demand and the of when any given supply order will actually arrive. distribution when there is demand. On the surface, it would appear the problem is Such statistical models allow for the determination of solved—but not for the first time—because there are safety stock to load into the material requirements plan- also approaches using normal distribution theory.2 Then ning (MRP) system to achieve a targeted service level. again, perhaps we have a technically elegant solution no Experience has shown, however, that the actual service one wants. level achieved consistently runs a few percentage points lower than that advertised by statistical theory. Transient supplier delivery issues Digging into that situation, the process improvement It is frequently the case that supplier delivery problems team at my organization quickly identified the top factor come and go due not only to issues unique to the sup- behind actual parts availability consistently falling below plier, but also issues across the industry. For example, projections—imperfect supplier delivery performance. an industrywide slowdown can prompt suppliers to trim The statistical models presumed a supply order placed at capacity as a means of weathering the downturn. When the suppler lead time would always arrive on time. This the market returns, there is the problem of ramping back is simply not reality. up. During that period, suppliers across the industry are Perhaps particularly true in the service parts business, likely to be constrained for a period of time. with its low volume and intermittent demand, suppliers Suppose a statistical model incorporating past supplier do not always deliver needed supply orders precisely at delivery performance into the safety stock recommenda- their stated lead times. For example, a little data reveals tion is implemented. Assume the implementation is look- supplier delivery on a service part with a stated lead time ing back over a rolling period of several months of past of 10 weeks may look something like Figure 1. supply deliveries to draw its conclusions about what can This may seem outrageous in a supply chain in which be expected with future supply orders (see Figure 2). the purchaser commands considerable power, but it can Thinking this through, we see the problem. The pe- be common when the purchasing company does not riod of supply issues begins, but it takes some time be- have the volume necessary to exert much influence on fore enough delivery history is collected to trigger the the supplier. Remember: These are supply orders for statistical safety stock model’s recommendation to cover service parts having intermittent, low-volume demand, the deteriorated supply delivery performance. After the so the supply orders themselves are of very low volume. new, larger safety stocks are loaded into the MRP sys- Furthermore, if you are in the business of supporting tem, the new supply orders are now larger, but we are an installed base of aging technology, the industry has already well into the period of supply problems—mean- largely moved on so it is difficult to garner a great deal of ing it’s too late. attention from your supply base. When the supply issues improve, it would once again take time for the system to purge the period of poor de- What should you do? livery performance from history and reduce safety stock As it turns out, the statistical models for the determination recommendations back to values appropriate for the of safety stock can be expanded to incorporate supplier new delivery reality. Finally, and worst of all, the period delivery distributions, such as the one depicted in Figure in which actual inventory on hand reflects the additional 1. By interpreting the chart to say that with the placement safety stock target values likely begins and ends far too of the next supply order, there is a 75% chance the order late. In fact, it is even possible the period of holding the will arrive at lead time, a 15% chance it will arrive one incremental safety stock misses the duration of delivery week late and a 6% chance it will arrive two weeks late, we problems entirely. 28 QP • www.qualityprogress.com SUPPLY CHAIN So, we would have to anticipate the period of supplier delivery problems as well as the shape of the new delivery perfor- Supplier delivery performance given 10-week stated lead time / Figure 1 to load the higher safety stock targets into the MRP system in time to receive supply timed with the period of delivery problems. All of this looks highly impractical. Philosophical objections Suppose we limit the use of a statistical safety stock model incorporating supplier delivery performance to only situations in Percentage of deliveries mance histogram (as depicted in Figure 1) 100% 80% 60% 40% 20% 0% 10 which there is sustained expectation of ongoing supply lead time inconsistency. As it turns out, this can elicit an objection based on not wanting to reward a supplier for bad delivery performance. The thinking goes as follows: Poor supply lead time consistency leads to increased safety stock targets, which 11 12 13 Delivery performance in weekly increments 14 Incorporating past supplier delivery performance / Figure 2 Period of delivery issues means incrementally larger orders to the Period delivery issues impact SS calculation supplier, which means more business for the supplier and, therefore, a reward for poor delivery. The counterargument is Period actual SS on hand is impacted that it is the company’s job to protect the customer from the reality of underlying issues with supply. This second line of thinking is similar to Time SS = safety stock the well-understood practice of mechanical tolerance stackups. In mechanical tolerance analysis, the engineer accounts for the variation in buyer may place the order seven or eight weeks in ad- the manufacturing processes for each of the components vance of the need date. While understandable, this prac- going into the assembly. Because component manufactur- tice raises issues around manual intervention by the buy- ing processes are not perfect, the variation in output is er over MRP system ordering recommendations unless mathematically combined to create the expected varia- the MRP system itself allows the buyer to enter an early tion in the final assembly. Similarly in inventory manage- order time buffer. ment, we have a tolerance on expected demand and a tolerance on exactly when we will get supply. Genichi Taguchi raised a similar issue in the parameter design phase of new product development, arguing These differing perspectives—do not hold incremental that parameters should be selected to account for ex- safety stock to buffer against supplier delivery inconsis- pected variability.3 In this case, our parameter is the lead tency versus, yes, do so to protect the customer—create time at which supply orders are placed, and we are se- tension within the supply chain organization. lecting a nominal value that is robust enough for delivery performance variation. Practical reality The interesting thing is that such early order practices In practice, it is common for buyers who are aware of mean essentially holding safety stock to cover supplier past lead time inconsistencies to place the supply order delivery inconsistency while never calling it that. To be early. If the supplier’s stated lead time is six weeks, a sure there is no visible increase in the official safety June 2013 • QP 29 stock targets, but equally as certain, there is an increase In the real world, supplier delivery is likely to ebb in average inventory held on hand. Also, the practice and flow due to dynamics in the industry. The organiza- suffers from similar problems the safety stock model tion is likely to push back at the idea of visibly hold- had with transient supply issues, such as the timing of ing incremental safety stock inventory to buffer against when to add such an early order buffer and when to supply uncertainty. Lean principles suggest to resolve the delivery is- take it off. If the organization is not using a formal statistical sues. Perhaps a financial analysis would enable a cre- model, another option is for buyers to rely solely upon ative supply contract containing incentives for on-time their judgment in factoring supply and demand varia- delivery even during periods of industrywide supply tion into the setting of safety stock targets. constraints. This approach leads into the topic of struc- This makes it less visible that supplier delivery per- tured contracts.4 formance played any role in sizing the safety stock buf- Certainly, we want a win-win relationship with our fer. So the practical reality is that the organization subtly suppliers, and integral to that objective is a foundation goes about buffering delivery problems with incremental of social capital between customer and supplier in the inventory. While bordering on heretical, is the subtle ap- supply chain.5 It just may be that the supplier is already proach more acceptable to the organization? making a substantial commitment by providing the delivery performance depicted in Figure 1 (p. 29), consider- An unwanted solution ing the end game is servicing customers using ancient What is the appropriate strategy for achieving assurance technology. of supply in the face of inconsistent supplier delivery Left alone, the organization will do what it has performance? A computer Monte Carlo simulation of learned—ordering early or engaging in other practices the MRP system demonstrates that our statistical safety that handle the reality of imperfect delivery in less than stock model—enhanced to cover supplier delivery varia- a completely visible fashion. By visible, I mean delineat- tion—performs as the theory advertises. Yet, there is lit- ing up front the incremental inventory investment re- tle motivation to implement it. Such a computer simula- quired to handle imperfect supplier delivery. tion presumes supplier delivery will follow a prescribed It may sound absurd to you if your frame of reference is a supply chain in which the purchasing company’s probability distribution. volumes provide clout, but I still vividly Impact of increasing minimum order quantity / Figure 3 35 that part of his job was to grovel at the feet of the supplier, begging for parts. After working through the Six Sigma define, measure, analyze, improve and control process, is the solution to simply main- 30 Inventory on hand recall a buyer telling me many years ago tain those excellent groveling skills? 25 In the face of lean 20 The process improvement team pulled together and brainstormed the ques- 15 tion: How do you reduce the impact of imperfect supplier delivery? Answer: 10 Reduce the frequency at which you need 5 0 a delivery by increasing the order quantity. Great, but this comes at a cost—in- 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 Time period MOQ = minimum order quantity 30 QP • www.qualityprogress.com MOQ = 10 MOQ = 30 creasing the average inventory held and the risk of eventual excess inventory write-off. A balance must be struck, raising minimum order quantity (MOQ) but SUPPLY CHAIN doing so only to a level supported by demand forecasts the supplier of the need for a last-time buy. As the indus- while confident of eventually consuming the purchase try moves on to new technology, suppliers are frequently quantity. unwilling to maintain production capability for the ag- This simple concept is depicted in Figure 3. Note that ing segments of their portfolios. As a result, a supplier given expected demand of one per period when ordering will offer one final build. Now we must determine how in quantities of 10, we are subject to the supplier’s de- many parts to buy to balance the following two compet- livery performance every 10 periods. By increasing our ing business objectives:6 order quantity to 30, we are now subject to the supplier’s 1. High probability of not stocking out before the end delivery performance only every 30 periods. The net re- of the advertised service repair life of the product. sult will be improved on-time service parts delivery to 2. Low risk of being left with excess and obsolete the end customer. inventory write-off at the end of service life. So, we replaced a potentially complex solution in the Label the number of parts required to meet customer form of a statistical safety stock model with a simple solution. To be fair, however, we would need to compare the average on-hand inventory expected be- Three-year demand forecast / Figure 4 4 tween the statistical safety stock methapproach raised our average on-hand inventory substantially higher than the statistical alternative, while elegant in its simplicity, the MOQ solution has a clear drawback. Speaking of drawbacks, have we for- Forecast demand od and the MOQ solution. If the MOQ 3 2 1 gotten the seven forms of waste from 0 lean training? 1.Transportation. 1 3 5 7 9 2. Inventory. 11 13 15 17 19 21 23 25 27 29 31 33 35 Month 3.Motion. 4.Waiting. 5.Overproduction. 6.Overprocessing. 7.Defects. Here, we are consciously increas- Poisson distribution for ultimate usage during 36 months to end of support life / Figure 5 ing inventory and seemingly ignoring One of the realities of service parts management is the dreaded notification from 130 126 122 118 114 110 106 102 98 94 90 86 82 The last-time buy 78 make other decisions. 74 part has become more tenuous, we 70 when the simple act of purchasing the 66 cess vision. At the end of the life cycle, UU quantity at 95% assurance of supply = 105 62 erate more true to the classic lean pro- Most likely UU quantity = 89 58 when volumes were high, we could op- 0.045 0.04 0.035 0.03 0.025 0.02 0.015 0.01 0.005 0 50 trade-offs. Early in the part’s life cycle, Probability density management. Those are the real-life 54 the principles of just-in-time inventory Quantity UU = ultimate usage June 2013 • QP 31 SUPPLY CHAIN As the industry moves on to new technology, suppliers are frequently unwilling to maintain production capability for the aging segments of their portfolios. service repair demand until the end of service life In the scenario that service repair demand consumes the ultimate usage (UU) quantity. Suppose the part all of the last-time buy purchase, we could cannibalize for which we must make a last-time buy is currently those instruments for the service parts, getting our cus- running an average demand of three per month, and tomers back up and running despite it being years since we see from history that the Poisson distribution does the product was in production. a reasonable job of modeling the variation in actual monthly demand. By considering the expected distribution of ultimate use along with the potential for obtaining supply via Suppose we expect a slow decline in the demand unconventional means, such as cannibalization, we can for this part as our installed base shrinks due to cus- make a decision balancing the goals of availability and tomers moving to our newer product generations. As- write-off risk. suming we are three years away from the eventual end of support life, the forecast would look something like Lesson learned Figure 4 (p. 31). Like most of us, I remember humorous expressions Taking advantage of the knowledge that the sum of my parents used. I recall my dad saying something was Poisson distributions is itself Poisson, we can model “easy as falling off a log.” As a Black Belt, I periodically the uncertainty in the ultimate use of this part over the teach lean Six Sigma Green Belt classes. I now make three years to end of support life with the distribution sure to point out to my students that just because the depicted in Figure 5 (p. 31). Note that as the average of root cause has been identified, arriving at the solution a Poisson distribution gets large, the distribution be- is not necessarily as easy as falling off a log. QP comes more bell-shaped (normally distributed). In this example, the mean or expected value of the distribution is 89, which is the single best forecast for the UU quantity during the next three years. If we would like some assurance that our last-time buy will cover all demand, we might decide to purchase at the 95th percentile, leaving a 5% chance of stocking out and disappointing a future service repair customer. Unfortunately, that would increase our risk of ending up with excess and obsolete inventory. Placing a lasttime buy now for a quantity of 105 would leave us with References 1.Robert Goodall Brown, Advanced Service Parts Inventory Control, Materials Management Systems Inc., 1982. 2.Donald J. Bowersox, David J. Closs and Omar K. Helferich, Logistical Management—A Systems Integration of Physical Distribution, Manufacturing Support and Material Procurement, third edition, Macmillan Publishing Company Co., 1986. 3.Raghu N. Kackar, “Off-Line Quality Control, Parameter Design and the Taguchi Method,” Journal of Quality Technology, October 1985, pp. 176-188. 4.Bill McBeath, “2012—the Year of Uncertainty: Part Two,” ChainLink Research, January 2012. 5.Roger McGrath Jr. and William L. Sparks, “The Importance of Building Social Capital,” Quality Progress, February 2005, pp. 45-49. 6.Gregory A. Kruger, “Using Six Sigma to Reduce Excess Service Parts Inventory While Maintaining Service Levels,” presentation at the ASQ Lean Six Sigma Conference, 2010. an expected future write-off of 105 − 89 = 16 parts. Are there other alternatives? Finding balance Once again, our process improvement team came through with an answer. The company owns complete products that may have been part of the sales demo fleet or used by other divisions of the corporation. 32 QP • www.qualityprogress.com Gregory A. Kruger is a statistician and supply chain inventory analyst for Agilent Technologies in Colorado Springs, CO. He has a master’s degree in statistics from Iowa State University in Ames. Kruger is an ASQ member. OFFICE USE ONLY Membership Application MBKDF32 PRIORITY CODE _______________________ Order Number ________________________ Member Number ______________________ Industry: ❑ Healthcare ❑ Service ❑ Government ❑ Education ❑ Male ❑ Female ❑ Manufacturing Preferred Mailing Address: ❑ Home ❑ Mr. ❑ Ms. ❑ Mrs. ❑ Dr. ❑ Business / / Date of Birth _________________________________ M D 1 Member Type: ❑ Full $143 2 The one geographic Section included with Full membership will be determined by your primary address.* Y ____________________________________________________________________ First Name Middle Initial Last Name/Family Name ____________________________________________________________________ City, State/Province Zip+4/Postal Code Country ____________________________________________________________________ Home Address Apt./Ste. ____________________________________________________________________ City, State/Province Zip+4/Postal Code Country ____________________________________________________________________ Area Code/Business Telephone Area Code/Home Telephone ____________________________________________________________________ Preferred Email Address Fax Which one of the following best describes your title? ❑ Accountant ❑ Administrator ❑ Advisor ❑ Analyst ❑ Associate ❑ Auditor ❑ CEO ❑ Chemist ❑ Clinician ❑ Consultant ❑ Contractor ❑ Controller/ Comptroller ❑ Coordinator ❑ Director ❑ Engineer ❑ Facilitator ❑ Foreman ❑ General Manager ❑ Inspector ❑ Instructor ❑ Machinist ❑ Manager ❑ Mechanic ❑ Nurse ❑ Owner ❑ Physician ❑ President ❑ Principal ❑ Professor ❑ Programmer ❑ Retired ❑ Scientist ❑ Six Sigma Black Belt ❑ Six Sigma Green Belt ❑ Specialist ❑ Statistician ❑ Student ❑ Superintendent ❑ Teacher ❑ Technician ❑ Unemployed ❑ Other cccc cccc cccc Contact ASQ to change your assigned Section. 3 Forum or Division Selection As part of your Full membership you receive participation in one topic- or industry-specific Forum or Division. Use the list below to indicate the Forum or Division number and name. included _____ ______________________ $______________ (#) Name Additional Forums and Divisions may be added to all levels of membership. Please indicate in the list below the additional Forums or Divisions you would like and total the number you have selected. ❑ Audit (19) ❑ Automotive (3) ❑ Aviation, Space and Defense (2) ❑ Biomedical (10) ❑ Chemical and Process Industries (4) ❑ Customer-Supplier (15) ❑ Design and Construction (20) ❑ Education (21) ❑ Electronics and Communications (5) ❑ Energy and Environmental (11) ❑ Food, Drug, and Cosmetic (7) ❑ Government (22) ❑ Healthcare (18) ❑ Human Development and Leadership (13) ❑ Inspection (9) ❑ Lean Enterprise (23) ❑ Measurement Quality (17) ❑ Product Safety and Liability Prevention (25) ❑ Quality Management (1) ❑ Reliability (8) ❑ Service Quality (16) ❑ Six Sigma (26) ❑ Software (14) ❑ Statistics (12) ❑ Team & Workplace Excellence (27) Additional Forum and Division selections: Full or Associate member ________ x $10 = $________________________ total Total of all items (1-3): Mailing Lists ❑ Occasionally ASQ shares its mailing list with carefully selected quality-related organizations to provide you with information on products and services. Please check this circle if you do not wish to receive these mailings. Payment Information ❑ Check or money order (U.S. dollars drawn on a U.S. bank) Make check payable to ASQ. ❑ MasterCard Member Referred By: _______________________ Member Number ❑ Visa ❑ American Express (Check one) ____________________________________________________________________ Cardholder’s Name (please print) ______________________________________________ Card Number WHY DID YOU JOIN? To help us understand what’s important to you, please tell us the top three reasons why you became an ASQ member. ❑ ❑ ❑ ❑ ❑ ❑ ❑ ❑ ❑ ❑ Career Development Certification Pricing In-person Networking Involvement in ASQ’s Cause Involvement in SRO Knowledge/Information Leadership Opportunities O nline Networking/Communities Product Discounts Training $ _________ Please submit your application with remittance to: ASQ does not sell email addresses to third parties. _____________________________________ Member Name $___________________ Sections (geographic) may be added to any member type for an additional $20.00 each. Visit www.asq.org for a listing of available Sections. , , _ $_______ Sections ____________________________________________________________________ Company Name Job Title ____________________________________________________________________ Business Address Ste. ❑ Associate $85 ___________________ Exp. Date ______________________________________________ Cardholder’s Signature ____________________________________________________________________ Cardholder’s Address ASQ P.O. Box 3066 Milwaukee, WI 53201-3066 USA or fax to 414-272-1734. You may also join online at www.asq.org or by calling ASQ Customer Care at USA and Canada: 800-248-1946 Mexico: 001-800-514-1564 All other locations: +1-414-272-8575 New memberships are effective upon receipt of payment. New members receive one year of membership from the date they join. Members are billed prior to the anniversary date of their membership for next year’s dues. Memberships, even those paid by employers, are nontransferable. All prices are subject to change. In becoming an ASQ member, you have the duty to follow the ASQ Code of Ethics and Society governing documents. Parts of the Process Evaluating the impact of the number of parts on a measurement process In 50 Words Or Less • Repeatability and reproducibility (R&R) studies can be used to pinpoint the parts of a measurement system that might need improvement. • By using simulation, there is practically no difference in using five or 10 parts in many R&R studies. REPEATABILITY AND REPRODUCIBILITY (R&R) studies are based on the concept of variance components estimation. This is a remarkable achievement of statistical science because people usually think in terms of means and trends, not variation around a mean. These types of studies provide useful information about which human or gage component of the measurement process should be improved. These studies are a crucial step before starting a statistical process control procedure. They are conducted so observations being charted will provide meaningful information and not just measurement error. statistics by Víctor M. Aguirre-Torres and Maria Teresa López-Alvarez R&R studies require assembling a protocol that includes performing blind measurements on N parts by M operators and repeating the whole setup n times. If all the operators measure the same set of parts, the study is said to be crossed. Otherwise, it is nested. This article addresses crossed studies, but the methods described can be easily extended to nested studies. Usually, the recommendation for R&R studies is to have 10 parts, two or three operators, and two or three replicates.1-3 This recommendation comes from the automotive industry, a sector in which most measurements are relatively inexpensive and easy to gather. June 2013 • QP 35 An in-depth study by Minitab4 shows simulationbased results that relate the number of parts with the 10 samples are used, this means two full working days for three people. relative error in the estimation of the part-to-part varia- 2. Cost. Most chemical determinations require one tion. For 10 parts, the study mentions that the 90% con- or more reactants, which are expensive because they fidence interval has a relative error of 35 to 40% and should be of high quality. Also, most analytical tests that you would need to use 135 parts to get a relative are destructive and the testing material can also be error of about 10%. expensive. The focus of this article is not oriented toward vari- 3. Delicateness of samples. In analytical mea- ance component estimation. Although R&R studies surement processes, testing is often destructive, and make use of variance components, the end result is samples suffer deterioration over time. Consider not the separate estimation of the components but an measuring humidity. If the process has 10 operators estimation of a function of them. in the three shifts, and 10 parts are repeatedly used, Variance component estimation is an extensively this implies 200 samples. Even if the evaluation of developed science in and of itself. The estimation of humidity doesn’t take very long, you may not obtain variance components is important in animal breeding 200 determinations quickly enough to guarantee all or sampling from finite populations with sample sizes the samples of the characteristics will be preserved of hundreds or thousands. For R&R studies, however, during the entire measurement process. If testing we look for evidence that the relative sizes of variance time is too long, the samples can be expected to components are acceptable. change, and this adversely affects the results of the Usually, the end result of an R&R study is the percent R&R study. R&R (%R&R), which could be the ratio of the standard 4. Certification process. When certifying a mea- deviation of the measurement error relative to the total surement process during a lean Sigma or Six Sigma standard deviation, which is called %R&R1, or the ratio project, the number of R&R studies that must be per- of the 99% interval due to measurement error divided by formed could be very large. In practice, we have found the span of the tolerance, which is called %R&R2. a production line requires a minimum of 10 to 15 gage There are guidelines to evaluate the measurement R&R studies. If this is multiplied by the total number system based on the %R&R. In the automotive indus- of lines and processes in a plant, this could result in an try, a measurement system with the %R&R less than enormous amount of work devoted just to certify the 10% is usually considered acceptable.5 For example, measurement systems. this would be restrictive in the food industry. In our This article intends to provide evidence that the end experience, the 30% limit is taken as a minimum re- result of an R&R study is basically the same, whether quirement at Kraft Foods. Here, let’s consider that if it’s for five or 10 parts. To do this, we focus on the prob- this percentage is less than 30%, the measurement ability of acceptance of the measurement system— system is said to be acceptable. Otherwise it is con- that is, the probability that the %R&R is less than 30%. sidered marginal or unacceptable—depending on how far it is from 30%. We refer to the probability of acceptance because when we run an R&R study, we observe only a sam- If we focus on another industry besides automo- ple from the process and, hence, we do not observe tive, such as the food industry, it is important to define the real %R&R. In fact, the estimate of %R&R is a an adequate size for the R&R study that considers the random quantity that is subject to random variation, following crucial aspects: which is characterized by means of a frequency dis- 1. Time required to complete the R&R study. tribution. When the measurement requires analytical determi- We will show that not only the probability of ac- nations, the time required to make the measurement ceptance is close for the two numbers of parts, but is usually long because it involves several steps. Con- also that the profiles of the frequency distributions are sider the determination of protein. It is first a diges- similar. To do so, we then describe the model and the tion, then a titration and finally a distillation. Using five estimation procedure to be considered. Later, we will samples (parts), three operators and one replication show the simulation results of %R&R for different sce- require a minimum of eight hours for three people. If narios and the two numbers of parts. 36 QP • www.qualityprogress.com statistics The model The model that we consider for the measurements is a random effects model without interaction given in equa- Frequency distribution of % R̂&R1 for system one / Figure 1 tion one: yijk = µ + τi + βj + εijk Equation 1 i = 1,…, N; j = 1,…, M; k = 1,…, n. 10 els the part-to-part variation, βi represents the operator σ2β +and σ2ε ε models the variation variation or reproducibility, ijk 100%. %R&R1 = 2 due to the gage or repeatability. there is no inσ2τ + σ2β + σAssume ε Density The parameter µ represents the overall mean, τi mod- teraction between the operator and parts, which will be yijk = µ + τi2+ βj + εijk Equation Equation 1 Equation 1 yijkN;= jµ=+1,…, τi + β + εijk 1,…, n. i = 1,…, M; j k = i = 1,…, N; j = 1,…, M; k = 1,…, n. 5 explained later.6 i = 1,…, yN; j==µ 1,…, 1,…, n. σβ2εjk+=εijk +_σ2τβ i++M; Nijk _ 22 2 %R&R = 100%. σ + σ SS = – y , y ( ) 1 β ε P i.. =2 1,…, i%R&R = 1,…,i=1 k2 β = 1,…, = µ 2+M; τ + + εijk n. =N; jyσ ijkτ + σ β + iσ ε 100%. j 1 2 2 2 σ + σ + σ N – 1 . MSP =i SS /( ) τ =jµ=β = 1,…, yijkN; +1,…, τi +ε M; βj +k ε=ijk1,…, n. P When a tolerance (TOL) the correspond2 is available, 2 σ + σ β ε i = 1,…, N; j = 1,…, M; k = 1,…, n. ing %R&R%R&R is defined as equation three: = 100%. 1 2 2 2 2 ++σσ σ2τ +σσ βσ β2 + ε σ ε2 ε 100%. %R&R1 = _5.15_ 2β 2 %R&R2 M= 5.152 2σ β2 σ+2 σ+2ε2 σ100% σ SSO = , β +β σ ε ε y.j.σ– τy+ ( ) TOL j=1 %R&R = 100% 100%. %R&R = 2 2 2 1 TOL 2 σ +2 σ 2 SSO /(M=– 1).σ τ +β σ β +ε σ ε 100%. MSO =%R&R 1 2 2 2 2 Notice the constant 5.15 corresponds to the span of a 2 σ + σ 5.15 βσ τ + εσ β + σ ε 100%normal distribu%R&R = Nzero _ a22 standard _ for 99% interval around 2 SSP = – yσ) β, + σ2ε y TOL (5.15 i=1 _ i.. _ data N 2 tion. After %R&R the experimental available, the analSS = – y) , are100% y = ( P i.. 2 i=1 2 2 TOL σ + σ = SS N – 1 . 5.15 /( ) _ _ β ysis begins MS with the computation of the P N P M n 2 εsums of squares 100% %R&R = SSMS = – y . y ( ) = SS N – 1 . ) notation 2 2 j=1 T i.. 2 P /( k=1 σ β + σfor TOL 5.15 (SS). ConsiderP thei=1 usual average over ε _dot N _ 2 100% %R&R = SS = (−y2i..– ythe ) , overall a subindex, Pand let TOL mean. The sum of i=1 y be _ 2 N _ SSmean y (y_i..––1(MS) squares and = SSPsquare MS _).) 2, due to part variation is M/( N P SSO = i=1 (_y.j.N– _y_) , _ 2 j=1 M represented by: = (yN – (y1y)i..2. ,– y) , SS =SS = SS MS P P /( OP .j. i=1 _1 . –_SS SSE = SSMN–– SS 2 , = SS=j=1 MSO SS O /( T ( y ) P– y ) , O P =/(SS N)i... – 1). MS /( i=1 = SS M – 1 MS P P MSEO= SSEO_/(NMn _ 2 – N – M + 1) . M SSO = MSP (=y.j.SS – Py/() N, – 1). j=1 _ 2 M _ The SS MS and MS SSOO = (yto.j.––operators SSdue 1y).) , _ is _represented as: /(M ON j=1 M n _(y2i..– y)2. SST = M _ i=1 j=1 k=1 SS = y – y_) , _ 2 N M n ( = SS M – 1 . MS /( ) O .j. 2 = {0,O MSj=1 – y) . σ̂ SS = Omax – MSE(y}/Mn. P _ k=1 _ 2i.. τ T M i=1 j=1 SS = yM.j. –– y1))., ( O = SS MS /( j=1 O O N = SS M /( M –_1). _ 2 as: MS O O The total SS, of course, isnrepresented SST = (yi..– y) . i=1 j=1 k=1 _ _ 2 N M n SS =SSE = SST – SSP (–yi..SS – Oy, ) . i=1 j=1– MSE k=1 }/ Nn. σ̂ 2β =T max { MS 0, O SSE = SS , _ 2 NT – SS M P – SS n O_ MSESS = SSE = /(NMn – N – M(y+i..–1)y. ) . T Finally, the SS=and attributable i=1 j=1 k=1 to gage variation is: MSE SSEMS – N – M + 1 . 2 /(NMn ) _ _ N M n SST = (yi..– y) . i=1 j=1 k=1 SSE = SST – SSP – SSO, Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ Σ MSE =SSE SSE=/(SS NMn – N –– SS M +, 1). – SS T – MSE P O 2 2 = max { 0, MS }/Mn. σ̂σ̂ = MSE. P ε τ 2 MSE = SSE NMn – N – M From the formulas for expected σ̂ = max { 0, MS – MSE }/ Mn. /( ),. practice the SSE =P SST – SSP MS, – +SS1in τ O MSE SSE = SSE = SS – SS–PN– –SSM , + 1) . /(NMn T O MSE SSE–/(MSE NMn}/–Mn. N – M + 1) . σ̂ 2τ = max {0,=MS P PAi 2= P (% R̂&R1 ≤ 30%). σ̂2β = max {0,MSO – MSE }/Nn. σ̂ 2τ = max {0, MSP – MSE }/Mn. σ̂ β = max {0,MSO – MSE }/Nn. σ̂ 2τ = max {0, MSP – MSE }/Mn. Equation 1 σ2β + σ2ε Equation Equation 31 0 %R&R = 1 σ2 +2σ2ε 2 100%. 2 β σ + σ + σ ε 100%. τ β %R&R1 = 0 0.2 σ2τ + σ2β + σ2ε Values 0.4 0.6 Equation 2 Equation 2 Equation 1 MC1_10 MC1_5 2 Equation 1 Five parts (MC1_5) and 10 parts (MC1_10). %R&R = 7.3% Equation 3.1 σ2β2+ σ2ε and reproducibility 5.15 repeatability %R&R =Equation percent 1 Equation 3 %R&R2 = Equation σ2β + σ2ε 100% 5.15 TOL Equation 3 %R&R2 = Equation 1100% TOL Equation 2 Frequency distribution of %R̂&R 1 for system Equation _ _ 2 Equation 3two Σ Σ SSP = Equation , (yi..– y)3.2 i=1 _Equation Equation _ 23 N SS = – y , y ( ) P i.. 2.5MS i=1 /( N = SS – 1) . P P Equation MSP = SSP /(N – 1). N 2 1 / Figure 2 Equation 3. Equation 3. 2 2 Equation 3 Equation 3.1 Equation 33.1 _ 2 M _ SS = Σ (yEquation – y) , 3 1.5 O Equation j=1 _ .j. _ 23.3 M SSO = Σ (y.j. – y) , 3 SSj=1O /(Equation M – 1)3.1 . MSO =Equation 1MSO = SSO /( M – 1) . Equation 3.1 Equation 3.2 Equation Equation 3.2 3.1 0.5 Equation 3.4 _ _ 2 N M n Equation SST = Σ Σ – y) . (y 3.1 i=1 j=1Σnk=1 _ i.. _ 2 N M SST = Σ Σ Σ (yi..– y) . i=1 j=1 k=1 0 Equation 3.2 2 Density = µ +2),τβi +~N(0,σ βj + εijk2) and ε ~ N(0,σ 2) ijk Then, assume τiy~N(0,σ ijk ε yijkN; =jµ=τ+1,…, βj2ε +k ε=βijk1,…, n. στ2βi j+ M; σ 5.15 i = 1,…, are independent. These are the variance components in100% %R&R2 = i = 1,…, N; j = 1,…, M; k = 1,…, n. volved in the R&R study.TOL The %R&R is represented as equation two: yijk = µ + τi + βj + εijk Equation 3. Equation 3. Equation 3. Equation 3. 0 0.5 Equation 3.2 Values Equation 3.3 3.2 Equation MC2_10 MC2_5 Equation 3.3 Equation 3.5 SSE = SST – SS – SSO, P FiveEquation parts (MC2_5) and 10 parts (MC2_10). %R&R 3.2 SSE = SS – SS–P N – SS , + 1) . MSE = =SSE –M /(NMn T repeatability O and %R&R percent reproducibility MSE =Equation SSE /(NMn 3.3 – N – M + 1) . 1 1 = 41% Equation 3. Equation 3. variance components are estimated as: 3.4 Equation 3.3 • Part-to-partEquation variation: Equation3.6 3.4 σ̂ 2τ = max {0, Equation MSP – MSE }/3.3 Mn. }/ σ̂ 2τ = max {0, MS – MSE Mn. Equation 3.3 P Equation 3.4 Equation 3. Equation 3. • Reproducibility or operator-related variation: Equation 3.4 Equation3.7 3.5 Equation σ̂ 2β = max {0,Equation MSO – MSE Nn. Equation 3.5 }/3.4 σ̂ 2β = max {0,MSO – MSE }/Nn. Equation 3.4 Equation 3.5 Equation Equation43.5 3.6 Equation Equation 3.6 σ̂ 2ε = MSE. σ̂ 2 = MSE. Equation 3.5 ε Equation 3. Equation 3. June 2013 Equation • QP 37 3. Equation 3. σ̂ 2β = max {0,MSO – MSE }/Nn. Equation 3.4 Equation 3.6 SSE = SST – SSP – SSO, σ̂ 2 = max {0,MS – MSE }/Nn. MSE =β SSE /(NMn –O N – M + 1). σ̂ 2ε = MSE. The simulation • Repeatability or gage-related variation: Let MC be the number of Monte Equation 3.5 Carlo replications. The σ̂ = max {0, MS – MSE }/Mn. σ̂ 2ε = MSE. P 2 τ Equation 3.7 For the measurement process to be useful, the vari- simulationP procedure is straightforward: Given σ2τ, σ2β, = P (% R̂&R ≤ 30%). Ai 1 and σ2ε, we generate MC sets of observations (see equa- ance component due to part-to-part variation should tion one) with τi ~N(0,σ2τ ) , βj~N(0,σ2β) and εijk~N(0,σ2ε), be much larger than other two).components. The =P % R̂&R ≤ 30% σ̂ 2β = Pmax {0,(the MS – 1MSE }/Nn. Ai O %R&R is estimated by substituting the estimates of the all independent, and Equation then estimate 4 equation four with: variance components in equations two or three, denotˆ ˆ ed by % R&R and % R&R . 1 2 Equation 3.6 P̂Ai = # (% R̂&R1 ≤ 30%)/ MC. First, consider PA1. Table 1 shows the results for sev- The exact distribution of these random variables is = # (% R̂&R1 ≤ 30%)/ MC. 2 Ai particularly = P̂MSE. difficult toσ̂obtain, because truncation is inε eral scenarios that represent measurement processes volved. Making the transformation nondifferentiable, and different %R&R1. Figures 1 and 2 (p. 37) show a we resort to simulation to compute: smoothed version of the histograms of the %R&R1 for PAi = P (% R̂&R1 ≤ 30%). Equation 4 with different Equation magnitudes 3.7 of variance components ^ (equation four) systems one and two when the number of parts is either five or 10. Equation 4 This is called the probability of acceptance of The estimated probabilities of acceptance corre- the measurement process. Other interests may spond to the area to the left of 0.3. Similar results are lead to consideration of other probabilities such P̂Ai 1= ≤# (% R̂&Ror ≤ P(% 30%)/R̂&R MC.1 > 50%), which as P(% R̂&R 10%) 1 obtained for systems three, four and five. For all cases, could be handled similarly to that in equation four, gram is obtained using the command “density plot” from hence we focus on equation four in this article. the R (2012) package.7 The area below any curve is one. Equation 4 we used MC=10,000. The smoothed version of the histo- Table 1 shows the probability of acceptance is practically the same for both numbers of parts. Figures 1 and 2 show the frequency distribution of the random Estimated probabilities of acceptance for different scenarios, %R&R1 / Table 1 st System sβ sε %R&R1 variable is similar for cases in which the measurement system is very good, such as system one, or poor, such as system two. The figures also show the distribution ^ of %R&R1 is just a little narrower for 10 parts. Parts = 5 Parts = 10 P̂A1 P̂A1 same variance components, but with two different tol- Now consider PA2. Table 2 shows the results for the 1 0.44 0.012 0.03 7.3% 0.9964 0.9999 erances. Figures 3 and 4 show a smoothed version of 2 3 1 0.9 41% 0.3284 0.3224 the histograms of the %R&R2 . 3 3 1 2 60% 0.0584 0.0413 In these cases, the agreement between distributions 4 4 1 1 33% 0.4813 0.5124 is even closer and, hence, the probabilities of accep- 5 8 0.55 0.2 7.3% 0.9847 0.9977 %R&R = percent repeatability and reproducibility st sβ sε TOL %R&R2 tance are also closer. Notice for system six and for both numbers of parts, there are some experiments in ^ which the %R&R2 is greater than 100%. If we include an interaction term, the results would Estimated probabilities of acceptance for different scenarios, %R&R2 / Table 2 System ^ be the same. Therefore, we conclude that as far as the probability of the outcome is concerned, it will not change noticeably if we use five or 10 parts in the study. The same approach could be used to evaluate Parts = 5 Parts = 10 P̂A2 P̂A2 the impact of changing the number of operators or the number of samples. As a consequence, if we have a good measurement 6 4 0.55 0.2 10 30.14 0.4291 0.4278 system, the probability that we accept it is basically 7 8 0.55 0.2 20 15.07 0.8862 0.8837 one with five or 10 parts. If we have a lousy measure- %R&R = percent repeatability and reproducibility 38 QP • www.qualityprogress.com ment system, the probability that we reject it is near statistics one for both five and 10 parts. If the measurement process is marginal (in the sense that the %R&Ri is a little above 30%), the probability of acceptance is near 50% for both Frequency distribution of %R̂&R2 for system six / Figure 3 number of parts, and hence the outcome is like that of a coin toss. 2 ^ If we use as a guideline %R&Ri<10%, the results should be similar because the frequency distributions are close. 8 Worth the effort? By focusing on the probability of acceptance of a measurement process, changing from 10 to five parts does not Density 1.5 1 modify this quantity substantially. Hence, this brings into 0.5 question the usefulness of doubling the experimental ef^ fort. The plots also show that the %R&Ri could have a large variability that decreases very little when we increase the 0 number of parts from five to 10. 0 ^ The distribution of values for the %R&Ri is similar and, 0.5 1.5 Values MC6_10 MC6_5 in some cases, very similar for both five and 10 parts. Five parts (MC6_5) and 10 parts (MC6_10). %R&R2 = 30.14% %R&R = percent repeatability and reproducibility Conclusions are similar if taken as a guideline to accept ^ a process if the %R&Ri<10%. All of these considerations are particularly important in situations in which measurements require analytical processes. QP ACKNOWLEDGEMENT 2 Frequency distribution of %R̂&R2 for system seven / Figure 4 Aguirre-Torres thanks Asociación Mexicana de la Cultura A.C. for its financial support in preparing this article. 4 REFERENCES AND NOTES 3 Density 1.Automotive Industry Action Group (AIAG), Measurement Systems Analysis (MSA) Manual, third edition, Chrysler, Ford, General Motors Supplier Quality Requirements Task Force, 2003. 2.John Raffaldi and Steven Ramsier, “5 Ways to Verify Your Gauges,” Quality Magazine, Vol. 39, No. 3, 2000, pp. 38-42. 3.Pingfang Tsai, “Variable Gauge Repeatability and Reproducibility Study Using the Analysis of Variance Method,” Quality Engineering, Vol. 1, No. 1, 1988, pp. 107115. 4.“Gauge R&R Study (Crossed),” white paper, Minitab 16 Statistical Software, Minitab Inc., 2012. 5.Ibid. See p. 11 for an example. 6.Douglas C. Montgomery, Design and Analysis of Experiments, John Wiley and Sons Inc., 2001. See chapter 12 for a more detailed discussion of the model that contains the interaction. 7.“The R Project for Statistical Computing,” software environment, R version 2.14.1, 2012, www.r-project.org/index.html. 8.Ibid. The simulation is programmed as an R function and is available from the authors. Email medmund@asq.org to request the additional information. 2 1 0 0 0.2 0.4 0.6 Values MC7_10 MC7_5 0.8 Five parts (MC7_5) and 10 parts (MC7_10). %R&R2 = 15.07% %R&R = percent repeatability and reproducibility VÍCTOR M. AGUIRRE-TORRES is a professor at the Instituto Tecnológico Autónomo de México. He holds a doctorate in statistics from North Carolina State University in Raleigh. MARIA TERESA LÓPEZ-ALVAREZ is an associate principal scientist at Kraft Foods in Mexico City. She holds a doctorate in statistics from North Carolina State University in Raleigh. López-Alvarez is a senior ASQ member and a Six Sigma Master Black Belt. June 2013 • QP 39 Getting a Good Read In 50 Words Or Less • The internet has forced libraries to focus on enhancing customer service through quality management. • LibQUAL+ is a tool to evaluate library customer service and engage patrons as partners to close the gap between experienced and expected service. • Examples of university libraries using LibQUAL+ illustrate using customer feedback to improve service. service quality Engaging patrons as partners helps libraries achieve total quality in service by Todd Bruns and Rendong Bai in the late 19th century, philanthropist Andrew Carnegie provided funding for the establishment of hundreds of libraries in the United States.1 His libraries were the first to feature open stacks—shelves of books that library patrons could browse by themselves. Before Carnegie’s libraries, most libraries had closed stacks, meaning librarians would locate items on behalf of the patrons and retrieve it for them. This change to open stacks enhanced the average person’s access to information and greatly improved the quality of service that libraries offered. Although librarians are generally service-oriented, their traditional training defines service to the patron as a book retrieved from the closed stacks rather than collaboration between the librarian and the patron. The patron is rarely perceived as being an equal partner in the transaction. June 2013 • QP 41 To achieve total quality in library service value, that customer needs were often not addressed. value must be technically defined quality as assessed The quality of a library collection can certainly be by the librarian—the value of the library collection and considered an indirect service to the patron, but with the accuracy of reference service—as well as function- the arrival of the internet and Google, libraries were ally defined quality as defined by the patron—the pa- no longer the sole source of information resources.6 To tron saying: “I received good service.” To achieve the compete, libraries must develop a strategy for assess- latter, a true librarian-patron partnership is required. ing and responding to customer perceptions of quality 2 Libraries began to see service competition for the of service. first time in the 1990s, due to the internet and Google. Service industries, such as libraries, had been slow The necessity to be more customer-focused became ap- to adopt quality management principles. By the 1990s, parent, and this resulted in the creation of LibQUAL+, a systematic approach to implementing total qual- a tool to quantitatively measure the quality of customer ity was underway.7 The concept of quality manage- service.3 ment was adapted and defined to fit within the library While LibQUAL+ serves as a first step toward in- realm,8-10 and these early attempts embraced quality creasing quality and value for patrons, library commit- management concepts, such as the formation of qual- ment to responding to customer survey results is es- ity circles among staff11 and the reorganization of the sential. LibQUAL+ can be used by a range of libraries, library staff hierarchy.12 but for purposes of this article, we will focus on its use in academic libraries. By the late 1990s, however, it became apparent librarians were “better at collection evaluation than ser- vice evaluation.”13 Although some libraries conducted Libraries embrace quality patron surveys, this was the exception and not the rule. Historically, the main role of a librarian has been to maintain the quality of the collection itself rather than Tool development tend to customer needs or expectations.4, 5 Librarians As patrons became customers and libraries found focused on developing their collections and buying themselves striving to prove their value, they needed books and journals to fit the needs of their patrons. to establish a means of quantifying and assessing cus- As noted by researcher Charles R. Martell Jr., librar- tomer needs and satisfaction. ies have been “more inclined to invest in things [than For many years, service sector industries relied on in people],” meaning staff development and assessing a simple Likert scale when testing customer satisfaction.14 Customers were asked to rate their experiences from one to five. What became apparent, however, was Number of libraries using LibQUAL+ per year / figure 1 that not only was it necessary to ask customers their perception of the quality of service, but it was also necessary to ask what their expectations were. Measuring the gap between experienced service and expected 400 350 Initiative, launched in 1999 by the Association of Re- 250 search Libraries (ARL) with a grant from the U.S. De- 150 100 2002 2003 2004 2005 2006 42 QP • www.qualityprogress.com 2007 2008 2009 2010 The increase in competition and the need to assess customer expectations led to the New Measurements 300 200 2000 2001 service is the “true measure of satisfaction.”15 partment of Education. Under this initiative, libraries used and adapted the ServQUAL tool to suit library services, resulting in the LibQUAL+ survey. LibQUAL+ was developed by ARL as a survey that 50 would test libraries’ strengths and weaknesses and, 0 because it was standardized, could also be used for libraries to compare themselves to their peer institutions.16 The survey contained 25 core questions in four service quality primary areas: access to information, affect of service, service that needed improvement to meet customer’s the library as a place and personal control of informa- expectations, but that additional analysis was needed: tion. The author of an article in the Journal of Library Ad- Each question was comprised of three parts: the ministration said, “The wealth of information to be customer’s perceived level of service, the customer’s gained from the survey results complicates the iden- minimal expectation of service and the customer’s de- tification of areas of clearest opportunity for improve- sired expectation of service. Since its development, ment.”21 the LibQUAL+ survey has evolved to comprise 22 A commitment to continual improvement is a key core questions in three primary areas: effect of service (nine questions), information control (eight questions) and the library as place (five questions) (see Table 1, p. 44). LibQUAL+ was first used in 2000. By 2003, 309 libraries used the survey (see Figure 1), To compete, libraries must develop a strategy for assessing and responding to customer perceptions of quality of service. and more than 1,000 libraries used the tool by 2007.17 Libraries may opt to participate annually in the survey or par- aspect of total quality, and Vanderbilt identified the ticipate in specific years. need for further evaluation via focus groups, follow- Although many libraries have used LibQUAL+ to assess patron satisfaction, it’s important to remem- up surveys, and follow-up with peer institutions that scored higher in the LibQUAL+ survey.22 ber that LibQUAL+ is simply a first step in the quality Two action items were identified for immediate improvement process. Although a reliable and valid response:23 First, in response to complaints about li- assessment tool, LibQUAL+ survey respondents are brary hours from undergraduate students, Vanderbilt self-selected, meaning that regardless of survey size launched a trial program making one of the campus there will always be some question of selection bias.18 libraries open 24 hours a day. In an executive summary LibQUAL+ provides libraries with significant amounts report, Vanderbilt identified a library and developed a of data, but follow-up activities with customers are a budget for a trial project in spring 2003.24 Beginning in necessity, as demonstrated by the experience of two spring 2003, the Vanderbilt Science and Engineering institutions that used LibQUAL+: Vanderbilt University Library established a 24-hour Monday through Friday in Nashville and the University of Pittsburgh. schedule that is still in place today, demonstrating the Vanderbilt University and the University of Pitts- success of that initiative. burgh libraries participated in the 2002 LibQUAL+ Second, a service quality improvement program survey of 46 ARL members across the United States. was proposed, which was to be developed by three Both academic libraries published papers in 2004 de- library directors at Vanderbilt. The service qual- tailing the results of their respective surveys. These ity improvement program team met regularly in 2003 LibQUAL+ surveys were chosen as a starting point to with the charge that “the team will limit its activities contrast different academic library approaches to im- to identifiable issues that seem to be easily fixed. Is- proving service, comparing information collected and sues that are not easily fixable or that deal with orga- reported during the same time period, and assessing nizational structure or policy will be referred to the the direction of each institution afterward. Library Management Council.”25 Reports of the team 19, 20 from 2003 ended with the Sept. 12, 2003, meeting. No Case study: Vanderbilt University publicly available final report or conclusions appear Vanderbilt University’s library reported results of the to exist. 2002 LibQUAL+ survey with a heavy emphasis on de- Vanderbilt has since participated in additional tailed statistics. The library noted it identified areas of LibQUAL+ surveys in 2006 and 2009. The 2006 survey June 2013 • QP 43 resulted in a number of substantial initiatives, including: • The addition of 14 work stations.26 The 2009 survey was conducted by new staff under • Implementation of a new electronic discovery the direction of a new dean of libraries and featured tool, “DiscoverLibrary,” that combined the library’s a heavier focus on marketing, campus outreach and catalog and other resources, simplifying patron project management than the previous surveys.27 searches. • Creation of a library delivery service for faculty. Case study: University of Pittsburgh • The addition of group study rooms. In contrast to Vanderbilt’s initially cautious approach in 2002, the University of Pittsburgh library 22 core questions of the current LibQUAL+ survey / table 1 listed several new initiatives and demonstrated a real commitment to quality improvement as a response to its 2002 survey. A particularly quality manage- Library service quality survey questions ment-oriented viewpoint is presented in Ratings: 1 (low) to 9 (high) Service level definitions: • Minimum—the number that represents the minimum level of service you would find acceptable. • Desired—the number that represents the level of service you personally want. • Perceived—the number that represents the level of service your library currently provides. the Pittsburgh library’s response to cus- When it comes to ... Question # Effect of service tomer perceptions: “Even if you disagree with the information offered [based on customers’ perceptions], you can’t say their perceptions are wrong. Your challenge is to address these perceptions ... .”28 A specific example illuminates the two institutions’ different responses to 1 Employees who instill confidence in users 4 Giving users individual attention sities’ surveys showed a negative balance 6 Employees who are consistently courteous between customers’ minimal require- the 2002 LibQUAL+ results. Both univer- 9 Readiness to respond to users’ questions ments for having complete collections of 11 Employees who have the knowledge to answer user questions journal titles and customer perceptions 13 Employees who deal with users in a caring fashion of the journals offered by the library. For 15 Employees who understand the needs of their users Vanderbilt’s library, this meant acknowl- 18 Willingness to help users 22 Dependability in handling users’ service problems Information control edging the negative balance coupled with a statistical analysis showing the complaint primarily came from faculty and that this complaint was common for 2 Making electronic resources accessible from my home or office 5 A library website enabling me to locate information on my own 7 The printed library materials I need for my work showed the same negative balance, but 10 The electronic information resources I need the response was more proactive: The 14 Modern equipment allows easy access to needed information library followed up with customers and 16 Easy-to-use access tools allow me to find things on my own discovered patrons wanted complete col- 19 Making information easily accessible for independent use lections of specific journals. 20 Print and electronic journal collections I require for my work Library as place many libraries.29 In contrast, the Pittsburgh library also The library then developed a dual service response to address this need. First, electronic back issues of the journals 3 Library space that inspires study and learning 8 Quiet space for individual activities 12 A comfortable and inviting location 17 A getaway for study, learning or research it was provided.30 This dual approach 21 Community space for group learning and group study improved access to the specifically re- 44 QP • www.qualityprogress.com being requested were obtained. Second, the off-site storage facility was relocated closer to campus, and shuttle service to service quality quested journals and made obtaining print back issues Customer-focused approach more convenient for patrons, beginning the process of Although Vanderbilt had responded to its first improving customers’ perception of service. LibQUAL+ survey tentatively by adapting only one (al- Pittsburgh’s response exemplifies a critical aspect beit quite important) identifiable change in response of quality management—embracing customer-defined to customer desires, subsequent survey responses value. For the customer, the product or service must be what the customer wants, of expected quality and be delivered courteously and promptly.31 By taking steps to obtain the specific e-journals customers wanted, Pittsburgh addressed the issue of providing the product customers desired. Pittsburgh’s response exemplifies a critical aspect of quality management—embracing customer-defined value. The relocation of the off-site storage facility improved speed of access, addressing demonstrated a movement toward a more engaged customer desires for faster access. Rather than consid- approach. In particular, the implementation of “Dis- ering lack of access to journals a common complaint, coverLibrary” demonstrates the merging of customer- Pittsburgh followed the quality management philoso- defined functional quality and librarian-defined techni- phy by letting the customer define the value and then cal quality. responding accordingly. At the University of Pittsburgh, a customer-focused The Pittsburgh library also identified other areas for approach was in place from the beginning. By ad- improving service and took steps to do so. A team was dressing customer perceptions in customer’s terms, established to rethink and improve public services. It Pittsburgh fully used its LibQUAL+ survey results to followed up the LibQUAL+ survey results by inves- improve customers’ perceptions of the service value. tigating national research studies and holding open The commitment to service quality has led Pittsburgh meetings with staff. Following up on LibQUAL+ results to focus efforts on developing local assessment tools proved to be pivotal in establishing new services, in that fit its patron base. particular the creation of a peer-to-peer library service To further ensure success, libraries must not sim- plan in which students could get assistance in their ply focus on patron perception discovery tools such as dorms and in the library from other students. LibQUAL+, but also engage patrons as full partners in The Pittsburgh library has since participated in an- all library endeavors. nual LibQUAL+ surveys (with the exceptions of 2008 and 2010), although further results and initiatives have Patrons as partners: UCLA not to date been published. As of 2011, Pittsburgh opt- In 2001, the University of California at Los Angeles ed to develop its own survey “to focus … survey efforts (UCLA) library needed to obtain a new integrated li- toward [their] individual community’s concerns and be brary system (ILS), a system that combines elements able to utilize the data in a more agile manner.”32 for acquiring, cataloging and loaning materials. The This shift away from the LibQUAL+ tool demon- library made the commitment to involve faculty and strates the importance of connecting with the patron students in the process of assessing different ILS op- to ensure quality is a bigger issue than the use of a tions. particular tool. While there are clear positives of us- After three ILS options had been identified, the ing LibQUAL+, particularly for libraries that have overview committee tasked with the project estab- little experience assessing customer-defined value, lished an advisory group of functional sponsors— ultimately a library must determine the best way to faculty and students who assisted the overview com- keep an ear to the ground and maintain that patron mittee. The functional sponsors worked with the connection. committee to craft a survey focusing on questions June 2013 • QP 45 service quality important to customers and to develop test runs of the ILS options that focused on how patrons would use the service.33 Bringing in patrons as actual partners added clarity to the survey questions and identified which features of the ILS were most important to patrons. Ultimately, the UCLA library’s selection was influenced by the participation of its customers. This example demonstrates the quality management strategy of bringing in customers as partners in innovation.34 By involving patrons in the entire process of deciding on an ILS, UCLA was addressing customer needs from the very beginning. The inclusion of faculty and students in crafting a customer survey about the ILS ensured UCLA was asking the right questions. Involving customers in the test runs of three different ILS options gave patrons a voice in the process, ensuring the chosen system had the attributes customers desired. Committing to quality The development of LibQUAL+ as an assessment tool has assisted libraries to move closer to achieving quality management. As a tool, LibQUAL+ is beneficial, but embracing quality management organizationally and committing to service improvement as defined by customers is crucial. Libraries that commit to a fully engaged quality management philosophy can provide their patrons with the products and services they want, at a higher Journal, Vol. 113, No. 7, 1988, p. 52. 12. Fitch, “Turning the Library Upside Down: Reorganization Using Total Quality Management Principles,” see reference 9. 13. Saunders, “The LibQUAL+ Phenomenon: Who Judges Quality?” see reference 3. 14. Ibid. 15. Ibid. 16. Ibid. 17. Ibid. 18. Ibid. 19. Amy Knapp, “We Asked Them What They Thought, Now What Do We Do? The Use of LibQUAL+ Data to Redesign Public Services at the University of Pittsburgh,” Journal of Library Administration, Vol. 40, No. 3/4, 2004, pp. 157-171. 20. Flo Wilson, “LibQUAL+ 2002 at Vanderbilt University: What Do the Results Mean and Where Do We Go From Here?” Journal of Library Administration, Vol. 40, No. 3/4, 2004, pp. 197-240. 21. Ibid. 22. Ibid. 23. Jean and Alexander Heard Library, “LibQUAL+ Survey: Results From the Survey, 2002,” 2002, http://libstaff.library.vanderbilt.edu/Admin/LQ/ LibQUALReport.htm#Appendices (case sensitive). 24. Jean and Alexander Heard Library, “LibQUAL+ 2002 Executive Summary,” 2002, http://libstaff.library.vanderbilt.edu/Admin/LQ/LibQUALReport.htm (case sensitive). 25. Jean and Alexander Heard Library, “Service Quality Improvement Team,” 2011, http://libstaff.library.vanderbilt.edu/service. 26. Jean and Alexander Heard Library, “LibQUAL+ 2009,” www.library. vanderbilt.edu/libqual2009/faq.html. 27. LibQUAL+ Share Fair, “Proceedings from the 2009 American Libraries Association annual conference,” 2009, www.libqual.org/documents/admin/ Share_Fair_booklet_2009_2.pdf (case sensitive). 28. Knapp, “We Asked Them What They Thought, Now What Do We Do? The Use of LibQUAL+ Data to Redesign Public Services at the University of Pittsburgh,” see reference 19. 29. Wilson, “LibQUAL+ 2002 at Vanderbilt University: What Do the Results Mean and Where Do We Go From Here?” see reference 20. 30. Knapp, “We Asked Them What They Thought, Now What Do We Do? The Use of LibQUAL+ Data to Redesign Public Services at the University of Pittsburgh,” see reference 19. 31. David L. Goetsch and Stanley Davis, Quality Management for Organizational Excellence, sixth edition, Prentice Hall, 2010, pp. 144, 154. 32. Personal communication with John Fudrow, University of Pittsburgh Library, Feb. 24, 2012. 33. Terry Ryan, “Turning Patrons Into Partners When Choosing an Integrated Library System,” Computers in Libraries, Vol. 24, No. 3, 2004, pp. 6-8, 54, 56. 34. Goetsch, Quality Management for Organizational Excellence, see reference 31. quality than they expect and in a timely manner. QP Bibliography References 1.George Sylvan Bobinski, Carnegie Libraries: Their History and Impact on American Public Library Development, American Library Association, 1969, p. 191. 2.William B. Edgar, “Questioning LibQUAL+: Expanding its Assessment of Academic Library Effectiveness,” Libraries and the Academy, October 2006, Vol. 6, pp. 445-465. 3.E. Stewart Saunders, “The LibQUAL+ Phenomenon: Who Judges Quality?” Reference and Users Quarterly, Vol. 47, No. 1, 2007, pp. 21-24. 4.Charles R. Martell Jr., “The Client-Centered Academic Library: An Organizational Model,” Journal of Academic Librarianship, Vol. 9, No. 5, 1983, pp. 292-293. 5.Saunders, “The LibQUAL+ Phenomenon: Who Judges Quality?” see reference 3. 6.Ibid. 7.Kitty Mackey and Terry Mackey, “Think Quality! The Deming Approach Does Work in Libraries,” Library Journal, Vol. 117, No. 9, 1992, pp. 57-61. 8.Ibid. 9.Donna K. Fitch, Jean Thomason and Elizabeth Crabtree Wells, “Turning the Library Upside Down: Reorganization Using Total Quality Management Principles,” Journal of Academic Librarianship, Vol. 19, No. 5, 1993, pp. 294-299. 10. Rush Miller and Beverly Stearns, “Quality Management for Today’s Academic Library,” College and Research Libraries, Vol. 55, No. 7, 1994, pp. 406-422. 11. Janice Desirey, Christopher Dodge, Nancy Hargrave, Randi Larson and Susan Nesbitt, “The Quality Circle: Catalyst for Library Change,” Library 46 QP • www.qualityprogress.com Zbaracki, Mark J., “The Rhetoric and Reality of Total Quality Management,” Administrative Science Quarterly, Vol. 43, 1998, pp. 602-636. Todd Bruns is the institutional repository librarian at Eastern Illinois University in Charleston. He has a master’s degree in library and information studies from the University of Wisconsin-Madison and a master’s degree in technology management from Eastern Illinois University. Rendong Bai is an assistant professor in the School of Technology at Eastern Illinois University in Charleston. He received a doctorate in computer science from the University of Kentucky in Lexington. Standards Outlook BY John E. “Jack” West The Missing Key Clearing up misunderstandings about ISO 9001 conformance past perceptions about conformity product designs be validated to to ISO 9001 suggest an organization may ensure they will meet require- have a good quality management system ments for given applications, and (QMS) that conforms to ISO 9001, but it requires the product be verified it can still produce junk. In fact, ISO to ensure those requirements are 9001 contains many requirements that, met. The overall requirements taken together, should provide reason- of clause 7.3 are summarized in able assurance that the output of the Table 2 (p. 48). system—the organization’s products and services—will indeed meet customer Pinpointing the problems requirements. What, then, is the issue? There There is also the myth that the design and development clauses of ISO 9001 can- are several: Innovation and product not be applied to highly innovative product definition. Some experts developments because they are too restric- express the opinion that clause tive. Yet, we continue to see product and 7 does not adequately address service disasters in the marketplace that situations in which the customer could readily have been prevented by judi- does not know he even needs the cious application of these same design and product, much less its detailed development requirements. functional requirements. For ex- First, the scope of ISO 9001:2008 clear- ample, many products available identify product requirements that are ly states the intent of the document to today were not even a gleam in the user’s necessary but not stated by the customer. meet customer requirements (see Table eye a few years ago. It is argued that Also note that the product design must 1). ISO 9001, clause 5.3, also requires that the formal design process of ISO 9001 is include as inputs the organization’s own the quality policy include commitments too narrow to include such innovative requirements for the product, and that to meeting requirements and continual designs, so it is often ignored. the product design must be verified and improvement of the QMS. Clause 7 on product realization A careful reading of ISO 9001 reveals validated to ensure that the product this concern is not valid. What can stuffy meets specified requirements and will be requires identification of customer old ISO 9001 do to control such a situ- suitable for a given application. requirements. Clause 7.3 requires that ation? A lot, actually. Note the need to No product development should go forward in the absence of a robust Excerpt from ISO 9001:2008, Clause 1.1—scope, general / process to define the design inputs and outputs, and to verify and validate prod- Table 1 “This International Standard specifies requirements for a quality management system where an organization a) b) uct functionality. Design review. After requirements are documented and a product development plan has been created, the actual needs to demonstrate its ability to consistently provide product that meets customer…requirements, and work of developing a product can pro- aims to enhance customer satisfaction through the effective application of the system, including processes for continual improvement of the system and the assurance of conformity to customer and applicable regulatory requirements.” often ignored or treated in an extremely ceed. One activity of the process that is superficial manner is design review (subclause 7.3.4). June 2013 • QP 47 standards outlook The purpose of this requirement is to intended to address more than just the Design reviews are intended to iden- ensure timely release of a new product question of whether the product will tify issues, discuss possible resolutions that fully meets the needs of the custom- meet specified requirements. An effective and determine appropriate follow-up ers. It also can contribute significantly to design review should address all the abili- early enough to minimize cost and sched- reduced cost. The intent of the standard ties associated with a new product—man- ule impact. is to involve all appropriate individuals as ufacturability, deliverability, testability, soon as it is feasible so they can under- inspectability, capability, serviceability, of design output to input requirements stand and address life-cycle issues early repairability, availability and reliability— to ensure the development project is on in the process. as well as issues related to inventory and track. Often this step is just ignored but, production planning. if done properly, it determines objec- Design and development review is Verification is simply a comparison tively whether the requirements will Validation—summary of requirements / Table 2 7.2.1 Determination of requirements related to the product 7.3.1 Design and development planning It is often an incremental process and, for complex innovative products and services, can require years of work. In The organization shall determine: • Requirements specified by the customer, including the requirements for delivery and postdelivery activities. • Requirements not stated by the customer but necessary for specified or intended use, when known. • Statutory and regulatory requirements related to the product. • Any additional requirements determined by the organization. fact, for entrepreneurial start-ups, this The organization shall plan and control the design and development of product. Required activities include determining: • Design and development stages. • Review, verification and validation activities appropriate for each design and development stage. • Responsibilities and authorities for design and development. It is also required that interfaces be managed to ensure effective communication. Verification is not complete until you can 7.3.2 Design and development inputs Inputs relating to product requirements shall be determined. These inputs shall include functional and performance requirements. 7.3.4 Design and development review Systematic reviews of design and development shall be performed: • To identify any problems and propose necessary actions. • To evaluate the ability of the results of design and development to meet requirements. 7.3.5 Design and development verification be met by the ultimate product design. Verification shall be performed to ensure that the design and development outputs have met the design and development input requirements and records shall be maintained. 7.3.6 Design and development validation Design and development validation shall be performed to ensure that the resulting product is capable of meeting the requirements for the specified application or intended use, when known. 7.3.7 Control of design and development changes Design and development changes shall be identified and records maintained. The changes shall be reviewed, verified and validated, as appropriate, and approved before implementation. Source: ANSI/ISO/ASQ Q9001:2008 (ASQ Quality Press) 48 QP • www.qualityprogress.com step develops confidence in the startup company’s business case—its very reason for being. When verification is complete, the organization should have high confidence its product will work as intended. answer the question: Will the product do what I claim it will do? Validation is a confirmation the product will meet actual application needs. This is best conducted directly with users and may involve having them use product prototypes or development models. For services, it may involve test sites where the service is piloted with actual clients. User feedback can be critical in making the best final product design decisions. When users provide negative feedback, you have a golden opportunity for designers to use their innovative talents to develop customer-friendly solutions. It is especially important to rigorously address verification and validation for the design and development of innovative new products or extensions of existing product lines in which customer requirements are vague or nonexistent. Customer dissatisfaction with newly released products can be harmful—even devastating—to a new product launch and can even negatively influence the marketing of unrelated products. Control of the design and development process (see Figure 1) can at least minimize Design and development planning process / figure 1 the occurrence of widespread “infant mortality” (for example, new batteries in Design and development planning a new airplane). Design validation is not complete until you can face your investors and tell them Design verification how the product will meet your potential customer’s needs for specific applications. Will the customer be willing to pay enough for the new product to cover all Customer needs Design inputs our front-end costs with a reasonable return on investment? Process capability and capac- Design process phases and reviews Design outputs Product Risk analysis and preventive action ity. There is a note in clause 7.1 of ISO Design validation 9001:2008 that says: “The organization may also apply the requirements given in 7.3 to the development of product realization processes.” Perhaps the most significant difficulty with ISO 9001:2008 is that this note does expenditures that had been grossly take corrective action to eliminate the underestimated. causes of these problems should not An organization is not ready for prod- make claims of conformity to ISO 9001. not make the use of the design process uct or service rollout until the product We must ensure our systems deliver mandatory for production and service realization and delivery processes are conforming product to our customers. delivery processes. Often, this notion is capable of delivering conforming product The standard requires it, and the credibil- completely ignored until the organiza- at the needed volume. Note there are not ity of ISO 9001 certification demands it. It tion discovers it cannot produce enough two things here. There is only one: ability is the output of our QMS that matters to conforming product to meet customer to deliver the required volume of product our customers. demand, or worse, cannot deliver in suf- that meets customer needs. ficient volume to achieve enough positive Unfortunately, these messages have not adequately penetrated the world of Keep the product in mind the quality professional. There is con- We need to stop perpetuating the myth siderable hope, especially for new small intended. Pilot runs are successful and that organizations can truly be in confor- businesses. Venture capitalists and other delight test customers. But what happens mance with ISO 9001 and still produce investors are asking new entrepreneurs when the demand for volume delivery product that does not meet customer re- questions related to these subjects before hits the organization? In my experience, quirements. Why does the myth persist? committing to investments. Due dili- scale-up of a new product is often the Because standards implementers and gence—and perhaps more importantly, Achilles’ heel of the organization. At the auditors tend to focus on conformance common sense—demand it. QP point when solid process capability and to the details of ISO 9001 and often lose capacity are needed, it is too late to think sight of the basic requirements. Never about these subjects. They must be ad- lose sight of the product. cash flow to survive. Prototypes get tested and work as dressed during product design. I have even seen situations in which A claim that an organization conforms to ISO 9001 should mean the organi- high-tech products could be produced zation’s customers will consistently easily in low volume, but the entrepre- receive a product or service that meets neurs involved never realized that scale- its requirements. An organization that up of their great idea was technologically consistently produces nonconforming impossible or required capital equipment product and is unable or unwilling to John E. “Jack” West is a member of Silver Fox Advisors in Houston. He is past chair of the U.S. Technical Advisory Group to the International Organization for Standardization Technical Committee 176 and lead delegate of the committee responsible for the ISO 9000 family of quality management system standards. He is an ASQ fellow and has co-authored several ASQ Quality Press books. June 2013 • QP 49 Measure For Measure BY Christopher L. Grachanen Know Your Surroundings Environmental influences impact tests and calibration Ambient conditions during tests Initial impact as a rule, specified at 20°C (68°F)—often and calibration activities can have a big in- The acceptance of a metallic rod used in a referred to as standard temperature. This fluence on measurement results, depend- precision mechanical device requires that is the reason many precision measure- ing on the measurement scenario. Some of its length be held to 1 inch plus or minus ment laboratories are controlled to 20°C these influences may be obvious—some 50μin (1 +/- 0.005%). The rod is made (68°F)—to forego having to apply correc- not so obvious. Some of these influences of aluminum to conserve weight, and is tion factors for dimensional and volu- affect the device being measured; some manufactured and measured in a machine metric measurement made at other than affect the measurement system; and some shop with an ambient temperature of 25°C standard temperature. affect both—though not necessarily pro- (77°F). After leaving the machine shop, portionally between the two. the rod is taken to a quality inspection lab deemed the critical influence affecting to be measured again and is found to be the measurement. If the aforementioned out of tolerance. measurements had been performed in an Environmental influences include temperature, humidity, vibration, barometric pressure, ambient particulate concentra- The quality inspection lab has an ambi- In this example, temperature was environment subject to excessive ambi- tion, magnetic fields, electrical fields, ent temperature of 22°C (72°F). Due to ent particulate concentration (as can be ambient light and sound. Depending on aluminum’s inherent linear temperature possible in a machine shop with milling the measurement application, there may expansion coefficient of 23.1E-06 (23.1μin operations), foreign material may rou- be only one environmental influence that per inch per 1°C), this equates to 139% tinely lodge between the rod’s surface and an operator should be concerned with of the 50μin tolerance (3°C difference x the styluses of the measurement device, regarding unwanted measurement biases. 23.1μin = 69.3μin) that can be attributable adversely affecting measurement results. to just the temperature difference be- In this scenario, ambient particulate con- is significant requires an understanding tween the two locations without consider- centration would be another environmen- of the measurement system and what is ing any other source of measurement bias. tal influence to consider. being measured. A simple example will It is noteworthy that linear temperature Determining if a particular influence illustrate this point. expansion coefficients for materials are, Continue monitoring Not only is it important to evaluate environmental conditions initially when measurements are being made, but it also may be necessary to monitor environment conditions during the duration of measurements. This depends on the measurement application and the length of time to make the measurements. Another example will illustrate this point. The calibration of air discharge electrostatic (ESD) simulators is influenced by moisture content in the air as characteristically disclosed by manufacturers in terms of relative humidity operating range. ESD simulator calibration is characteristically based on relative agreement between the simulator’s output impulse and the International Electrotechnical Commis- 50 QP • www.qualityprogress.com Determining if a particular influence is significant requires an understanding of the measurement system and what is being measured. sion’s accepted waveform specifications started at 30% relative humidity could in- evaluating the measurement results of for the ESD phenomena. Calibration of an clude some measurement readings taken others and consequently making business ESD simulator may take from 30 minutes outside the manufacturer’s operating lim- decisions based on their evaluations. to 2.5 hours, depending on the simulator’s its (a no-no when calibrating to manufac- sophistication and installed options. turer’s published specifications). It should surement data with descriptive verbiage It is always prudent to accompany mea- be noted that most humidity monitoring in regards to the measurements made. For reputable manufacturer specifies one of devices (sensors) have an accuracy instance, the following descriptive ver- its popular ESD simulators over a relative specification of 2 to 5% relative humidity, biage could be used in the first example: humidity range of 20 to 80%. It has been adding to the measurement error. Continuing with this example, a “Measurements were made with digital caliper (serial No. 12345) on Feb. 14, 2012, my experience that generic room temperature controllers can create relative Avoid assumptions after cleaning and a one-hour soak time at humidity swings of 10% or more during an The examples show that environmental 23°C and 45% relative humidity.” hour while attempting to maintain a set influences can cause measurement errors temperature (dependant on surrounding and can even invalidate a calibration measurement helps ensure everybody is ambient conditions, environmental system when deviating beyond acceptable limits. on the same page and helps avoid errone- design/feedback and set point tempera- It is always a best practice to question ous assumptions, which can be a real cost ture). measurement applications in terms of avoidance in terms of product reworks environmental influences. This advice and field failures. QP Unless relative humidity is continuously monitored during the calibration, it also extends to those who do not actually is conceivable that a 2.5-hour calibration make measurements but are tasked in more to measure QP has featured the Measure for Measure column for more than a decade. Check out past columns, including those written by Christopher L. Grachanen, under the Departments and Columns tab at www.qualityprogress.com. Including descriptive verbiage about Christopher L. Grachanen is a master engineer and operations manager at Hewlett-Packard Co. in Houston. He earned an MBA from Regis University in Denver. Grachanen is a co-author of The Metrology Handbook (ASQ Quality Press), a senior member of ASQ, an ASQ-certified calibration technician and the treasurer of the Measurement Quality Division. Do the Write Thing in 2013 We’re looking for writers who want to contribute articles about these topics: September Standards and auditing If you want to write for a specific issue, please submit your article at least three months before the issue date. If you miss that deadline, don’t worry. Send the article and we might use it in a different issue. And if there’s a topic you would like to write about that’s not listed as one of October Social responsibility our featured topics in the editorial calendar, don’t let that dissuade you November Global quality from submitting the manuscript. We will publish all accepted articles, whether they align with an issue theme or not. Visit www.qualityprogress.com and click on “Author Guidelines” at the bottom of the page for more details. QUALITY PROGRESS June 2013 • QP 51 Quality in the First Person BY John W. Jennings III It’s a Process DMAIC helps manage diabetes More than half of Americans will have diabetes or be pre-diabetic by 2020.1 I have body’s ability to regulate its BSL. Statistical thinking supports BSL control Design of experiments can determine factor levels for minimizing glucose level been a Type 2 diabetic for 40 years—about with three basic assumptions: variation after exercise. This approach also as long as I’ve been in the quality field. • Work occurs in a system of intercon- can help determine calorie limits, meal Currently, there is no cure for diabetes. But nected processes. it can be managed. I found that statistical • Variation exists in all processes. thinking and the define, measure, analyze, • Understanding and reducing variation improve and control (DMAIC) method can are the keys to success. times and what to eat. Analysis of variance (ANOVA) can analyze the effects of medication changes. In August, I began a new weight loss medication, Victoza. Using ANOVA single-factor reduce variation in diabetes management Measure test, I evaluated my BSL averages, which An Excel spreadsheet is useful for recording are shown in Figure 2. The null hypothesis Define BSL readings, medication use and comments. represents June and July daily averages Insulin converts glucose into cell nutrients. I use Excel functions to calculate averages without Victoza. The test shows u1 and u2 Type 1 diabetics produce little to no insulin and standard deviations on a daily, weekly equals August’s daily averages with Victoza and need insulin injections. Type 2 diabet- and monthly basis. Figure 1 shows an ex- (u3). Because the null hypothesis is ac- ics either produce too little insulin or have ample of a truncated data collection matrix. cepted, I can determine the medication has and create control. become insulin resistant. They control their not affected my BSL. blood sugar levels (BSL) by modifying their Analyze diet, exercising and, in some cases, taking Landmark studies published in the 1990s Improve medication. showed the benefit of tightly controlled The hardest part about managing diabetes Type 1 and 2 diabetics use a device to BSLs in the management of both types is changing your lifestyle into a disciplined measure their BSLs. Most diabetics aim to of diabetes.4, 5 Key factors that influence process. That usually means participation in maintain a BSL between 80 and 150 mil- BSL—exercise, weight and medication— an exercise program, controlling carbo- 2 ligrams per deciliter. Monitoring provides are identified by doctors. Modifications to hydrate intake, monitoring BSLs, taking input data for statistical tools to evaluate the these factors help establish a zone of tight medication and sometimes abstaining from level of control the diabetic has on his or her control. alcohol and nicotine. 3 Doctors set BSL goals and diabetics achieve them with lifestyle changes. A dietician helps plan a proper diet and identify foods to avoid. Medication may be a part of the management plan. Support from your family, friends and medical team is crucial. Control Statistical process control (SPC) is used to monitor and analyze variation in manufacturing processes. Processes are in control when there are no special causes of variation present. A diabetic’s BSL is similar to a process with special cause variation because BSL regulation does not occur automatically. 52 QP • www.qualityprogress.com Data collection matrix / figure 1 Blood sugar levels (BSL) and insulin dosages Date AM BSL Dosage Noon BSL Dosage PM BSL Comments Dosage Night BSL Dosage Victoza started 8/31/2012 9/1/2012 124 16 138 20 279 36 204 9/2/2012 147 16 106 20 185 32 158 9/3/2012 119 16 160 22 156 32 158 9/4/2012 154 18 116 20 99 30 130 9/5/2012 172 18 106 20 89 30 157 Average 141.8 134.8 143.8 161.4 Standard deviation 20.9 36 46 42.4 Most diabetics track their BSL daily and may need to modify their food intake, exercise and medication accordingly. Figure 3 is an X-bar chart of my daily BSL averages 2 Ate too much in afternoon Started new Apridra pen Started new Lantus pen ANOVA single factor test ANOVA: Single factor α / figure 2 0.05 Ho: u1= u2= u3 Summary in September. There were several days with Groups Count Sum Average Variance daily averages on the high end due to special June 30 4,462 149 444.6713 cause variation. My BSL was managed with July 30 4,350 145 386.7152 more insulin, less food and more exercise. August 30 4,278 143 199.9885 It is possible for diabetics to experience a high quality of life. The key to achieving Accept null hypothesis ANOVA because p > 0.05 (means are the same) treatment goals is a combination of lifestyle Source of variation changes, close monitoring and heeding doc- Between groups tors’ advice. DMAIC and statistical think- Within groups health. QP References 1. Bill Berkrot, “Half of Americans facing diabetes by 2020: report,” Reuters, Nov. 23 2010, www.reuters.com/ article/2010/11/23/us-unitedhealth-diabetes-idusTRE6AM0NH20101123. 2. American Diabetes Association, www.diabetes.org. 3. “Hyperglycemia: When Your Blood Sugar is Too High,” Diabetic Gourmet, http://diabeticgourmet.com/articles/206.html. 4. National Institute of Diabetes and Digestive and Kidney Diseases, “The Diabetes Control and Complications Trial,” Health Trends, 1994, http://diabetes.niddk.nih.gov/dm/pubs/control. 5. B.M. Margetts, United Kingdom Prospective Diabetes Study (UKPDS) 13, BMJ, 1995, www.ncbi.nlm.nih.gov/ pubmed/7833731. John W. Jennings III is a compliance manager for Vertex Business Services in Richardson, TX. An ASQ fellow, Jennings is an ASQ-certified Six Sigma Black Belt, quality engineer, reliability engineer, software quality engineer, quality auditor, quality technician, biomedical auditor and a manager of quality/organizational excellence. He is an RABQSA-certified ISO 9001:2008 lead auditor, is certified by the National Association of Radio and Telecommunications Engineers as an electrostatic discharge control engineer, and is a senior examiner for the Texas Quality Award program. DF MS F P-value F crit 0.832202 0.439 3.101296 572.208 2 286.104 29,909.87 87 343.7917 ANOVA = Analysis of variance SS = Sum of squares DF = Degrees of freedom MS = Mean square F = (found variation of the group averages)/ (expected variation of the group averages) P-value = The probability of obtaining a test statistic at least as extreme as the one that was actually observed, assuming that the null hypothesis is true. F crit = The value of F at the threshold probability α of mistakenly rejecting a true null hypothesis. Daily blood sugar level averages 211 / figure 3 201.6 UCL 191 171 151 Upper specification limit 150 CL 145.5 131 111 91 71 LCL Lower specification limit 89.3 80 9/1/2012 9/2/2012 9/3/2012 9/4/2012 9/5/2012 9/6/2012 9/7/2012 9/8/2012 9/9/2012 9/10/2012 9/11/2012 9/12/2012 9/13/2012 9/14/2012 9/15/2012 9/16/2012 9/17/2012 9/18/2012 9/19/2012 9/20/2012 9/21/2012 9/22/2012 9/23/2012 9/24/2012 9/25/2012 9/26/2012 9/27/2012 9/28/2012 9/29/2012 9/30/2012 approach for making decisions about your Daily average mg/dl ing provides a disciplined and data driven SS mg/dl = milligrams per deciliters June 2013 • QP 53 Career Corner BY Henry J. Lindborg Stake Your Ground Unearthing the origins of stakeholder management Stakeholder analysis has become stakeholders to ensure alignment with of Pennsylvania in Philadelphia, where he standard practice among quality prac- strategic objectives. Higher education’s realized stakeholder theory was underde- titioners. Whether conducting strategic Academic Quality Improvement Program veloped and deserved attention. “I never planning, exploring an organization’s lead- of the Higher Learning Commission has a looked back,” Freeman said. ership and governance using the Baldrige similar requirement. As management tools criteria, chartering Six Sigma projects, and approaches become more sophisticat- Separation fallacy assessing enterprise risk or auditing cor- ed and commonplace, we may forget their In the 1980s, concerns about economic porate social responsibility, it’s important origins and broader implications. This competitiveness led corporations and to identify individuals or groups affected month, let’s take a look at the discipline of government to embrace strategic formu- by an organization’s actions. In light of stakeholder management. las, such as Michael Porter’s “five forces” volatility and complexity in organiza- Modern stakeholder thinking arose in model.3 Freeman sought to understand the tions’ internal and external environments, the 1930s in midst of a legal debate on the ethics of strategic decisions and wondered stakeholder analysis can guide them to act responsibilities of corporations to their if it was possible to integrate ethics with responsibly and mitigate risk. stockholders and other constituencies. So- organizational strategy. He discovered ciety began to ask, “What do corporations widespread “separation fallacy”—busi- in ASQ’s body of knowledge for quality owe those who aren’t investors?” Legally, ness decisions were separate from ethical managers. It is practiced by revenue-driv- stockholders came first. As health, safety, considerations. Making decisions without en corporations and nonprofits alike—in- environmental and community relations factoring in ethics can lead to dire conse- cluding the United Nations and the World evolved, other stakeholder groups were quences, as we’ve witnessed in dramatic Bank. The Malcolm Baldrige National included. Few CEOs today would repeat instances of corporate misconduct—from Quality Award requires applicants to iden- J.P. Morgan’s statement, “I owe the public Enron and WorldCom to Lehman Brothers tify and differentiate the needs of various nothing,” even if they privately agree.1 Holdings Inc. and Bernie Madoff. A stakeholder approach is enshrined Though the Stanford Research Institute is credited with first using the modern Everyone counts definition of stakeholder in 1963, the con- To Freeman, the essence of business is cept wasn’t fused with management think- value creation. Maintaining relationships ing until R. Edward Freeman published his with customers, suppliers, employees, influential book, Strategic Management: A financiers and communities encourages Stakeholder Approach, in 1984.2 Freeman leaders to spark conversations about how remains a thought leader in stakeholder their companies create value and manage management. Recently, I interviewed Free- the needs of all stakeholders. Tradeoffs man and we discussed stakeholder theory never work. “Great companies figure out a and its connection to quality. win-win strategy,” he said. Freeman, a professor at the Darden 54 QP • www.qualityprogress.com Freeman’s thinking evolved as quality School of the University of Virginia in gained ground in the United States. He Charlottesville, studied mathematics and sees his work as complementary to the philosophy at Duke University in Dur- quality movement. When I asked Freeman ham, NC, before earning a doctorate in for his opinion on W. Edwards Deming, he philosophy from Washington University called him a hero and a true philosopher. in St. Louis. Postdoctoral work took him “Deming’s principles of driving fear out to the Wharton School of the University of the workplace and supporting human We don’t need more theory. We need more stories where we don’t see tradeoffs. the profound knowledge embedded in stakeholder management and Deming’s approach to quality and the potential for transformational change when putting those principles in action. They may find their passion for service and organizational change in the deeper dimensions of stakeholder management, which is gather- dignity have a moral basis,” he said. While to make red blood cells. We need them, but ing new energy in business thinking and Freeman thinks Deming’s teachings are they are not our purpose. Risk takes our practice. QP “incredibly important,” he finds it interest- eyes off of what we want to do. We have ing that the core principles of his work are to remain engaged, open, committed to continually revisited. “Have we become so purpose and answer the key question: Is messed up that we need to be told what this worth doing?” Freeman hopes stakeholder manage- 5-year-olds know?” he said. Our conversation shifted to risk manage- ment will continue to integrate with other ment and its entry into the quality arena. disciplines. “We don’t need more theory. “Risk isn’t interesting,” he said. “What’s We need more stories where we don’t see more interesting [and important] is how tradeoffs,” he said. Freeman cited the con- organizations engage multiple stakeholders. scious capitalist movement and values- [Risk] enters the equation when share- based vision outlined in Whole Foods co- holder value needs to be maximized to founder John Mackey’s book, Conscious satisfy Wall Street.” Freeman said he thinks Capitalism: Liberating the Heroic Spirit risk and profit hold too much power. “It’s of Business,4 as an example. as if one were to say the purpose of life is Share Your Quality Journey Quality professionals sometimes forget References 1. Roland Marchand, Creating the Corporate Soul, University of California Press, 1998. 2. R. Edward Freeman, Stakeholder Theory: The State of the Art, Cambridge University Press, 1984. 3. Michael E. Porter, Competitive Strategy, Free Press, 1980. 4. John Mackey and Rajendra Sisodia, Conscious Capitalism: Liberating the Heroic Spirit of Business, Harvard Business School Press, 2013. Henry J. Lindborg is executive director and CEO of the National Institute for Quality Improvement in Fond du Lac, WI, which provides consulting in strategic planning, organizational development and assessment. He holds a doctorate from the University of WisconsinMadison and teaches in a leadership and quality graduate program. Lindborg is past chair of ASQ’s Education Division and of the Education and Training Board. He also chairs the IEEE-USA’s Career Workforce Policy Committee. QP occasionally includes an interesting, personal quality story in its “Quality in the First Person” column. If you are interested in sharing your story—how you got into the quality field, how it has helped your organization or your career or how quality has enhanced your personal life—email manuscripts@asq.org. June 2013 • QP 55 Special Advertising Section TECHNOLOGY 2013 Software Showcase & Directory The speed at which technology moves today is blinding. Blink, and you might miss the boat. Stay a step ahead with the 2013 Software Showcase & Directory. This guide brings together the best and most innovative software providers today in the quality field. Inside you’ll find extensive listings featuring: • A description of the providers • Contact information • Areas of focus, from benchmarking to inspection June 2013 • QP 57 Company Contact Information Advanced Systems Consultants x x x x x x x x x x x x x x x x x x x x x x x x OTHE R TECHNOLOGY AUDIT ING BALD RIGE AWA RD BENC HMA RKIN G CALIB RATIO N CAPA BILIT Y STU COM DIES PUTE RS CONS ULTIN G CORR ECTIV E AC TION CUST OME R SER VICE CUST OMIZ ED S OFTW DATA ARE ACQU DEVE ISITIO LOPM DESIG N ENT N OF E X PERIM DOCU ENT ME N T MA NAGE E-CO MME ME N R CE T GAUG E R EP E A TABIL INSP ITY A ECTIO ND R N EP R O ISO 9 DUCIB 000 ILITY ISO 1 4000 MAN AGEM ENT MEA SURE ME N T PR EV ENTIV E AC TION PROB LE M SOLV ING PROC ESS D OCUM QS-9 ENTA 000/T TION S 169 /MAP 49 QUAL PING ITY A SSUR QUAL ANCE ITY F UNCT ION A R ECA LL M ND D ANAG E P LO YMEN RELIA E ME N BILIT T T Y SIMU LATIO N SIX S IGMA SOFT WAR E STAT ISTIC AL M ETHO STAT DS ISTIC AL PR OCES SUPP S CO LIER NTRO QUAL L ITY A TAGU SSUR CHI T ANCE ECHN TEST IQUE S P R EP A Training RATION Special Advertising Section P.O. Box 5257 Scottsdale, AZ 85261 Phone: 480-423-0081 www.mpcps.com ASC provides the right combination of consulting, training and implementation in statistical-engineering improvement methodologies, thus helping our clients improve their productivity, efficiency and profitability. CyberMetrics Corporation x 1523 W. Whispering Wind Drive, Suite 100 Phoenix, AZ 85085 Phone: 800-777-7020 www.cybermetrics.com x x x x x More than 12,000 facilities worldwide choose CyberMetrics software solutions to manage assets, calibrations, preventive maintenance and supplier quality while maintaining standards compliance and reducing costs. DataNet Quality Systems 29200 Northwestern Highway, Suite 350 Southfield, MI 48034 Phone: 866-4WINSPC www.winspc.com x x x x x x x x x x x x x x x x x x x x x x x x x x x x DataNet Quality Systems delivers real-time quality control software that empowers manufacturers to make the highest quality product for the lowest possible cost. EtQ Inc. 399 Conklin Street, Suite 208 Farmingdale, NY 11735 Phone: 516-293-0949 www.etq.com x EtQ’s enterprise quality and compliance management software identifies, mitigates and prevents high-risk events in the quality system. Modules include CAPA, document control, training and more. 58 QP • www.qualityprogress.com x x x x x x x x x x x x x x The World Standard. No Deviation. GAGEtrak has a solid track record spanning over 25 years decades and is used at over 12,000 facilities worldwide. Accurate, reliable and recommended by auditors, GAGEtrak is the world standard for calibration management software. Available in Desktop, Server and Cloud versions, GAGEtrak will help you effectively manage your metrology assets, monitor scheduled and unscheduled calibrations, minimize costs, improve efficiency and assure compliance with quality standards ISO 9001, ISO/TS 16949, ISO/IEC 17025, ISO 13485, AS 9100 and 21 CFR Part 11. Contact us today for your free 30-day trial and you’ll see why GAGEtrak is the #1 calibration management software solution in the world. We’re certain of it. Calibration Management Software by CyberMetrics 1-800-777-7020 www.cybermetrics.com June 2013 • QP 59 Company Contact Information Hertzler Systems Inc. x Phone: 574-533-0571 www.hertzler.com Minitab Inc. Quality Plaza 1829 Pine Hall Road State College, PA 16801-3008 Phone: 800-448-3555 www.minitab.com x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x Minitab® Statistical Software, Quality Companion by Minitab™ and Quality Trainer by Minitab™ provide a complete solution for Six Sigma and quality improvement projects. PQ Systems 210B E. Spring Valley Road Dayton, OH 45458 Phone: 937-885-2255 www.pqsystems.com x x x x x x x x x x x PQ Systems provides quality control software and training to help all industries demonstrate proof of their quality performance. Products include SQCpack®, GAGEpack® and CHARTrunner Lean®. QI Macros SPC Software for Excel x QiSOFT x x x x x x x x x x x x x Phone: 888-468-1537 www.qimacros.com 2 E. Bryan Street, Suite 1515 Savannah, GA 31401 Phone: 912-790-7990 www.qisoft.com Working with manufacturers worldwide to improve profitability by establishing quality as a strategic focus, QiSOFT’s manfacturing intelligence software supports the drive for continuous process and product excellence. 60 QP • www.qualityprogress.com x x x x x x x OTHE R TECHNOLOGY AUDIT ING BALD RIGE AWA RD BENC HMA RKIN G CALIB RATIO N CAPA BILIT Y STU COM DIES PUTE RS CONS ULTIN G CORR ECTIV E AC TION CUST OME R SER VICE CUST OMIZ ED S OFTW DATA ARE ACQU DEVE ISITIO LOPM DESIG N ENT N OF E X PERIM DOCU ENT ME N T MA NAGE E-CO MME ME N R CE T GAUG E R EP E A TABIL INSP ITY A ECTIO ND R N EP R O ISO 9 DUCIB 000 ILITY ISO 1 4000 MAN AGEM ENT MEA SURE ME N T PR EV ENTIV E AC TION PROB LE M SOLV ING PROC ESS D OCUM QS-9 ENTA 000/T TION S 169 /MAP 49 QUAL PING ITY A SSUR QUAL ANCE ITY F UNCT ION A R ECA LL M ND D ANAG E P LO YMEN RELIA E ME N BILIT T T Y SIMU LATIO N SIX S IGMA SOFT WAR E STAT ISTIC AL M ETHO STAT DS ISTIC AL PR OCES SUPP S CO LIER NTRO QUAL L ITY A TAGU SSUR CHI T ANCE ECHN TEST IQUE S P R EP A Training RATION Special Advertising Section Give everyone in your organization the power to improve it. Minitab 16 Statistical Software www.minitab16.com QualityProgress_HalfPage_03-25-13.indd 1 3/25/2013 1:46:44 PM June 2013 • QP 61 Company Contact Information RealityCharting® OTHE R TECHNOLOGY AUDIT ING BALD RIGE AWA RD BENC HMA RKIN G CALIB RATIO N CAPA BILIT Y STU COM DIES PUTE RS CONS ULTIN G CORR ECTIV E AC TION CUST OME R SER VICE CUST OMIZ ED S OFTW DATA ARE ACQU DEVE ISITIO LOPM DESIG N ENT N OF E X PERIM DOCU ENT ME N T MA NAGE E-CO MME ME N R CE T GAUG E R EP E A TABIL INSP ITY A ECTIO ND R N EP R O ISO 9 DUCIB 000 ILITY ISO 1 4000 MAN AGEM ENT MEA SURE ME N T PR EV ENTIV E AC TION PROB LE M SOLV ING PROC ESS D OCUM QS-9 ENTA 000/T TION S 169 /MAP 49 QUAL PING ITY A SSUR QUAL ANCE ITY F UNCT ION A R ECA LL M ND D ANAG E P LO YMEN RELIA E ME N BILIT T T Y SIMU LATIO N SIX S IGMA SOFT WAR E STAT ISTIC AL M ETHO STAT DS ISTIC AL PR OCES SUPP S CO LIER NTRO QUAL L ITY A TAGU SSUR CHI T ANCE ECHN TEST IQUE S P R EP A Training RATION Special Advertising Section x 731 Commerce Street, Floor 2 Tacoma, WA 98402 Phone: 877-722-2770 www.realitycharting.com Apollo Root Cause AnalysisTM methodology and RealityCharting® software provide a better understanding of problems and identifying effective solutions used by organizations around the world! SigmaXL x 110 Cumberland Street, Suite 356 Toronto, ON M5R 3V5 Canada Phone: 416-236-5877 www.sigmaxl.com x x x SigmaXL + Excel = easy statistical and graphical analysis. SigmaXL and DiscoverSim are cost-effective, powerful, user-friendly and ideal for Lean Six Sigma training! Free 30-day trial. StatPoint Technologies Inc. P.O. Box 208 Orlean, VA 20128 Phone: 540-229-0090 www.statgraphics.com x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x Statgraphics Centurion XVI—make sense of it all. Leading companies worldwide trust Statgraphics statistical software to give voice to their data. Make informed decisions that drive operational efficiencies, ensure quality and preserve precious resources. StatSoft 2300 E. 14th Street Tulsa, OK 74104 Phone: 918-749-1119 www.statsoft.com x StatSoft provides award-winning, industry-specific enterprise solutions that integrate analytics with your data repositories for process monitoring, root cause analysis, quality improvement and sophisticated graphical reporting. 62 QP • www.qualityprogress.com x x Apollo Root Cause Analysis™ By Dean L. Gano Free Book Free Software Solutions Go to realitycharting.com/rcbook Go to realitycharting.com/downloads for free software and 30-day trial provideds a text entry interface that simply asks the questions a facilitator would ask. As answers are provided a Realitychart is created. There is no training needed to use this application and it is free to everyone. Using RealityCharting® increases the quality of your analysis by providing expert guidance and structure supporting the Apollo Root Caues Analysis method. Enter the code below to get your copy CODE: RCASQ13 Version 6 includes RC Coach™ A self-paced online learning experience designed to teach how RealityCharting® can be used to understand and solve your most pressing problems. www.realitycharting.com ◆ (US) 206-922-7910 ◆ (US) 877-722-2770 ◆ info@realitycharting.com Good Ad.new brown2_Layout 1 9/10/12 3:44 PM Page 1 Achieve efficiency that puts time on your side. Statgraphics® has long provided leading companies and governments worldwide with the software they need for Lean Six Sigma, quality improvement and statistical solutions of every stripe. We are committed to making your job easier and your projects successful. Take control of your business. Make sense of it all with the respected statistical data analysis solutions from Statgraphics. I I Statgraphics Centurion Sigma express, a DMAIC for Microsoft Excel Statbeans Java Library I Statgraphics .Net Services Statgraphics Online for ipad and beyond… I Training Consulting Learn more, find a local representative or download a free copy to evaluate and begin improving your business at: www.statpoint.com 2012 Statpoint Technologies, Inc. All rights reserved. The logo’s and service names mentioned herein are registered trademarks or trademarks of Statpoint Technologies. Copyright 2012 Statpoint Technologies, Inc. training. consulting. web services. six sigma anova regression spc doe Statistical Analysis Software forecasting visualization reliability June 2013 • QP 63 Company Contact Information Waypoint Global (formerly Powerway) x x x x x x x x x x 6910 N. Shadeland Avenue Indianapolis, IN 46220 Phone: 317-624-4000 www.waypointglobal.com Waypoint Global Suite—with more than 1 million users worldwide—is a proven software solution for QMS, with 16 software modules driven by the Suite Desktop command center. Get complete visibility of potential quality and process issues before they become a problem… MANUFACTURING INTELLIGENCE SOFTWARE Pre-emptive Quality and Process Control See potential problems before they cost you time and money Automatic Data Capture Paperless shop--oor, frees up time and improves data integrity Real-time SPC Analytics Reduce manufacturing waste, rework and raw material usage OEE Performance Monitoring Monitor and nd ways to correct the ‘Six Big Losses’ Instant Capability Studies Review product and process capability in seconds Supplier quality monitoring Prevent substandard materials entering your process 64 QP • www.qualityprogress.com x OTHE R TECHNOLOGY AUDIT ING BALD RIGE AWA RD BENC HMA RKIN G CALIB RATIO N CAPA BILIT Y STU COM DIES PUTE RS CONS ULTIN G CORR ECTIV E AC TION CUST OME R SER VICE CUST OMIZ ED S OFTW DATA ARE ACQU DEVE ISITIO LOPM DESIG N ENT N OF E X PERIM DOCU ENT ME N T MA NAGE E-CO MME ME N R CE T GAUG E R EP E A TABIL INSP ITY A ECTIO ND R N EP R O ISO 9 DUCIB 000 ILITY ISO 1 4000 MAN AGEM ENT MEA SURE ME N T PR EV ENTIV E AC TION PROB LE M SOLV ING PROC ESS D OCUM QS-9 ENTA 000/T TION S 169 /MAP 49 QUAL PING ITY A SSUR QUAL ANCE ITY F UNCT ION A R ECA LL M ND D ANAG E P LO YMEN RELIA E ME N BILIT T T Y SIMU LATIO N SIX S IGMA SOFT WAR E STAT ISTIC AL M ETHO STAT DS ISTIC AL PR OCES SUPP S CO LIER NTRO QUAL L ITY A TAGU SSUR CHI T ANCE ECHN TEST IQUE S P R EP A Training RATION Special Advertising Section The ASQ Global State of Quality R e s e ar c h Discoveries 2013 Now Available! The ASQ Global State of Quality Research is an unprecedented research project pioneered by ASQ. It evaluates how the quality discipline is being practiced and performed around the world. The ASQ Global State of Quality Research: Discoveries 2013 is the first in a series of three reports, and is now available for download. Discoveries 2013 provides groundbreaking data on quality operations and their impact on organizations worldwide. Gathered from 2,000 organizations in more than 22 countries, it uncovers trends throughout the world, providing benchmark data to help you compare your organization to the current state of quality. You’ll also gain a solid understanding of local, regional, and global landscapes, allowing you to pinpoint new growth opportunities. Download the report today at www.globalstateofquality.org. training CERTIFICATION CONFERENCEs MEMBERSHIP PUBLICATIONS QPToolbox Portable analyzer Spectro Analytical Instruments’ Spectroscout portable energy dispersive x-ray fluorescence (ED-XRF) analyzer enables rapid, laboratory-class elemental analysis of environmental and geological samples in remote locations. The portable analyzer is ideal for on-site environmental and geological analysis. In shale gas boundary layer detection, for instance, it allows for data-based direction changes that increases drilling productivity. The Spectroscout portable ED-XRF analyzer is encased in rugged housing with its X-ray source protected. Its features include a large sample compartment, x-ray tube, onboard processor, and high-yield battery pack. An optional integrated video system allows spot testing and image storage. It provides precise measurements from trace to minor and major concentra- • Email: spectro.info@ametek.com. Gage calibration software tions across a range of relevant elements • Visit: www.ametek.com. GAGEpack from PQ Systems is gage cali- from heavy elements, such as uranium, bration software that maintains complete to light elements, such as sodium. The Extension tables histories of measurement devices, instru- touchscreen interface provides access to R&R Fixture Xtension tables are designed ments and gages. The software features analysis results and measured spectra. for the mounting of portable coordinate tools such as calibration schedules and measuring machines or laser scanning reports, alerts about failed and past-due machines. The tables feature rigid alumi- calibrations, gage location and status, num extruded structures that can hold tracking repair records, audit trail for a lot of weight for any size, material or traceability and a task tab with a “to-do” shape of part you can inspect. list. It also features an optional email alert The tables are available with ¼-20 or M8 threaded plate and components for 66 QP • www.qualityprogress.com system that sends reminders. The “to-do” list indicates items requir- locating and elevating parts off the surface ing attention for a user-selected period of for full inspection of the part, whether time. It includes the ability to estimate due using touch probe or laser. R&R offers a dates for gages based on the number of variety of components to be used with the uses rather than time, and then it prints Xtension Tables. worksheets directly from the list. The list • Call: 616-847-6045. can be filtered by action and ordered with • Visit: www.rrfixtures.com. a simple click of column headers, and also can be sent by email attachment to designated receivers. If a temperature humidity USB monitor is connected, GAGEpack will record the temperature and humidity for a new calibration or verification event. Acceptable temperature and humidity ranges during calibration also can be defined in the software. • Call: 800-777-3020. • Visit: www.pqsystems.com. Vibration monitor meter detect a problem before a machine is concealed by the encoder end plate. The VibChecker vibration monitor meter damaged. • Call: 847-519-4702. from Quality Bearings and Components • Call: 888-276-4787. • Visit: www.heidenhain.us. is a compact-sized instrument designed • Visit: www.qbcbearings.com. Flexible zone control for quick vibration surveys. The meter provides portable vibration measure- Rotary encoders Intelligrated has announced the ZoneFlex, ment in the 10-1,000 Hz frequency Heidenhain has released the ECN/EQN a flexible zone control for the company’s range. Measurement results are au- 400 series rotary encoders with Profibus Accuglide and IntelliQ zero-pressure tomatically evaluated against Interna- and Profinet interfaces for use in manu- accumulation conveyors. The system tional Organization for Standardization facturing and process automation. The makes the best use of product flow of standards. Results can be stored for encoders feature a rugged blind hollow varied package sizes and weights for documentation and follow-up. shaft with a 12mm diameter. high-density accumulation and increased Models ECN 413 and EQN 425 rotary encoders are resistant to shock up to ZoneFlex modules improve material 1,000 m/s2 and vibration up to 100 m/s2, handling efficiency and allow for carton in accordance with the German Insti- control, product handling and increased tute for Standardization (DIN). Providing conveyable cartons. Clip-in mounting, protection to IP 64, they are designed for switch configuration and LED diagnostics an ambient temperature range from –40° allow operators to replace, modify and C to +70° C in accordance with DIN with monitor zone controllers. shaft speeds up to 6,000 rpm. In regards to connectivity, users of VibChecker is used to trend vibration throughput. ZoneFlex is designed to move previously accumulated product downstream these encoders can choose between by monitoring and automatically respond- levels on fans, motors, pumps, compres- three radial M16 cable glands or three ing to accumulation conditions. sors and other types of rotating process radial M12 connectors. Node addresses • Call: 513-881-5239. machinery. Vibration measurement can are set through rotary decimal switches • Visit: www.intelligrated.com. Got a quality product? Send your product description and photo to vellifson@asq.org. June 2013 • QP 67 QPReviews Enterprise Analytics: Optimize Performance, Process and Decisions Through Big Data pathway to future management thinking. bringing forward a summary and recom- All leaders and corporate officers should mendations. As a resource, the book is read this book to understand the future well-organized and offers effective sec- Thomas H. Davenport, editor, FT Press, and develop their action plans accord- tion headings to guide the reader. There 2012, 288 pp., $44.99 (book). ingly. is also a useful author and subject index. Davenport’s career Bill Baker as a thought leader Speed to Excellence and business process Santa Fe, NM analyst allows him to The STEM agenda is considered a continuing initiative and not a trial curriculum. This content can be used by those in the K-12 school programs as well as academic programs in colleges and universities. ing trends in big data Advancing the STEM Agenda: Quality Improvement Supports STEM and integrating those Cindy P. Veenstra, Fernando F. Padró and explorations related to teaching learning with a future. With Julie A. Furst-Bowe, edittors, ASQ Quality programs. corporate backing, Press, 2012, 304 pp., $38 member, $63 list continue pushing the envelope by identify- he formed the International Institute for The academic areas of interest relate to the subject fields and the pedagogical Jerry Brong (book). Ellensburg, WA Analytics where extensive research is per- This book presents formed to understand big data, its impact insights, operational and current position in the industry. suggestions for pro- Standard Work for Lean Healthcare grams and guidance Thomas L. Jackson, editor, Productivity sisting of 18 chapters. The book starts with The book is divided into five parts con- in moving a STEM Press, 2011, 124 pp., $29.95 (book). an overview of big data and proceeds with (science, technol- Standard work is one applications, technologies, the human side ogy, engineering, of the fundamental and case studies. Big data are character- and mathematics) concepts in lean, and ized as “data that are either too volumi- curriculum forward. this book explains nous or too unstructured to be managed The book is a collection of selected papers how it can be applied and analyzed through traditional means.” from the July 2011 ASQ Education Divi- and implemented in This is a challenge and an opportunity sion conference on Advancing the STEM a healthcare setting. to get a complete view of your custom- Agenda in Education, the Workplace and The book has seven ers by linking and projecting needs from Society. Papers selected from the confer- computer clicks. ence present perspectives about STEM, the definition of standard work and its ap- strategies for implementation and sug- plications in healthcare. In part four, the discussion turns to how big data analysts could be structured in gestions for bridging the gap between the organizations. The challenge is to leverage workplace and educational programs. their knowledge and generate value, defin- Value from this book can be derived chapters that cover The book is readable with margin assists that delineate sections with examples, key points and how-to steps. ing roles and responsibilities and gover- from the reports by the authors of Given that this book’s intended reader nance in different-sized organizations. Part individual papers on their strategies and is someone learning lean methods, the five presents some insightful case studies tactics working with a STEM agenda, margin assists are good aids. and the organizations’ journeys. overcoming challenges, assessing learner This book outlines the terminology and In addition, the how-to sections of the and program outcomes, and preparing for book go into great detail on how to fill importance of dealing with big data and follow-steps. The editors of this collec- the charts, such as the standard work its impact on being competitive. It is the tion of papers offer a concluding chapter sheet, percentage load chart, standard 68 QP • www.qualityprogress.com work combination sheet and the stan- subject, but it does deliver on the purpose ment his baby-step approach to dealing dard work instruction sheet. and value of the internal audit process. I with typical managerial challenges. These Overall, this book is worthwhile for would recommend this book to all person- include boosting morale, cutting costs, anyone who is interested in implementing nel either managing or performing audits. improving quality and service, develop- standard work in a healthcare setting. And unlike the first edition, you can stop ing new products, increasing sales and searching. reducing healthcare expenses. He further Brian Cocolicchio New City, NY Wayne Sander Dove Quality Consulting How to Audit the Process-Based QMS Dousman, WI connects with managers by frequently relating the discussion to actual business situations or past clients. Maurer’s relatively short work emphasizes the key points that are nicely 208 pp., $44 member, $72 (book, second The Spirit of Kaizen: Creating Lasting Excellence One Small Step at a Time edition). Robert Maurer, McGraw-Hill Professional, anyone new to kaizen, as well as internal 2012, 176 pp., $22 (book). kaizen leaders in any organization. Dennis R. Arter, Charles A. Cianfrani and John E. (Jack) West, ASQ Quality Press, 2012, I had the pleasure of presented in language that any reader will find helpful. This book is recommended for reviewing the first Maurer presents a Dale Farris edition of this book. relatively brief—but Groves, TX At that time, I be- powerful—expla- lieved it was a good nation of how the source of information weighty principles for those who need- of kaizen can help ed to understand the ensure significant basics of auditing and and lasting change how to set up an audit program at their in an organization. organization. But for quality practitioners, I Following a succinct introduction to suggested they continue searching for the kaizen, Maurer explains how to imple- Recent Releases Permission to Forget, 10th Anniversary Edition: And Nine Other Root Causes of America’s Frustration with Education Lee Jenkins, ASQ Quality Press, 2013, 210 pp., $30 member, $51 list (book). process to audit the process approach. As with a lot of things, time helps us mature and continuously improve. This is true of the book’s second edition. The authors have hit the mark with this book. They are direct and frank in their approach to auditing. Auditing is about understanding requirements, looking for opportunities for improvement and best practices, asking questions, gathering information, analyzing what is seen and heard, forming opinions, and reaching conclusions. And believe it or not, they do a good job of explaining how to do it all. The book is not lengthy and doesn’t go into a great deal of explanation on every Advertisers Index Advertiser PagePhone Web Advanced Systems Consultants 59 480-423-0081 www.mpcps.com CyberMetrics Corporation 59 800-777-7020 www.cybermetrics.com EtQ Inc. 1 516-293-0949 www.etq.com Minitab Inc. 61 800-448-3555 www.minitab.com PQ Systems Inc. 61 800-777-3020 www.pqsystems.com QiSoft 64 912-790-7990www.qisoft.com Quality Council of Indiana 10, 11 800-660-4215 www.qualitycouncil.com RealityCharting 63 877-722-2770www.realitycharting.com SigmaXL 56 888-744-6295www.sigmaxl.com StatPoint Technologies Inc. 63 540-229-0090 www.statgraphics.com StatSoft Inc. OBC 918-749-1119 www.statsoft.com Villanova University Online IFC 800-341-7106 www.villanovau.com/efficiency June 2013 • QP 69 For information on placing an ad, contact Media Sales at 866-277-5666. ProfessionalServices Lean Six Sigma Training and certification online by Thomas Pyzdek Save 10%! Enter coupon code ASQ10 at checkout www.sixsigmatraining.org /store +1 520-204-1957 Quality Institute of America, Inc. Since 1994 Complete Quality, Environmental, Safety, Six Sigma, Consulting, Auditing, and Training Services. Comprehensive Quality Management Software (QISS). • QISS-based ISO-Easy Program • Quality Management Software (QISS) Automating Document and Records Control, Communications, Nonconformance, Corrective-Preventive Actions, Calibration, Maintenance, Training, Audits, and Management Reviews. PH: 281-335-7979. Houston, TX Email: Sales@qisssoftware.com, www.qi-a.com www.qisssoftware.com Quality Systems Registrars, Inc. Providing Registration Services to: AS 9100, ISO 9000, ISO 14000, ISO 18000 and RCMS/RC 14000 QSR will be moving and would like to announce the new office address: 22375 Broderick Drive, Suite 160 Sterling, VA 20166 Phone: 703-478-0241 • Fax: 703-478-0645 www.qsr.com IMDS Data • We can create your data and submit within 72 hours • We do the work and invoice you • References found on our website • Based in the United States www.imdsdata.org • 734-205-8874 TQM Associates Inc. “we’re here to support you” • Established in 1994 • Women-Owned • Quality Assurance Professionals • Across the U.S. and Worldwide • Temporary or Permanent • Source Inspection • Surveys • Audits • Expediting Visit us at: www.qualitytng.com Email sales@qualitytng.com for brochure 800-424-4729 3990 Old Town Ave. #C109 Fax 619-297-3251San Diego, CA 92110 tqmassociates.com email: stephk@tqminc.net Ph: 248-641-7030 Fax: 248-641-7031 PO Box 611 Troy, MI 48099-0611 70 QP • www.qualityprogress.com Classroom Training Aids… g Quincunx Boards g Sampling Bowls g Catapults g Deming Funnels g And lots more … ProfessionalServices For Accredited Certification Look for the Symbols of Quality EAGLE Registrations Inc. EAGLE Food Registrations Inc. Highest Ranked Registrar in an independent customer survey! ISO 9001 · AS 9100 · ISO/TS 16949 ISO 14001 · ISO 13485 · OHSAS 18001 Safe Quality Food (SQF) · ISO 22000 FSSC 22000 · SQF Ethical Sourcing Call 800-795-3641 | www.eagleregistrations.com Consultants in Quality Inc. Experienced consulting for accreditation in: •ISO/IEC 17065 Conformity assessment for products, process and service •ISO/IEC 17025 Competency of test labs Contact: Richard Stump • stumpRB@aol.com For information on placing an ad, contact Media Sales at 866-277-5666. Quality Systems Enhancement, Inc. Auditing/Consulting/Training 9 9 9 9 9 ISO 9001, AS 9100, ISO 13485, ISO 20000 FDA QSR, Canadian MDR, European MDD Planning, Implementation, Training Process Improvements, Process Validation QMS Internal Audits, Mini-audits Camille Delmotte, MBA, President Phone and fax: 410-426-2269 info@qualityedgeconsulting.com www.QualityEdgeConsulting.com Standards • ISO Standards: 9001, 14001, 50001, 13485/21CFR 820, 15189, 17025, 27001, 31000, AS 9100/9110/9120, OHSAS 18001, TL 9000, TS 16949, etc. • Food Safety Standards: BRC, SQF, FSSC 22000, ISO22000, FDA Compliance • Chain of Custody Standards: FSC, SFI, PEFC • Recycling Standards: e-Stewards Certification, R2 Certification • Gap Analysis to All Standards Excellence Tools •Design of Experiments •Design for Six Sigma •Lean Enterprise •Lean Manufacturing •Lean Six Sigma •Sustainability •Bottom Line Improvement (Guaranteed) Other Tools/Services •Statistical Problem Solving •Project Management •Internal Auditing Services •Risk Management Industries Aerospace, Automotive, Casting, Energy, Healthcare, Environmental, Forest Products, Manufacturing, Packaging, Medical devices, Materials, Services, Telecommunications, Textile, etc. Quality Systems Enhancement Inc. 1790 Woodstock Rd. Roswell, GA 30075 www.enhancequality.com• baskar@enhancequality.com • 770-518-9967 Beijing Enertic Engineering Technical Co., Ltd. A 3rd party inspection company 10 years’ experience for boiler, steel structure, pipes, pumps, valves, etc. in China and Asia. China Headquarters: Tel: +86-1350-023-1596 Email: xqwang@enertic.com US representative: Tel: 843-810-4805 Email: james.zhang@linxyn.com Website: www.enertic.com The source for flash card study aids for certification exams: eAudits.org CQE, CSSBB, STATS, and more. Learn how to conduct audits remotely. www.qualityreviewinaflash.com Looking for Quality Professionals? Place a Recruitment Ad in Quality Progress and reach more than 100,000 readers from all over the world! Contact Media Sales at 866-277-5666. Experience ASQ TV on videos.asq.org TM June 2013 • QP 71 Back to Basics BY scott force Light Bulb Moment DMAIC powers new perspective Newly certified Six Sigma Black weren’t painted that could not be seen Leading the way Belts often struggle with understand- while standing upright. Lying on the floor As shown in Figure 1, the process map ing how the various quality tools work provided me with a new perspective. leads to the use of other tools that help together. It took me a few projects before Several months later while on my lad- break the process down. For example, I began to see linkage. As I teach and men- der replacing a light bulb, I found more failure mode and effects analysis can help tor new Belts, I use an analogy that hits areas that were missing paint. If it wasn’t uncover process steps that are error prone close to home. for the new point of view, I never would and pose greater risk to the customer. Key have noticed these minor defects. In lean inputs to the process map can be analyzed house. Before moving day, we conducted Six Sigma (LSS) projects, the define, using measurement systems analysis a final walkthrough with the contractor. I measure, analyze, improve and control (MSA) to determine whether there are remember pointing out areas on the walls (DMAIC) approach helps a team view a issues with data quality. that needed a touch up with paint, the process or problem from many different spackling smoothed out and other minor perspectives. In 2001, my family and I moved to a new cosmetic improvements. It wasn’t until After MSA, the team can conduct a process capability analysis to determine how DMAIC is a disciplined and systematic well the key process variables are meeting months later, while wrestling with our kids approach to drive process improvement. customer requirements. A cause and effect on the living room floor, that something While DMAIC can be used on its own, it matrix narrows the key inputs listed on caught my eye. I noticed little areas that also helps guide LSS projects. The method the process map and separates the trivial Tool linkage / figure 1 forces the team to many from the vital few. Process control view a process from plans document the controls in place in many different angles the current state and pinpoint where criti- and to turn to a vari- cal process controls are needed. ety of tools to solve FMEA/risk analysis Process control plan VA/NVA analysis Process map It is important for LSS practitioners to a problem. Issues, consider the DMAIC roadmap and to max- concerns, root causes imize quality tools along the improvement and solutions are un- journey. By adhering to this disciplined covered more quickly approach, there is a higher probability that and easily. the team will reach its goals and achieve Belts are taught to continuous improvement. Like a doctor di- always map the pro- agnosing an ailment or a carpenter build- cess being improved ing a home, you must use the tools at your in the early stages disposal to uncover variables preventing of the project. After the process from running at its best. QP the process map is Cause and effect matrix Measurement systems analysis complete, the Belt uses tools in his or her tool kit to evaluate the Process capability analysis VA = value added NVA = nonvalue added FMEA = failure mode and effects analysis 72 QP • www.qualityprogress.com process from various perspectives and determine what activities are value added and nonvalue added. Scott Force has more than 22 years of quality improvement experience in the healthcare, automotive and power equipment industries. He earned a bachelor’s degree in manufacturing engineering from Miami University in Oxford, OH. A senior member of ASQ, Force is an ASQ-certified quality engineer and a Sigma Breakthrough Technologies-trained Master Black Belt. Look at what’s new from ASQ Quality Press! Insights to Performance Excellence 2013–2014: Understanding the Integrated Management System and the Baldrige Criteria Statistics for Six Sigma Made Easy! This book helps leaders, performance excellence examiners, and organization-improvement practitioners to understand the 2013–2014 Baldrige Criteria and the linkages and relationships among the items. Item: P1487 This book helps make Six Sigma simple to understand and implement. Item: H1440 Statistical Process Control for the FDARegulated Industry The Lean Turnaround: How Business Leaders Use Lean Principles to Create Value and Transform Their Company The focus of this book is to understand and apply the different SPC tools in a company regulated by the Food and Drug Administration (FDA). Legendary business leader Art Byrne shares everything he has learned during his career implementing lean strategies at 30 companies in 14 different cultures. Item: H1445 Item: P1490 Learn more about these books through the Quality Press bookstore at asq.org/quality-press. training CERTIFICATION CONFERENCEs MEMBERSHIP PUBLICATIONS STATISTICA is the only process monitoring solution that leverages Big Data Predictive Analytics. Supports basic control charting, dashboarding, and predictive maintenance solutions for thousands of parameters. Quality Control, Process Monitoring, Multivariate Trending, Annual Product Reviews, Complaints Trending, Stability & Shelf Life Analysis, Validated Data Entry, and more. Open architecture readily supports your existing data warehouses, IMS, ERP, and MES systems. Dedicated solutions for validated deployment in medical device and pharmaceutical manufacturing. When quality matters, STATISTICA rules. www.StatSoft.com/v12 • 2300 East 14 Street • Tulsa, OK 74104 • USA • +1 (918) 749-1119 • © Copyright StatSoft, Inc. 1984-2013. StatSoft, and STATISTICA are trademarks of StatSoft, Inc. • QP201304r02