OC TOBE R 2018 | QUALIT YPROG RE SS .

COM

A Deeper How Machine Deming’s

Dive Into Learning Will Red Bead
Disruptive Redefine Experiment
Technologies Quality vs. Scorecards
page 24 page 30 page 48
|OCTOBER 2018

The official publication of ASQ

QUALITY 4.0

BUYING INTO
BLOCKCHAINn
How the revolutionary technology will change
supply chains—and the quality profession
VOLUME 51 | NUMBER 10

The Global Voice of Quality

 UPCOMING
CERTIFICATION
EXAMS
• Calibration Technician

• Pharmaceutical GMP
Professional

• Quality Auditor

• Quality Engineer

• Quality Improvement
Associate

• Quality Process Analyst

• Six Sigma Green Belt

• Software Quality
Engineer

GET YOUR
ASQ CERTIFICATION
Looking for a way to stand out in 2018? ASQ certification can help with
that. Choose from 18 certifications that are sure to not only differentiate you
from your competition, but can also result in a higher salary or help land a
promotion. It’s time to become a recognized expert in your field.
Apply for the December 2018 exams today!

2018 UPCOMING APPLICATION DEADLINE – November 2

TESTING WINDOW – December 1 ­– 17

For more information or to apply for an exam, visit asq.org/cert.

 CHECK OUT THESE GREAT
ITEMS AVAILABLE FROM
ASQ QUALITY PRESS!
THIS MONTH’S PICKS!
Adaptive Space: The Toyota Kata
How GM and Other Practice Guide
Companies Are Positively In this long-awaited
companion to the
Disrupting Themselves groundbreaking book
and Transforming into Toyota Kata, Mike Rother
Agile Organizations takes you to the next level
of developing a business
To edge out the competition
mindset and capabilities
in today’s disruptive environment,
for the 21st century. The
you need to ensure that your
Toyota Kata Practice Guide
company is agile. This book
walks you through the process of making improvement,
provides a model that will inspire
adaptation, and even innovation routine behavior.
employees to explore new ideas, empower the most creative
people and teams to spread their ideas across the organization, Item: P1643
and operationalize the entrepreneurial spirit so adaptability
is a constant.
Item: P1642

Learn more about these books through the

Quality Press bookstore at asq.org/quality-press.
FEATURES
18 The Time Is Now 30 Revolutionary Impact?
The basics of blockchain technology and why quality Machine learning is already transforming quality
professionals should pay attention to this innovation— improvement activities, and more is in store for
one that’s sure to optimize supply chains and more. quality professionals.
by Narahari Rao by Johannes Freiesleben

24 Let’s Get Digital 21st Century Excellence

37
How the fourth industrial revolution will reshape the Updating and adapting W. Edwards Deming’s system
way we think about quality. of profound knowledge for today’s service systems.

INSIDE
by Nicole Radziwill by Bob Kennedy

October 2018
VOLUME 51 ● NUMBER 10

EXCLUSIVES
at qualityprogress.com
Added Insight
Additional figures to illustrate a
comparison between Deming’s red
bead experiment to red-yellow-
green scorecards, the topic of
this month’s Six Sigma Solutions
column, “The Improvement of
Scorecard Management” (pp.
48-52).

One More to Go
An additional table that
accompanies this month’s Back
to Basics column, “The Right
Questions,” (p. 64).

Pick and Choose

Review a selection of QP articles
recommended by editors.

Back to Basics
Volviendo a los fundamentos en
español.

qualityprogress.com ❘ October 2018 QP 3

INSIDE
October 2018

8 10 45 64

6 Seen & Heard 14 Career Coach 58 Marketplace

Taking a chance on a pizza parlor.

8 Expert Answers 60 Footnotes

The differences between problem 45 Statistics Spotlight
solving and root cause analysis. Care is needed when extrapolating
patterns with straight lines. SPECIAL SECTION

56 STANDARDS AND AUDITING

10 Progress Report GUIDE
The ratings race. 48 Six Sigma Solutions
PLUS Deming's red bead experiment vs.
Getting to Know … red-yellow-green scorecards.
John Karlin

53 Standard Issues
12 Mr. Pareto Head Making sense of different
interpretations of ISO 9001:2015.

53

Graphic Designers Article Submission Process express permission of Quality

Cori Haagensen, Sandy Wyss Quality Progress is a peer- Progress. For permission, write
reviewed publication with 85% Quality Progress, PO Box 3005,
Advertising Production of its feature articles written by Milwaukee, WI 53201-3005, call
Barbara Mitrovic quality professionals. To learn 414-272-8575 x7406, fax 414-272-
more about article submission and 1734 or email reprints@asq.org .
Editor in Chief and Publisher Digital Production Specialists the manuscript review process, Photocopies and Reprints
Seiche Sanders Julie Stroik, Julie Wagner visit www.qualityprogress.com Article photocopies are available
@ASQ_Seiche under “Tools and Resources” and
Media Sales Naylor LLC from ASQ at 800-248-1946. To
866-277-5666 click on Author Guidelines. purchase bulk reprints (more than
Associate Editor Mark Edmund
@ASQ_Mark Editorial Offices 414-272-8575 Photocopying Authorization 100), contact Barbara Mitrovic at
Authorization to photocopy items ASQ, 800-248-1946.
Assistant Editor Lindsay Mail for internal or personal use or the
Dal Porto @ASQ_Lindsay Quality Progress/ASQ List Rentals
internal or personal use of specific Orders for ASQ’s member and
600 N. Plankinton Ave. clients is granted by Quality
Manuscript Coordinator nonmember buyer lists can be
Milwaukee, WI 53203 Progress provided the fee of $1
Valerie Ellifson purchased by contacting Michael
Telephone Fax per copy is paid to ASQ or the Costantino at the Infogroup/Edith
Copy Editor Amanda Hankel 800-248-1946 414-272-1734 Copyright Clearance Center, 222 Roman List Management Co.,
Email editor@asq.org Rosewood Drive, Danvers, MA 402-836-6626 or fax
Art Director Mary Uttech
01923, 978-750-8400. Copying 845-620-1885.
Creative Services Manager for other purposes requires the
Cathy Milquet
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 $17 per copy; nonmembers $25 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. © 2018 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 ASQ, PO Box 3005, Milwaukee, WI 53201-3005. Printed in USA.

4 QP October 2018 ❘ qualityprogress.com

FOREWORD
The quality inside Quality Progress

Quality 4.0 and You

What the revolution means for quality

I’m no stranger to the subject of Bitcoin. My significant other dramatically alter quality, how business gets done and the global
has a minor obsession with compulsively monitoring the erratic workforce. Yet the tangible links to our organizations and jobs
swings within his Bitcoin coffers and talking about the topic remains nebulous. That’s because the full deployment of many
with anyone who will listen. Admittedly, I’ve often tuned out of these technologies is in process or still on the horizon. “Let’s
as he went on and on about the novelty of completing nearly Get Digital,” p. 24, details the major technologies that will influ-
instantaneous monetary transactions, in any currency, via your ence the profession in years to come: cloud computing, big data,
smartphone without the involvement of banks or governments, virtual reality, augmented reality, blockchain, additive manufac-
but when I came across the term used in the context of quality, turing, artificial intelligence, machine learning, Internet Protocol
my ears perked up. Version 6, cyber-physical systems and the internet of things.
Blockchain, the technology that enables Bitcoin transac- There is much to know, and much to do to prepare.
tions, will inevitably revolutionize supply chains of the future, We hope this month’s issue helps bring these new and emerg-
as detailed in this month’s cover story, “The Time Is Now,” p. 18. ing technologies into sharper focus as they
Read about how blockchains work, and how the technology will relate to quality and your future.
solidify the integrity of tomorrow’s supply chains. As the article’s
author maintains, it’s not a matter of “if,” it’s “when.”
ASQ defines the overarching concept of Quality 4.0 as the
“future of quality within the context of the exponential growth
of technological advancement and the unprecedented rate of
change that those advancements are causing.” Seiche Sanders
There’s no arguing that this concept—Quality 4.0—will Editor in Chief and Publisher

TCC Daniella A. Picciotti, QMS Alliance Brady Boggs, Randy Brull, Larry Haugh,
Secretary William J. Troy, ASQ Jim Jaquess, R. Dan Reid, Richard Stump
Directors Technical Reviewers
Donald Brecken Ferris State University Naveen Agarwal, Ashraf Ali, Suresh Anaganti,
Heather Crawford Apollo Endosurgery M. Onur Artan, Andy Barnett, Matthew
Jim Creiman Northrop Grumman Corp. Barsalou, David Bonyuet, David Burger,
Ha C. Dao Emerson Climate Technologies Brooks Carder, Bernie Carpenter, L.N. Prabhu
ASQ’s Vision James Kittredge Adaptimmune US Chandrasekaran, Ken Cogan, Linda Cubalchini-
By making quality a global priority, an Travis, Ahmad Elshennawy, Mark Gavoor,
Scott Moeller GI Supply
organizational imperative and a personal ethic, Kunita Gear, Daniel Gold, T. Gourishankar,
Raul Molteni Molteni Consulting
the American Society for Quality becomes the Roberto Guzman, Ellen Hardy, Lynne Hare,
community for everyone who seeks quality Luis G. Morales Verizon Telematics Inc.
Victoria Jones, Trevor Jordan, Ray Klotz, T.M.
technology, concepts or tools to improve Mark Moyer CAMLS Kubiak, William LaFollette, Scott Laman, Pradip
themselves and their world. Barrie Simpson Genentech Access Solutions Mehta, N.S. Narahari, Arind Parthasarathy,
JoAnn Sternke Pewaukee School District Larry Picciano, Gene Placzkowski, Tony Polito,
ASQ Administration John Vandenbemden Q-Met-Tech Peter Pylipow, Imran Ahmad Rana, Narahari
CEO William J. Troy Allen Wong Abbott Vittal Rao, John Richards, James Rooney,
Senior Leadership Ayman Sakr, Manboubeh Samghabadi, Brian
Administration Scullin, Abhijit Sengupta, Amitava Sengupta,
Ann Jordan Brian Savoie
William J. Troy, CEO Mohit Sharma, A.V. Srinivas, Adrian Tan, Joe
Lindsey Linder Jim Templin Seiche Sanders, Publisher Tunner, B. Vaithiyanathan, Manu Vora, Keith
Chair Elmer Corbin, IBM QP Editorial Review Board Wagoner, Jack Westfall, Doron Zilbershtein
Chair-Elect Benito Flores, Randy Brull, Chair
Universidad De Monterrey Administrative Committee
Past Chair Eric Hayler, BMW Manufacturing To promote discussion of issues in the field of quality and ensure coverage of all responsible points of view, Quality
Treasurer Francisco “Paco” Lopez, Metalsa 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
SAC Sylvester (Bud) Newton Jr., Alcoa constitute endorsement of that particular product or service by ASQ.

qualityprogress.com ❘ October 2018 QP 5

SEEN&HEARD
Reader reactions from around the world

ADDITIONAL ACTIONS
“Pitfalls and Pratfalls” (September 2018, pp. 42-47) is well
written and addresses important points. One suggestion is to
include the concepts of correction (an interim measure), correc-
tive action (decided based on the root cause analysis to prevent
recurrence) and preventive action. The concept of preventive
action can be elaborated and the fact that corrective action
can’t be a preventive action in all cases must be emphasized.
Sujatha Prakash, Manapakkam, India

CONCEPTS FOR CLIENTS KEEP IT SIMPLE

In response to “Career Coach: “Passport to Success” (July
Build Integrity In” (July 2018, pp. 2018, pp. 20-23) is a great arti-
16-17): Great article. I’ll especially cle. It was straightforward and
keep in mind some concepts— easy to understand—the hallmarks
tone at the top, toxic triangle, of good writing (and an example of the
Robert Sutton’s The No [Jerk] author’s “keep things simple” attribute).
Rule—when I work with clients, Thanks for your consistently high-quality
especially with management. and informative magazine.
Jonathan Patridge, Hopkins, MN Bob Austenfeld, Kobe, Japan

THE REACTION GAUGE

Hundreds of millions of tons of plastic is produced every year, and
last month's much of it ends up in landfills or the ocean. To combat its negative
question ê
this month's question effect on people and the environment, some governments are
banning single-use plastics and organizations are turning to alter-
ê

native materials. What are some of the pros and cons of reducing
With billions of vendors selling their plastic use? How might these changes affect manufacturers?
items on Amazon, many consumers rely
on reviews to guide their purchasing
decisions. And just one bad review—or no Mike Kuklewski, Netherlands, says: Harry Rowe, Indianapolis, writes:
reviews—can significantly affect a prod- In Belgium and France, a plastics Due to the large variety of types of
uct’s sales. Manufacturers are beginning
recycling organization annually recycles plastics, it can be difficult to separate
to realize that product reviews are a pow-
erful key performance indicator (KPI). more than 50,000 tons of plastics into waste plastic by type. It’s also more
What are some other new or nontradi- reusable granulates. Most of these difficult to return it to a form usable
tional KPIs and what do they measure?
secondary resources are intended for for manufacturing. A lot of recycled
the auto industry. The rest is sold to plastics get turned into things like
organizations providing goods to the plastic decking and park benches.
building and agricultural industries. Some “recycling” plants just burn paper
Join the discussion on myASQ at They export granulate to most Euro- and plastic to generate electricity. The
my.asq.org, or on LinkedIn at www.
pean countries as well as the overseas market for recycled plastics is growing,
linkedin.com/groups/3633.
market. It’s a small drop on the giant largely due to a desire for social respon-
plate of plastics waste, but a fine exam- sibility rather than economics. But it
ple of how industry could create its own isn’t nearly as attractive as recycling
circular economy. aluminum.

6 QP October 2018 ❘ qualityprogress.com

 REGISTER NOW
AND SAVE $100!
Early-bird
pricing ends
January 24, 2019.
asq.org/lssc

Lean and Six Sigma This year’s sessions and

workshops will focus on:
in the Digital Age
• The “How” of Lean and Six Sigma:
Organizations are transforming at an Methods for Implementing Projects
exponential rate! Join ASQ to explore the and Initiatives
challenges and opportunities presented by • The Application of Lean and Six Sigma
changes in the workforce, the workplace, in Nonmanufacturing Environments
and the markets organizations serve. • The Human Side of Lean and Six Sigma
This year’s dynamic program will offer • Lean and Six Sigma Fundamentals
best practices, tested solutions, and proven • Advanced Content Masters Series
results for applying lean and Six Sigma
tools and principles in the digital age. Save $100 and register today:
You can’t afford to miss it! asq.org/lssc
 Subject matter experts take on your quality-related queries

THIS MONTH'S QUESTION OUR RESPONSE

What are the In practice, problem solving and RCA are interchangeable.
differences between Technically, problem solving involves assessing a process from end
to end—from the event occurrence to the sustained prevention of
problem solving recurrence. A typical problem-solving approach is the eight disci-
and root cause plines method.
analysis (RCA)? RCA, on the other hand, involves investigating the incident, and
What are some identifying and confirming the root causes. Some organizations also
differentiate between RCA and root cause corrective action, which
tips for effectively involves identifying and confirming the root cause, and implementing
and efficiently corrective actions.
facilitating Here are some of the most important aspects of facilitating prob-
problem solving, lem solving and improvement projects:

RCA or both? Prior

Before problem solving, the problem-solving team must select a
trained facilitator. Some organizations expect the problem owner to
also function as the facilitator, but that could be problematic. Problem
owners typically are under pressure to resolve an issue as quickly as
possible, which means they may pressure the problem-solving team
to reach a temporary conclusion quickly rather than achieving a per-
manent, sustainable solution.
A facilitator, on the other hand, has a neutral role and can ensure
the RCA process is followed. The facilitator should complete the fol-
lowing tasks before the problem-solving team meeting:
++ Work with the problem owner to ensure the problem-solving team
has collected adequate background information and pertinent data
affecting business metrics.
++ Work with the problem owner to present background information
and the business impact to management for sponsorship. Typically,
problem-solving efforts fail without management’s commitment.
Having management’s support also will make it easier for the
problem-solving team to get the resources and support it needs to
follow through with corrective actions.
++ Train team members in the structured problem-solving process and
basic quality tools. Team members must understand various terms
such as symptom, root cause, containment, correction and correc-
tive action. It also is important for team members to understand the
stages of team dynamics.
++ Accommodate virtual team members who will be joining the meet-
ing via video or web conferencing. Assure them that they are as
important as onsite participants.
++ Work with the problem owner to ensure the problem-solving team
is cross-functional and includes supplier and customer representa-
tion when appropriate (this may require management support). A
diverse problem-solving team offers unique perspectives.

8 QP October 2018 ❘ qualityprogress.com

++ Be cognizant of the organizational
environment (layoff announce-
ments, mergers and acquisitions,
product recalls, legal proceed-
ings)—team members’ minds may
be elsewhere. Get the sponsor—a
management-level individual rele-
vant to the problem area—to attend
the meeting to answer questions
and alleviate any concerns.

During
During the problem-solving process, time management is important, team Reach out to virtual team members
the facilitator should: dynamics are more important for for their feedback and allow all par-
++ Revisit the project scope in every continuity of problem solving, so take ticipants to communicate directly in
meeting and ensure it doesn’t expand. breaks to defuse the situation. case of questions or concerns.
++ Understand the competing priorities ++ Don’t get hung up on a tool’s theoreti-
for resources in the organization cal approach—adapt it to the problem After
and be flexible. For example, ensure at hand. Engage participants by After the problem-solving project,
the timing and duration of meetings making it relevant. Modify five whys the facilitator should work with the
allow participants to perform their analysis from a single causal chain to problem owner and the sponsor to
daily tasks. a multi-branching fault tree analysis, acknowledge the team’s accomplish-
++ Set realistic ground rules for meetings for example. Or, modify cause and ments and communicate successes
(as an input from the team itself) effect diagram labels when analyzing and lessons learned with other
and enforce them. Implement a a software problem. problem-solving teams. Develop and
laptop-down policy, for example, so ++ Keep the use of tools to a minimum nurture the new batch of problem
participants stay focused on dis- and make them simple and easy. solvers—don’t let this become a for-
cussions and participation. Provide Problem solving is about getting gotten skill. And don’t forget to thank
10-minute breaks every hour so results, not a participant showing off management for its support through-
participants can catch up on emails, his or her ability to use complex tools out the process.
make calls and ensure business and analyses.
BIBLIOGRAPHY
continuity. ++ Stay neutral and resist aligning ASQ, “Problem Solving,” asq.org/learn-
++ Solicit input from quiet participants with the authority or influence in about-quality/problem-solving/overview/
by redirecting the conversation the room. Respectfully challenge overview.html.
ASQ, “What Is Root Cause Analysis (RCA)?”
and keeping everyone engaged. assumptions and hypotheses. Ask ASQ, asq.org/learn-about-quality/root-
If a participant is dominating the for data. Make everyone—not just a cause-analysis/overview/overview.html.
conversation, acknowledge his few members—feel important. Ramu, Govind, “Expert Answers: July 2018,”
Quality Progress, July 2018, pp. 8-9.
or her contributions and tactfully ++ Schedule regular meetings with Wikipedia, “Tuckman’s Stages of Group
request that he or she allow others the problem owner outside of the Development,” https://en.wikipedia.org/
to contribute. Use tools such as silent problem-solving activity to discuss wiki/Tuckman%27s_stages_of_group_
development.
brainstorming on sticky notes first to adherence to scope, progress,
get inputs from reluctant participants. escalation to sponsors, questions,
++ Observe participants’ body language concerns about participants’ behav- This response was written by Govind
Ramu, Program Manager, E2E Closed
and anticipate frustration, anger, ior or performance, or the need for Loop Learning, Google, Mountain View,
tiredness and dismissiveness. While additional training, for example. CA.

qualityprogress.com ❘ October 2018 QP 9

 A digest of
trends, research
& late-breaking
news

E-COMMERCE

The
Ratings
Race
The surprising effect of
Amazon reviews

In a world in which millions of brands compete to sell According to Shedletsky, manufacturers are still
similar items, product reviews can be a key decision adjusting to this new reality. “Whereas five years ago,
making factor for consumers trying to decide which they were able to focus on interacting with unhappy
product to buy. Do you choose product A because customers who left reviews on their own websites, now
it’s cheaper but has less favorable reviews? Or, do you they mine the information left by customers on Amazon
buy the more expensive product B with hundreds of for clues,” said Shedletsky.3
positive reviews? As part of her research, Shedletsky interviewed
And perhaps the most trusted and looked to source several industry leaders about this new metric, one
for product reviews is Amazon. So much so, in fact, that of whom was the vice president of operations at a
some organizations have started using the reviews and major cell phone manufacturer. He said: “If you have
ratings as key performance indicators. consistent issues across the reviews, and the company
According to marketing research firm BloomReach, doesn’t respond, this becomes a critical problem for
people trust Amazon reviews so much that they will customers … You can use Amazon reviews to quantify
look at product reviews on the online retailer’s site even how bad quality affects your sales.”4
if they’re going to buy the product somewhere else.1 Despite this news coming to the forefront only
Author Anna-Katrina Shedletsky said the need for recently, a 2006 research paper revealed the same
positive reviews can have detrimental effects. information. Researchers Judith Chevalier and Dina
“Amazon reviews are such a strong driver of sales Mayzlin studied the effect of Amazon book reviews on a
(even for brick and mortar store sales), that a few bad book’s sales. “An increase in the average star rating on
reviews soon after a product ships can have real drag Amazon.com over time results in higher relative sales of
on the sales and resultant revenue for a new product,” the book on Amazon.com over time (one month after
Shedletsky said.2 the reviews under consideration have been posted),”

10 QP October 2018 ❘ qualityprogress.com

the researchers wrote.5 AUTOMOTIVE
The researchers’ findings also revealed
that one-star reviews are the most
impactful. A Dimensional Research survey Will Tariffs Hurt Customer
Satisfaction With Cars?
backed up that finding: 86% of respon-
dents said that their decision to buy is
affected by negative reviews.6
Customer satisfaction with cars went up this year, but some fear that
Credibility proposed tariffs could put a dent in this gain.
Another factor working against manu- According to American Customer Satisfaction Index’s (ACSI) Auto-
facturers selling on Amazon is the vast mobile report, customer satisfaction with autos and light vehicles
number of positive reviews. BestReviews, increased 1.2% to a score of 82 on the ACSI scale of 0 to 100. An ASCI
a product review site, analyzed 360,000 official warned that proposed tariffs by the Trump Administration could
ratings for 488 products and found do some harm.
an overwhelming number of five-star “Proposed tariffs on auto imports add to the pressure of rising metal
reviews—66.3%, to be exact.7 costs for both international automakers and American-made cars using
But just how credible are those reviews? foreign parts,” said David VanAmburg, managing director at ACSI.
If its reviews and ratings have such a “We’ll be watching how the threat of higher prices affects customer
significant impact on product sales, satisfaction in the coming year.”
consumers obviously put a lot of stock in Subaru, Toyota and Honda topped the list of high customer satisfac-
them. tion scores.
Matt Moog of Power Reviews, which For more results, visit https://tinyurl.com/customer-sat-cars.
makes ratings and review software, said
that just one positive comment on a prod- ASQ

QUALITY 4.0 SUMMIT NEARS

uct increases its purchase rate by 65%. He
also said that one-third of people admit
they won’t buy a product if it hasn’t been The second annual ASQ Quality 4.0 Summit on Disruption, Innovation and
positively reviewed.8 Change will be held Nov. 12-13 in Dallas.
As Chevalier and Mayzlin’s research The conference features presentations, demonstrations and discussions
shows, however, when new products are led by technology experts, industry leaders and quality professionals. The
first released on Amazon, they don’t have four focus areas of the summit include:
any reviews or ratings yet. 1. Industry 4.0—Learn about the new forces of change in organizations,
“This creates a chicken-and egg-prob- markets and the future of work.
lem for Amazon and the product’s 2. Digital transformation—Examine methods, models and frame-
manufacturers,” the study said. “Users works organizations can use to successfully navigate today’s digital
will generally not buy a product when it transformation.
doesn’t have reviews, and if users aren’t 3. Managing change—Position your organization for change, implement
buying the product, there are no reviews needed changes and address the fast pace of change.
being written.”9 4. The future of quality—Uncover opportunities for quality professionals
To address the issue, manufacturers to lead innovation and to embrace disruptive technology’s impact on the
provide their products to reviewers for quality function and profession.
free or at a discounted price. Amazon For more information and updates on conference events, as well
allowed the incentivized reviews as long as announcements on keynote speakers, visit asq.org/conferences/
continued on page 12  quality-4-0.

qualityprogress.com ❘ October 2018 QP 11

The Ratings Race Clarification
The article “Valid or Not?” (July 2018, pp. 36-42) referred to a
continued from page 11
database that is shown to be associated with the Independent
as the reviewers disclosed that they received compensation in Association of Accredited Registrars (IAAR). As stated by the
exchange for their comments and ratings. IAAR, the database referenced is operated by an organization
However, some products were receiving suspiciously high that IAAR severed its relationship with in July 2015. Yet, the
numbers of five-star reviews just days—or even hours—after organization continues to make the database available despite
they were listed for sale on Amazon.10 IAAR’s requests. The link to the database has been removed
Amazon attempted to tackle the issue by banning incentiv- from the IAAR website and is no longer accessible from the
ized reviews in 2016, but sellers adapted. Now, they’re culling IAAR website. Since the dissolution, IAAR has made several
favorable reviews by reaching out to people via Facebook.11 attempts to ensure that the information is removed from the
A simple search for “Amazon Reviews” on Facebook reveals a web with no success or response from the organization.
lengthy list of groups dedicated to incentivized reviews. The authors of “Valid or Not?” were unaware of the dispute
regarding this publicly accessible database. However, this
Pressure on manufacturers serves as an example highlighting the need for the Inter-
According to Tommy Noonan of ReviewMeta, all of this spells national Accreditation Forum (IAF) database of accredited
failure for reputable sellers. certification, which will provide a reliable, global single source
“These days, it is very hard to sell anything on Amazon if you of truth to verify certificate validity.
play fairly,” said Noonan. “If you want your product to be com- As accredited certifications to ISO 9001:2008, ISO
petitive, you have to somehow manufacture reviews.”12 14001:2004 and many industry standards, such ISO/TS
Fraudulently rated products shoot to the forefront of Amazon 16949, were either withdrawn or expired on Sept. 14, 2018, it is
searches, pushing reputable sellers’ products to the bottom of important that organizations verify the validity of these cer-
the list. And if a product has negative reviews, consumers will tifications when relying on them for supply chain decisions.
likely pass over it altogether. Although some databases exist, some may be inaccurate or
In today’s world of one-click buying and instant gratification, unreliable. The authors encourage you to contact the organi-
this puts tremendous pressure on manufacturers to create the zation, the conformity assessment body or the accreditation
best quality product the first time, leaving very little margin for body directly to verify the status of accredited certification.
error. The authors thank the IAAR for making the public aware that
—compiled by Lindsay Dal Porto, assistant editor a database included in their article may be an inaccurate and
unreliable source for verification of accredited certification. The
EDITOR’S NOTE authors look forward to sharing more information regarding
References listed in this article can be found on the Progress the IAF database of accredited certification after it is available.
Report webpage at qualityprogress.com.

Mr. Pareto Head By Mike Crossen

12 QP October 2018 ❘ qualityprogress.com

QUALITY-RELATED
NEWS FROM AROUND
THE WORLD GETTING TO KNOW…

John Karlin
—powered by Lexis Nexis

Is That Really Tuna? Study Suggests

44% of Canadian Seafood Mislabeled
Almost half of the seafood sold in Canada is mislabeled, current position education

according to a new study. The Oceana Canada study, Clinical RN and team leader of Bachelor’s degree in nursing
released this week, looked at 400 seafood samples the therapeutic hypothermia from Oklahoma Panhandle
team in the neonatal intensive State University in Goodwell.
throughout Vancouver, Victoria, Toronto, Halifax and care unit (NICU) at Children’s
Ottawa, and found that 44% wasn't what it appeared to be, Hospital at Oklahoma State
University.
and it was nearly impossible to track from origin to plate.
For more, visit: https://tinyurl.com/ycassckt. What was your jötunn stares you in going somewhere
the face and threatens or doing something.
introduction to to take you down, To be motionless
It’s Cool to Be a Manufacturing quality? stare back and smile, in absolute silence,
improvise, adapt and even if just for a
Engineer Major Again overcome. moment, is golden. A
The number of individuals earning degrees in engineering It started with my glass of wine helps,
internal questioning, too.
has increased steadily over the past decade, and those “Why do we keep Do you have a
numbers will only continue to grow. In particular, one doing things the same
way without trying to mentor who makes Are you active in
subdiscipline of engineering is looking ever more appealing improve anything?” a difference in
to students: manufacturing engineering, which focuses on Frustrated, I thought ASQ?
there must be some your career?
improving the production of an item. Read the full story education or training I currently serve as
here: https://tinyurl.com/y7zj5gcu. on how to do things Susan Bedwell, who chair of the Okla-
better, which led me holds a doctorate of homa City Section.
to lean and Six Sigma, nursing practice, is
10 Things Customer Success and ultimately, ASQ. an advance practice
Leaders Do for Their Clients registered nurse and Have you
the clinical nurse spe- had anything
Customer service means customer success, and as the field Previous note­ cialist for our NICU.
grows, more leadership roles open up. Customer success worthy jobs? She has always pro- published?
vided me direction,
leaders need strong visionary and management skills, but resources, recommen- A handful of articles,
I worked as an ammu- dations, opportunities
leadership encompasses many vital components to serving nition specialist for and encouragement.
but most recently,
the U.S. Army. It was a “How to Apply 5S:
clients. There are 10 things customer service leaders do for I don’t think I would The Frightening
job many are capable be where I am now
their clients. Read more: https://tinyurl.com/ycep2769. of doing but few without her.
Fridge at Work,”
will do. This was my which appeared
introduction to supply on the website
chain, lot numbers goleansixsigma.com
To get a roundup of the week’s most noteworthy stories Any recent honors in August 2016.
and hazardous mate-
delivered to your inbox every Friday, subscribe to the QNT rials handling. I quickly or awards?
Weekly e-newsletter at asq.org/newsletters. learned that bullets
don’t fly without What was the last
Recently, I received
supply! the Academy of movie you saw?
Neonatal Nursing’s
ISO Excellence in Neona- “Thor” about seven
What’s the best tal Nursing Practice years ago.
New Guidelines Help career advice
you’ve ever
Award.

In Hiring Process received? Personal:

Quality quote:

New guidelines from the International Organization for Many see the world
Be relentlessly per- Wife Carol and two through the lenses of
Standardization (ISO) can help organizations determine how sistent. Don’t let any sons, Antonius and welfare and entitle-
well they recruit new employees. rejections, defeats, Michelozzo. ment. I see the world
ISO/TS 30411:2018—Human resource management—Qual- failures or discourage- through the lenses
ments ever deter you of opportunity and
ity of hire metric includes a range of options to measure the from your pursuits. What are your potential.
quality of a hire, which can be aligned to various business We all face obstacles favorite ways to
and organizational conditions. The document also includes in life. Some of these
obstacles are giants, relax?
international best practices. which we think we
For more on the guidelines, visit www.iso.org/ may not overcome. To just be still and
When a giant or a quiet. I’m constantly
standard/68220.html.

qualityprogress.com ❘ October 2018 QP 13

 Advice to advance your career

R I S K A N A LY S I S

A Passion for
Pizza—and Risk
Analysis
by Denise Wrestler
How a common tool helped one quality
professional pursue an unlikely dream

I’ve always dreamed of owning my own restaurant. My child- to finally try the amazing pizza we’d been
hood was filled with Play-Doh pizza sets that I would serve to the telling her about.
family dog. I’ve been told my first word was “pepperoni.” Instead, when we pulled up to the restaurant,
Just kidding—I’ve never wanted to own a restaurant. My idea we were greeted by a “Closed for good” sign
of making pizza is opening a box or removing a wrapper, and I’m posted on the window, chains on the doors and
certain my first words were “Document it or it didn’t happen.” (I darkness throughout the empty building.
was a talkative baby.) Our jaws dropped. It was like someone
But it seems like that’s how every dream-come-true story had ripped out our pizza-shaped hearts. My
starts—with a black-and-white flashback to someone’s early begin- six-year-old daughter had huge alligator tears
nings, foreshadowing a master in the making. welling up in her eyes, and my 10 year old was
Then there’s me: I have 15 years of quality and regulatory yelling, red in the face, about to burst into tears.
experience in the medical device field, my husband has 20 years’ Our family was heartbroken.
experience in finance, neither of us have any prior restaurant expe- The “For lease” sign posted in the window
rience—and we’re opening our first pizzeria. and the unpainted trim outline where the
So, how does a couple like us decide to open a pizza restaurant restaurant’s sign once hung were constant
anyway? For the past two and a half years, we’ve wondered the reminders of the deliciousness that used to be
same thing. Why would self-proclaimed sane and financially savvy there. We patiently waited for years, hoping
professionals make such a seemingly ridiculous decision? someone would bring it back to life.
Our story is a roller coaster—one I hope will inspire and excite It wasn’t until passing by that empty space
others to follow their passions and take risks, but only after the for the thousandth time that I came up with
appropriate risk analysis, of course. an idea. I frantically pulled out my cellphone
and called my husband. Our conversation went
Preheat oven to 300 degrees of inspiration something like this:
In December 2013, my mom came to visit me and my family. “Hey honey, you know the Pizza Inn we loved
Between the long sessions of holiday shopping, we decided to that went out of business?”
stop at our favorite local pizza joint, Pizza Inn. My mom was excited “Yeah,” he responded.

14 QP October 2018 ❘ qualityprogress.com

 “You know how we’ve been waiting for someone to revive it?” I 2. “Will you quit your jobs and do this full time?”
asked. 3. “How can you afford this?”
“Um, yeah,” he said tentatively. The questions were posed differently every
“Why don’t we bring it back?” time—some included curse words or tears of
After a stunned silence he said, “Because we don’t know any- concern—but they all surrounded the same
thing about owning a restaurant.” three topics. Our friends and family were right
“But we could learn!” to ask these questions, because we also had
That single conversation was the tiny snowflake that ultimately asked them many times.
created the giant snowball that changed our lives forever. A few I never believed in the age-old adage,
emails, conference calls and a very long franchisee application “Things happen for a reason.” But I do now.
later, we were on our way to reviving our beloved pizza restaurant. There is absolutely nothing inconsequential
My passion for not only pursuing a new money-making about how three major elements in our lives
endeavor, but also fulfilling a community demand was the driving all changed in such perfect harmony: Thanks
factor—the action that created the need for risk analysis in the first to the changing housing market and election
place. year, my husband’s income as a mortgage
Creating a list of all possible hazards and risks is the first step to broker began to suffer, our retirement portfolio
any good risk analysis. But for me, identifying the biggest risk—our of investment properties could no longer grow
inexperience—and being able to immediately mitigate it through and my job was relocated to Mexico. It was a
ridiculous amounts of training had to be addressed before our new trifecta of bad news.
endeavor could go any further. Or was it? Instead of returning to industry, I
decided to consult full time, which resulted in a
Gently knead (need) the dough flexible schedule and pay bump. Our retirement
When telling our friends and family about our new venture, the plan, cleverly disguised as a property manage-
same three questions always came up: ment organization, was due for a makeover,
1. “Why would you open a restaurant when you don’t know the and the housing market was primed and ready
first thing about owning one?” for liquidating and reinvesting. My husband,

qualityprogress.com ❘ October 2018 QP 15

more than ready for a career change, was now hopefully be our sizeable retire-
in the perfect spot to do so with minimal risk. ment fund.
With a clear path forward, the final concern Throughout this entire process,
that constantly resurfaced was our lack of we constantly remind ourselves
experience. Being business-savvy individuals, why we continue to persevere.
we didn’t throw our lives into this opportu- Like any good quality process, we
nity blindly. We researched, called, emailed, consistently recall our root cause.
calculated and planned. We read every line of We remind ourselves what our
every agreement that was put in front of us. We ultimate outcome will achieve.
learned and absorbed information like sponges. It’s about giving back. Often, just a glance at our
We worked on a mitigation plan for the risks we We’re excited to be able children’s faces, a random tweet
had identified. And the more we learned, the to promote community from an excited Pizza Inn fan and
more questions we had, which led to even more events in our restaurant. our dwindling bank account is all
mitigation. it takes to keep the motivation
Slowly but surely, we became more confident going.
in our venture. Our lack of experience even- I’ve written articles on risk assessment, career changes and
tually became a hurdle small enough to jump other life-changing events. I’ve practiced my own preaching
over. Our mitigation efforts proved effective, throughout my life—but this life-changing event has by far taken
lowering the severity of our risks to an accept- the cake. Unlike simple job changes, this risk assessment affects
able level and giving us the confidence to the greatest amount of my future. My analysis includes the lifespan
continue forward. of this project and—if my calculations are correct—will take me well
into retirement. I’ve given birth to another child in my life, a cheesy
Top with cheese, delicious one I’ll love unconditionally forever. The longevity of my
pepperoni, sweat and tears risk taking has been a huge consideration throughout this process.
According to my children, and several indi-
viduals who probably don’t know any better, Slice and serve
making pizza is fun and easy. How hard could it The signs are up—our restaurant is “coming soon.” The community
possibly be to roll out dough, slather on sauce, is excited about what we’re bringing to their town, and we’re so
sprinkle it with toppings and throw it in the excited to be the ones to deliver it.
oven? It’s not just about the mouthwatering pizza, it’s also about giving
As it turns out, very—it is very hard. It’s fun, back. We’re excited to be able to promote community events in
but it is far from easy. our restaurant, host kids’ nights and senior nights, and fundraiser
Just having a place to make pizza was nights for local nonprofits (fittingly named “doughraisers”).
challenging, and I won’t even mention our labor Being able to chase our dream—while also serving our com-
struggles. Finding people to actually make the munity—has become more rewarding than I imagined. How our
pizza was a feat in and of itself. actions are affecting others is proving to be a larger contributing
As we progressed in our endeavor, more and factor to our risk analysis than I initially budgeted.
more risks that we had overlooked were begin- I hope that my story of entrepreneurship and risk-taking inspires
ning to surface. There were days we asked others to think outside the box. It’s never too late to chase your
ourselves, “What in the heck are we doing?” passion. You’re never too old to gamble with your future. Just
There were moments we got cold feet and make sure the risks you take are appropriate for you. 
considered turning back. But throughout this
journey, we overcame every doubt and are still
trekking down that long, bumpy road to making
our dream a reality. The risks were plentiful, but
the benefits definitely outweighed all the risks
we were taking.
Denise Wrestler is an independent quality
Bake for 20-25 years assurance/regulatory assurance consultant
Our dream, like many others, was built on two for CYA Medical Device Consulting in Dallas.
She holds a bachelor’s degree in chemical and
specific, long-term goals—bring our favorite biomedical engineering from the University of
pizza place back into our community while California, Irvine. An ASQ member, Wrestler is an
also establishing the groundwork for what will ASQ-certified quality auditor and engineer.

16 QP October 2018 ❘ qualityprogress.com

 QUALITY 4.0 SUMMIT
on DISRUPTION, INNOVATION, and CHANGE
November 12 – 13, 2018 | Dallas, TX

They’re Ready for Change. Are You?

Here’s how ASQ’s Quality 4.0 Summit made an impact on last year’s attendees:

“ Enhanced my understanding of the

trends related to digital transformation.” “ It included a lot of data and information
sources as well as amazing networking.”

“ It served as a reminder of how the pace of change

will change our approach to quality improvement.” “ Great topics, great interaction,
and even better collaboration.”

“Provided me insight and help for getting the

message back to our leadership team.”

Find out how this year’s summit can change your organization.
Register today: asq.org/Quality4.0Summit
F E AT U R E
BLOCKCHAIN

The basics behind blockchain and why it will

transform supply chains | by Narahari Rao

18 QP October 2018 ❘ qualityprogress.com

H
Humankind has been through several time periods, including the Stone
Age, Bronze Age, Iron Age and Jet Age. Without question, we are now liv-
ing in the Information Age. Examining the enormous progress made in just the
past 10 years, the pace has been exciting and surreal. From the miniaturization
of computers to laptops to smartphones to the ubiquity of internet and the
associated retinue of internet-connected devices, we are living through what
seems to be a fast-forwarded movie.
Now added to this time of great progress are some new innovations that
have the potential to shape the next technological revolution. Among these
innovations are a few that have managed to capture people’s imagination
and occupy the high altar of media coverage: blockchain technology, artificial
intelligence (AI) and machine learning.
Blockchain technology is already the driving force behind cryptocurrencies,
such as Bitcoin. Machine learning is being used to train cars to self-drive, while
AI already has a place in many people’s homes in the form of personal assis-
tants, such as Alexa and Siri.
Much of what’s known about blockchain originated in a paper written by Just the
Satoshi Nakamoto.1 He introduced Bitcoin, a cryptocurrency that has become Facts
a familiar term to many, and along with it, the technology of blockchain was
born. From Nakamoto’s paper, the words “block and chain” are used sepa- Blockchain tech-
nology is one of
rately. But with use, the two words joined together to form “blockchain.” the latest innova-
The foundation for Bitcoin lies in the digital trust that blockchain brought tions changing the
way businesses
with it. Blockchain technology helped to decentralize the currency while buy and sell
simultaneously preserving anonymity of transactions and ensuring all transac- things.
tions were recorded in public spaces. Quality impli-
It is this power of storing information, maintaining anonymity and prevent- cations include
improvements
ing unauthorized changes that attracted people to discover alternative uses in supply chain
for blockchain technology. The immediate (and most logical) application of management
practices.
blockchain was in financial transactions. Increasingly, its power is being har-
nessed in various other sectors, such as healthcare, insurance, transportation For blockchain
and supply chains as well as the basic democratic exercise of voting. technology to
truly optimize
The ability of blockchain technology to keep a tamper-proof digital foot- supply chains,
print is what excites quality professionals. organizations
must keep an
open mind about
What is blockchain? learning and
understanding
Blockchain is “a mathematical structure for storing data in a way that is nearly how the technol-
impossible to fake.”2 Essentially, it’s a database that is validated by a larger ogy can affect
existing systems
community rather than by a central authority (that is, banks or governments). and processes.

qualityprogress.com ❘ October 2018 QP 19

 F E AT U R E
BLOCKCHAIN

At its very core, a blockchain is like a tra- of the product—from inception to consumption
ditional ledger that keeps records of all the and to eventual redemption. The most inno-
individual transactions made against the prod- vative part about blockchain is its verification
uct. These transactions can occur in any part of system and inherent security.
the supply chain. Blockchain technology has the potential to
The transactions could be any movement change the way we buy and sell things. The
of money, goods or secure data related to the verification and authentication of the product
product—that is, the purchase of raw materials, at every step will impart greater traceability, as
the blending of materials to make a new mate- well as the ability to trust the supply chain. It is
rial, purchase orders to buy new material, quality virtually impossible to change this data with-
inspection, certificate inspection, invoicing and out being detected by other users because no
bank payment. Each of these transactions is one person owns the entire data set. The other
stored digitally as a block within the ledger. great advantage of blockchain is that every
Each of the blocks is timestamped and becomes organization or vendor in this supply chain could
intrinsically linked to one another. The blocks are potentially use a different system, and it does
strung chronologically into a chain—hence the not affect the integrity of the blockchain itself.
term “blockchain.” Consider this scenario that’s relatable to any
To draw an analogy, if each transaction was industry—with a few big assumptions:
a picture, stringing them together and looking Company B wants to buy product X from
at them holistically renders a movie—a history vendor A. Product X has many raw materials,

FIGURE 1

Process flow with blockchain

Purchase order 1 Purchase order 2
Number: Number:
Company B Quantity: Vendor A Quantity:
Delivery request date: Delivery request date:

Purchase order Purchase order 1 Purchase order

• Supply product X
Procure product X Invoice 2 payment
• Invoice company B Invoice 1
Reference
Quality inspection purchase order:
Number: Quantity:
Inspect quantity: Cost:
Quality Rejected quantity:
Purchase

Invoice 2
order 2

Inspection details Material

shipment Material certificate
Heat certificate:
Batch:
Invoice 1 payment Distributor Material properties:
Reference
Accounts payable purchase order: Purchase order
Quantity:
Invoice processing Cost: • Supply base metal Invoice 2
Invoice 1 Reference
• Invoice vendor A purchase order:
Quantity:
Cost:

20 QP October 2018 ❘ qualityprogress.com

FIGURE 2

Visual representation of a blockchain

Purchase order 1
Purchase order 2
Material certificate
Invoice 2
Invoice 2 payment
Quality inspection
Invoice 1
Invoice 1 payment

including some base metals that vendor A must base metal to vendor A, its block includes the
source from a global metal distributor. Company sources of the raw material. This information
B issues a purchase order to buy a certain quan- gets tagged to vendor A’s purchase order
tity of product X from vendor A. and, consequently, to company B’s purchase
Because company B is buying this product order as well.
for the first time from vendor A, company B has ++ The metal, along with other raw materials, are
instituted a strict receiving quality inspection processed in vendor A’s plant, and product X
and certificate inspection as part of its process.
After the process and all the raw materials are
verified, including material dimensions, the
quality team accepts the raw materials into its
inventory. Vendor A submits an invoice against A great advantage of blockchain is that
the purchase order. Company B processes the every organization or vendor in this
invoice and submits it to the bank for payment supply chain could potentially use a
to complete the transaction. different system, and it does not affect
As shown in Figure 1, there are various the integrity of the blockchain itself.
transactions involved in this seemingly simple
scenario:
++ The creation of company B’s purchase order
is the first block in this process. The block
itself could have information pertaining to the
quantity of product X, cost and a need-by
date—information gleaned from the purchase
order. This block is timestamped and identi-
fied using a unique identifier. The blockchain
network verifies the veracity of this purchase
order and validates it.
++ Using this purchase order, vendor A sends
its own purchase order to the global metal
distributor, asking for this certain base
material (needed to make product X). This
creates a block inextricably linked to the
previous block (see Figure 2).
++ The global metal distributor buys its
metal from a few foundries around the
world. When it makes that shipment of

qualityprogress.com ❘ October 2018 QP 21

 F E AT U R E
BLOCKCHAIN

is fabricated. All the routings, machining opera- is meant to be used. What if this product
tions and internal quality inspections make it to suffers a metal fatigue failure? The history
product X’s production order, which in turn is tied of product X—including the batch and serial
to company B’s purchase order—thus creating a number of the metal used during produc-
new block. tion—can be identified using its blockchain.
++ Product X shows up in company B’s quality Potentially, other affected products using
inspection department, where the quality depart- the same metal from the same failed batch
ment can verify the entire supply chain, material and serial number can be identified quickly,
certificates, and vendor A’s internal manufactur- thus preventing other failures.
ing documents and quality inspection results. The The simple idea behind blockchain is that
quality department may perform its own quality information becomes part of a chain and
inspection, thus recording the results in another gets its own key, which matches the lock of
block linking it to the purchase order itself. the previous block. If the information in one
++ After product X is received into inventory, vendor block is tampered with, the key and lock
A submits an invoice against the purchase order. will not match anymore, thus effectively
The invoice is compared to the purchase order providing a high level of data security and
and validates whether the quantity and price on fraud prevention.
the invoice match the purchase order, thus creat-
ing another block in this chain. Supply chains and blockchains
You could extrapolate this scenario with prod- Who can forget the horse meat scandal
uct X being shipped to a remote location where it that rocked the United Kingdom in 2013,
when horse meat began to appear in the
beef hamburgers and meatballs sold in
supermarkets? It took weeks before the root
cause of this error was identified. The root
The simple idea behind blockchain cause was based in the labyrinth of Europe’s
is that information becomes part of processed meat supply chain. The meat, in
a chain and gets its own key, which one instance, supposedly traveled from a
matches the lock of the previous block. Romanian slaughterhouse to meat traders
in Cyprus and the Netherlands to a supplier
in France and then to a processing plant
in Luxembourg, thus opening the door for
the integrity of the meat supply chain to be
compromised.
A complex supply chain isn’t—and
shouldn’t necessarily mean—a misman-
aged one. However, a lengthy supply chain
requires a level of oversight that matches
its complexity. With a globalized supply
chain and thousands of components from
various sources going into our products, it is
imperative that we inject a sense of trace-
ability and quality into our supply chain. The
first step to quality lies in knowing where
the parts for our products are being sourced
from.
Of course, as with any technology, there
are downsides:
1. Adopting blockchain as an integral part of
the supply chain is a big departure from

22 QP October 2018 ❘ qualityprogress.com

 traditional processes. Much of supply chain It will enable transactions to be automatically validated
still dabbles in the 20th century in terms of without a need for human validation, thus bringing vast
traceability and verification. The adoption of improvements in efficiency and speed.
blockchain technology must first cross the The key to any successful adoption lies with the
hurdle of change management. technology running in the background—with the aver-
2. It is said that a chain is no stronger than its age user not needing to traverse the narrow bylanes
weakest link. If a vendor in your supply chain of understanding how the technology works. The key
is not judicious about entering accurate takeaway for all of us isn’t learning how to program a
information into the blockchain, your supply blockchain—it’s keeping open minds about learning and
blockchain is only as strong as that vendor. understanding how it will affect existing systems and
3. Everyone in the chain must adopt and adapt, processes.
otherwise the process breaks down. If a manufacturer using blockchain technology can
4. There is an obvious tradeoff in terms of cost. provide a quality product to its consumers with full
With any burgeoning technology, until it is traceability and assurance, it has already distinguished
widespread and commonly used, costs could itself from the competition. The future shall belong
be prohibitive. In the short term, big corpo- to those who can optimize and finetune these supply
rations may be the only ones that can sustain chains to their advantage.
this technology, with just portions of the Blockchain inherently drives big data, bringing
supply chain being blockchained. information on everything related to the efficiency of
5. As witnessed recently, no technology is 100% the entire process, how optimal (or suboptimal) the
secure. There are bound to be security con- transportation network is, the efficacy of the supply
cerns with blockchain technology, and they chain, vendor performance and product quality.
must be alleviated appropriately. The development of a cottage industry devoted to
Currently, we are slowly reaching the inflec- fixing or improving each of these subprocesses is not
tion point, after which blockchain technology too distant of a pipe dream. There already are organi-
will involve all of our global supply chains. The zations offering services relating to blockchain—akin
scalability—or the lack of it—is what is hold- to cloud services offered by some organizations. With
ing back early adoptions of this technology. a little bit of customization and adaptation, organiza-
Because by definition, a blockchain transaction tions can blockchain their transactions and their entire
must be processed by every computer con- supply chain.
nected to the network. Even though this leads All of this is coming soon to a supply chain near you.
to exceptional security, we have sacrificed effi- Bank on it.
ciency. Innovation in blockchain scaling is the
REFERENCES
leap that will lead to its widespread adoption. 1. Satoshi Nakamoto, “Bitcoin: A Peer-to-Peer Electronic Cash
Remember the transition from cash to plastic System,” bitcoin.org, https://bitcoin.org/bitcoin.pdf.
(credit and debit cards) for our daily commerce 2. Massachusetts Institute of Technology (MIT) Technology
Review editors, “Explainer: What is a Blockchain?” MIT
needs? Cards were notorious for their lack of Technology Review, April 23, 2018, www.technologyreview.
security because many scammers could easily com/s/610833/explainer-what-is-a-blockchain.
steal information and identities. There also was
EDITOR’S NOTE
a lack of vendors that initially accepted cards. The views and opinions expressed in this article are those of the
Today, using credit cards has become second author and do not necessarily reflect the official policy or position
nature to many. Credit cards seem more secure of his employer.
than ever before. Financial institutions that
issue them offer much more customer protec-
tion. You are more than likely to find a vendor
willing to accept plastic than not.
Much of what we know about e-commerce Narahari Rao is a business process
today would be absent if not for this plastic architect at Schlumberger in Houston. He
holds a master’s degree in mechanical
revolution. It is that same transition that we engineering from Texas A&M University in
must make in terms of blockchain technology. College Station. He is a member of ASQ.

qualityprogress.com ❘ October 2018 QP 23

F E AT U R E
QUALITY 4.0

Let’s Get
Digital
The many ways the fourth industrial
Just the revolution is reshaping the way we think
Facts about quality | by Nicole Radziwill

To adapt to
the disruptive
changes brought The technology landscape is richer and more promising
about by the than ever before. In many ways, cloud computing, big
fourth industrial
revolution, a data, virtual reality (VR), augmented reality (AR), block-
new perspective chain, additive manufacturing, artificial intelligence (AI),
on quality is
required. machine learning (ML), Internet Protocol Version 6 (IPv6),
cyber-physical systems and the Internet of Things (IoT)
Organiza-
tions must all represent new frontiers. These technologies can help
leverage Quality improve product and service quality, and organizational
4.0 tools—such
as artificial intel- performance.
ligence, machine In many regions, the internet is now as ubiquitous as
intelligence, per-
vasive computing electricity. Components are relatively cheap. And, a robust
and connectiv- ecosystem of open-source software libraries means that
ity—to improve
their quality and engineers can solve problems 100 times faster than just two
performance. decades ago.
Quality profes- This digital transformation is leading us toward connected
sionals possess intelligent automation: smart, hyperconnected agents
the skills required
to help their deployed in environments where humans and machines
organizations cooperate—and leverage data—to achieve shared goals.
succeed in the
fourth industrial This isn’t the world’s first industrial revolution. In fact, it is its
revolution. fourth—and the disruptive changes it will bring suggest we’ll
need a fresh perspective on quality to adapt to it.

24 QP October 2018 ❘ qualityprogress.com

qualityprogress.com ❘ October 2018 QP 25
 F E AT U R E
QUALITY 4.0

Quality 4.0: A fresh perspective Emerging trends in quality

Quality 4.0 is the name given to the pursuit of performance excellence Because the cost of enabling technol-
during these times of potentially disruptive digital transformation. It comes ogies (such as sophisticated sensors,
from “Industry 4.0”—a term coined at Hannover Fair in 2011 to describe the intelligent algorithms and the computing
fourth industrial revolution.1 power to leverage them) has decreased
That event emphasized the increasing intelligence and interconnect- so much over the past decade, organiza-
edness of smart manufacturing systems. It reflected on the newest tions can now begin making them part of
technological innovations, placing them in historical context and tracing their digital strategy.
the development of key technologies from the 1700s to the present. The process of digital transformation
During the first industrial revolution (late 1700s and early 1800s), innova- is revealing changes in how we perceive
tions in steam and water power made it possible for production facilities to customers and organizational bound-
scale up and expand potential production locations. Before then, manufac- aries. Organizations no longer will be
turing facilities had to be constructed along rivers so waterwheels could be defined solely by their employees and
used to generate power. business partners, but also by the cus-
By the late 1800s, the discovery of electricity and development of tomers who participate—without even
infrastructure enabled engineers to build machinery for mass production. explicitly being aware of their integral
Iron ore production increased, enabling machines themselves to be mass involvement—in ongoing dialogues that
produced. In the United States, the expansion of railways made it easier to shape the evolution of product lines and
obtain supplies and deliver finished goods. new services.
The widespread availability of New business models won’t neces-
reliable power sparked a renais- sarily rely on ownership, consumption,
The process of digital sance in computing. Toward the centralized production of products or
transformation is revealing end of World War II, digital com- centralized provision of services. The
changes in how we perceive puting started to emerge from its value-based approach will accentuate
customers and organizational analog roots. The third industrial the importance of trust, transparency
boundaries. revolution came at the end of the and security. New technologies, such
1960s with the invention of the as blockchain, as defined in the sidebar
programmable logic controller. “Quality 4.0 Tools,” will help us imple-
This made it possible to automate processes, such as filling and reloading ment and deploy systems to support
tanks, turning engines on and off, and controlling sequences of events those changes.
based on the state of the process and changing environmental conditions. Even though the term “Quality 4.0”
The World Economic Forum (WEF) has been keenly interested in these hadn’t been used yet, the quality impli-
changes. In 2015, it launched a digital transformation initiative to coordi- cations of the fourth industrial revolution
nate research that would help anticipate the effects of these changes on were first described as early as 2015 in the
business and society. WEF recognized that we’ve been actively experi- ASQ Future of Quality Report.2 The study
encing digital transformation since the emergence of digital computing in aimed to characterize the evolution of the
the 1950s: first with mainframes, then client-server computing and PCs, quality landscape over the next five to 10
followed by the advent of the internet and early e-commerce sites. years to prepare the quality community
Mobile devices and cloud computing led to a convergence of services, for the challenges of the future.
as multiple customer touch points (phone, fax, web and tablets) gradually The report’s authors described how
blended into the single view of the customer that most organizations now the health and viability of the entire
have. Just 20 years ago, organizations were barely able to link your phone calls industrial ecosystem would become
to customer service, emails and web form queries. Now, it’s taken for granted. everyone’s concern. They anticipated the
The first industrial revolution was characterized by steam-powered emergence of several new perspectives,
machines, and the second by electricity and assembly lines. Innovations including:
in computing and industrial automation defined the third. The fourth ++ A shift of emphasis in the quality pro-
industrial revolution brings us machine intelligence, pervasive computing, fession from efficiency, effectiveness
affordable storage and robust connectivity. How can we leverage them to and satisfaction to continuous learning
improve quality and performance? and adaptation.
++ Changing boundaries in and between

26 QP October 2018 ❘ qualityprogress.com

 Quality 4.0 Tools
organizations, and how information is THE FOURTH INDUSTRIAL REVOLUTION
shared between different areas, due to BROUGHT ABOUT NEW TOOLS AND
information availability and transpar- TECHNOLOGIES TO IMPROVE QUALITY.
ency (“shifting seams and transitions”).
++ Supply chain omniscience and visibility ++ Artificial intelligence: computer vision, language processing,
chatbots, personal assistants, navigation, robotics, making com-
into production processes (being able
plex decisions.
to monitor and respond to any element ++ Big data: infrastructure (such as MapReduce, Hadoop, Hive
in real time). and NoSQL databases), easier access to data sources, tools for
++ An increased emphasis on customer managing and analyzing large data sets without having to use
experience, participative markets (in supercomputers.
++ Blockchain: increasing transparency and auditability of trans-
which customers consume and produce
actions (for assets and information), monitoring conditions so
energy) and prosumerism (in which transactions don’t occur unless quality objectives are met.
customers participate in the design and ++ Deep learning: image classification, complex pattern recognition,
development of the products they want). time series forecasting, text generation, creating sound and art,
Each of these shifts must be addressed creating fictitious video from real video, adjusting images based
on heuristics (make a frowning person in a photo appear to smile,
in Quality 4.0. As a result, the way we
for example).
approach and tackle problems will ++ Enabling technologies: affordable sensors and actuators, cloud
evolve. For example, we must learn how computing, open-source software, AR, mixed reality, VR, data
to manage data over the lifetime of the streaming (such as Kafka and Storm), 5G networks, IPv6, IoT.
data rather than that of the organization ++ Machine learning: text analysis, recommendation systems, email
spam filters, fraud detection, classifying objects into groups,
that collects it. Our concept of voice of
forecasting.
the customer will expand, and we’ll find ++ Data science: the practice of bringing together heterogeneous
ways to listen to the voice of things as data sets for making predictions, performing classifications,
well because we’ll be able to learn about finding patterns in large data sets, reducing large sets of obser-
our customers from the connected vations to most significant predictors, applying sound traditional
techniques (such as visualization, inference and simulation) to
objects around them.3
generate viable models and solutions.

Why now?
Although the growth and expansion of
the internet accelerated innovation in the
late 1990s and 2000s, only now are we In addition, improved network infrastructure is expanding the extent
poised for the fourth industrial revolu- of connectivity, making it more widely available and robust. And unlike
tion. What’s changing? the 1980s and 1990s, there are far fewer communications protocols that
++ Production and availability of are commonly encountered, so it’s a lot easier to get one device to talk
information: More information is to another device on your network.
available because people and devices ++ Intelligent processing: Affordable computing capabilities (and process-
are producing it at greater rates than ing power) are available to analyze and interpret that information so it
ever before. Falling costs of enabling can be incorporated into decision making. High-performance software
technologies, such as sensors and libraries for advanced processing and visualization of data are easy to
actuators, are catalyzing innovation in find and, in many cases, easy to use. In the past, for example, software
these areas. developers had to write their own code for even common tasks. Now,
++ Connectivity: First and foremost, the they can use open-source solutions that are battle tested by many.
introduction of IPv6—which defines ++ New modes of interaction: The ways in which we acquire and interact
how data are sent from one computer with information also are changing. In particular, new interfaces, such
to another—has ensured that there as AR and VR, expand possibilities for training and navigating a hybrid
will be enough addresses to locate the physical-digital environment with greater ease.
billions of devices that are expected to ++ New modes of production: 3-D printing, nanotechnology and gene
connect to the internet. The informa- editing are poised to change the nature and means of production in
tion produced by these devices will be several industries. Technologies for augmenting or enhancing human
instantly accessible over the internet. performance (exoskeletons, brain-computer interfaces and even

qualityprogress.com ❘ October 2018 QP 27

 F E AT U R E
QUALITY 4.0

autonomous vehicles, for example) also will open new mechanisms for
innovation in production and distribution. New technologies, such as Predictive maintenance can
blockchain, have the potential to change the nature of production as help you anticipate equipment
well by challenging ingrained centralized perceptions of trust, control, failures and proactively reduce
consensus and value creation. downtime. Quality 4.0 initiatives
can help you assess supply
Discovery: The new role of quality chain risk on an ongoing basis,
What we recognize as today’s quality profession began during the middle or help you decide whether to
of the second industrial revolution, with the methods of scientific manage- take corrective action.
ment introduced by Henri Fayol in France and Frederick Winslow Taylor in
the United States. Factories needed methods to ensure assembly lines ran
smoothly so they produced artifacts to specification. Workers could now
know how to engage in the production process and costs were controlled.
As industrial production matured, those methods grew to encompass 3. Improve transparency, traceability and
the design of processes built to produce to specification. In the 1980s and auditability.
1990s, the adoption of personal computing once again changed the land- 4. Anticipate changes, reveal biases
scape. Organizations regrouped their quality efforts around the value of and adapt to new circumstances and
culture and active engagement in quality—and total quality management knowledge.
(TQM), lean and Six Sigma gained in popularity. 5. Evolve relationships, organizational
As connected, intelligent and automated systems are more widely boundaries and concept of trust to
adopted, we can once again expect a renaissance in quality tools and reveal opportunities for continuous
methods. The progression can be summarized through four themes: improvement and new business models.
1. Quality as inspection: In the early days, quality assurance relied 6. Learn how to learn by cultivating self
on inspecting bad quality out of the total items produced. Walter A. awareness and other awareness as
Shewhart’s methods for statistical process control helped operators skills.
determine whether variation was due to random or special causes. Quality 4.0 initiatives might help
2. Quality as design: Inspired by W. Edwards Deming’s recommendation you add intelligence to monitoring and
to cease dependence on inspection, more holistic methods emerged for managing operations, or enable remote
designing quality into processes to prevent quality problems before they monitoring to improve the productivity
occurred. or morale of your operators. Think about
3. Quality as empowerment: TQM and Six Sigma advocate a holistic how to add to human capabilities rather
approach to quality, making it everyone’s responsibility and empowering than how to replace people in your pro-
individuals to contribute to continuous improvement. cesses. According to Finnish sociologist
4. Quality as discovery: In an adaptive, intelligent environment, quality Esko Kilpi, “The real future of work is not
depends on how quickly we can discover and aggregate new data sources, in the industrial model of pursuit automa-
how effectively we can discover root causes and how well we can discover tion but in the post-industrial model of
new insights about ourselves, our products and our organizations. promoting augmentation.”4
Predictive maintenance can help
Quality 4.0 value propositions you anticipate equipment failures and
How can Quality 4.0 help your organization? Specifically, how can you proactively reduce downtime. Quality
improve the performance of your people, projects and products by imple- 4.0 initiatives can help you assess supply
menting enabling technologies such as AI, ML, robotic process automation chain risk on an ongoing basis, or help
and blockchain? you decide whether to take corrective
New technology always should be introduced with a clear articulation action. Quality 4.0 initiatives also can
of its desired benefits. A value proposition is a statement that explains help you improve cybersecurity: docu-
what benefits a product or activity will deliver and, sometimes, how it will menting and benchmarking processes
happen. Value propositions for Quality 4.0 initiatives fall into six categories, can help your organization detect
with No. 1 being the most significant: anomalies and understand expected
1. Augment (or improve upon) human intelligence. performance to more effectively flag
2. Increase the speed and quality of decision making. potential attacks.

28 QP October 2018 ❘ qualityprogress.com

FIGURE 1 one kind of ML algorithm, and deep learning is a special kind of complex

The ecosystem of
neural network that incorporates layers with special functions.
AI and ML are becoming popular now because not only is the software
more accessible and easier to apply, but it’s also easier to access the big

Quality 4.0 tools data that makes AI and ML so powerful.

Blockchain—a newly emerging technology—has the potential to improve
data quality and the quality of transactions. Statistics and data science
Artificial intelligence provide the firm foundations that should be applied to all problem solving.
Machine learning
Quality professionals:
Neural networks Leading the transformation
Deep learning
The introduction of AI and ML means that data-driven decision-making
can become more self aware. With better information, we’ll be better able
to adapt to changing environments and changing customer or stakeholder
Blockchain needs.
Big data Quality professionals are perfectly positioned to lead digital transforma-
Enabling technologies tion efforts because we have deep skills in:
Statistics and data science ++ Systems thinking.
++ Data-driven decision making.
++ Leadership for organizational learning.
++ Establishing processes for continuous improvement.
++ Understanding how decisions affect people: lives, relationships, commu-
The ecosystem of enabling nities, well-being, health and society in general.
technologies: Adding value This last skill is particularly important. Many ML algorithms must be
Automation isn’t an all-or-nothing pros- trained, and training is subject to personal and cognitive biases. Qual-
pect. A user can create a process that a ity professionals can anticipate positive and negative effects, helping
computer or intelligent agent executes, organizations protect against negative consequences while capturing
the computer can make decisions for an opportunities that will benefit everyone.
operator to approve or adjust, or the com- Quality professionals are distinctively good at structured problem
puter can make and execute all decisions. solving, data-driven decision making and leveraging cultural change to
Similarly, machine intelligence is a facilitate improvement. In Quality 4.0, these fundamentals will not change,
spectrum: An algorithm can provide even as the amount and variety of data increase. As a community, we are
advice, take action with approvals or uniquely positioned to help our organizations thrive in this new era. 
adjustments, or take action on its own.
REFERENCES
We must decide what value we want 1. Henning Kagermann, Wolf-Dieter Lukas and Wolfgang Wahlster, “Industrie 4.0:
to generate when we introduce various Mit dem Internet der Dinge auf dem Weg zur 4. Industriellen Revolution,” VDI
degrees of intelligence and automation Nachrichten, Vol. 13, No. 11, 2011, https://tinyurl.com/ly6vkgf.
2. ASQ, “2015 Future of Quality Report: Quality Throughout,” 2015, https://tinyurl.com/
into our processes. p22skhb.
To do this, we must understand the 3. Laurence Goasduff, “How to Listen to the Voice of ‘Things’ in the IoT,” Gartner, Dec. 20,
techniques and technologies of Quality 2017, https://tinyurl.com/y9hvfcyu.
4. Esko Kilpi, Twitter post, March 28, 2018, https://twitter.com/EskoKilpi.
4.0, as defined in the sidebar “Quality
4.0 Tools,” and how they relate to one
another. The relationships among these
areas are illustrated in Figure 1.
AI encompasses the most tools, which,
in many cases, can become so ordinary Nicole Radziwill is quality practice lead at Intelex
that they’re no longer considered AI, Technologies in Toronto and editor of Software
such as optical character recognition. ML Quality Professional. She is an ASQ fellow and an
ASQ-certified Six Sigma Black Belt. Radziwill is
algorithms make up some (but not all) one of ASQ’s Influential Voices and blogs at http://
of the domain of AI. Neural networks are qualityandinnovation.com.

qualityprogress.com ❘ October 2018 QP 29

F E AT U R E
MACHINE LEARNING

Just the Facts

Machine learning—the ability of artificial

intelligence to find patterns in data and
learn without being programmed—
is poised to become more widespread in
the near future.

The power of machine learning also can

advance quality improvement and qual-
ity maintenance activities—especially in
design and production—to find new and
innovative solutions.

All organizations—even cautious

ones—should explore machine learn-
ing options and prepare for possible
implementation.

30 QP October 2018 ❘ qualityprogress.com

REVOLUTIONARY
IMPACT?How machine learning could transform quality improvement 
| by Johannes Freiesleben

Machine learning, sometimes referred to as deep learning, is gaining more attention these days.
As a part of the field of artificial intelligence (AI), machine learning deals with algorithms that make
computers think—that is, make sense of huge amounts of unstructured data by themselves.
Similar to the human brain, these machines are trying to find patterns in data. Or, in statistical
terms, they perform a principal component analysis to distinguish between factors that can help
explain a phenomenon and those that cannot. Machine learning is already an integral part of many
applications used every day, from Facebook’s facial recognition algorithms identifying untagged
faces to Google’s search algorithms refining the user’s search by looking at patterns in that individu-
al’s previous searches.
Machine learning will only become more ubiquitous in the near future. It is, in fact, set to revolu-
tionize the way we live. Tesla’s recent advances with self-driving cars are a perfect example. These
cars automatically guide the passenger through complex and sometimes chaotic traffic situations
by interpreting large amounts of data while driving, enabling the cars to react swiftly to unforeseen
events that appear suddenly. The cars are equipped with hundreds of sensors that feed the com-
puter. Machine learning algorithms enable the cars to learn from the data and get better at handling
complicated situations and predicting their likelihood of occurrence.
Look to the future, and you will not only encounter self-driving cars, buses, trains or planes, but
a wide array of other possible uses of machine learning. Some examples include market research,
strategic controlling, medical analyses and legal services, to name just a few. Each will be concerned
with working through large amounts of data to find the right response.
Will machine learning also affect industrial production? And, most important to us as quality pro-
fessionals, will it improve and maintain design quality and production quality? This article addresses
the likely future impact of machine learning on these two quality dimensions.
Of course, this is a bit like reading tea leaves—the field of AI evolves quickly and with uncertain
outcomes. Nonetheless, we can deduce some of the principal fields of application for quality man-
agement solely from having a correct understanding of what machine learning is and where it can
help us in our field of investigation.

qualityprogress.com ❘ October 2018 QP 31

 F E AT U R E
MACHINE LEARNING

Knowing what machine learning is and how it works, and identi- FIGURE 1

Traditional programming
fying areas in quality management where its application will most
likely have a potentially revolutionary impact, can provide a helpful

vs. machine learning

guide for where practitioners should aim their efforts.

Intelligent machines
Since the digital revolution took off in earnest in about 1995, com- Traditional programming
puters have become ubiquitous aides in many daily endeavors.
Traditionally, a computer was viewed as a systematic processing Input (data)
machine, transforming input data into output by using algorithms Computer Output (data)
(that is, programs), which had been designed for this purpose by
Algorithm
humans. Given the data are fed to the computer in a consumable
format, its processing power yields predictable results effectively and
efficiently, vastly outperforming its human counterparts.
Machine learning
With the advent of machine learning, this picture is changing
(see Figure 1). Instead of feeding the computer data and programs Input (data)
and expecting output, the computer is now fed input data as well
Computer Algorithm
as output data to discover a systematic link between them—that
is, a functional algorithm. Machine learning is thus applicable Output (data)
whenever:
++ There are patterns between input and output data—but we
don’t know them. enriches the database and serves as a verifier to the
++ We cannot model the pattern mathematically—otherwise, initial insights. The learning process enables the com-
we would just resort to the familiar method of algorithm puter to progress from individual examples to broader
programming. generalizations—that is, to perform inductive reasoning.
++ We have a large amount of unlabeled or uncategorized data. The incorporation of prior knowledge in the form of
For almost any system collecting and analyzing real data, at training sets—implicating a certain bias of the learning
least one of these three points is relevant: process—is often deemed inevitable for the success
1. Knowledge might not always be explicit, as in the case of routine of learning algorithms by preventing the process from
tasks, such as driving, that cannot easily be broken down into reaching senseless conclusions.2
simple programs. What we described so far is supervised learning—
2. Traditional programming produces rigid codes, which do not that is, categorizations of data are given (for example,
adapt to changes over time or to different users. This is something hair length in the facial recognition example) and a
machine learning achieves by adapting to changing input data, training set is provided so the algorithm does not need
such as filtering spam emails and keeping up with constantly to start from zero.
evolving spam techniques. However, a result similar to what is achieved with
3. Large amounts of dynamic data cannot be processed by supervised learning can be done with unsupervised
human-written code because they are simply too complex, as it learning. Here, the algorithm makes sense of all data
is with weather forecasting. purely by itself. No categories or other classifications
The main idea of machine learning is enabling the computer are given, and there is no distinction between train-
to learn by itself, similar to how humans acquire knowledge. In a ing and test data. In July 2017, for example, Google
famous definition, Herbert A. Simon, an American economist and researchers programmed an unsupervised learning
political scientist, pointed out that, “Learning denotes changes algorithm so it could make a humanoid machine
in a system that ... enable a system to do the same task … more advance through difficult territory—one of the most
efficiently the next time.”1 This learning happens without previ- challenging tasks to teach a humanoid because there
ous expertise or access to it. Similar to a human without previous are seemingly unlimited variables involved—solely by
expertise, the computer learns from experience. setting an incentive to overcome obstacles.3
The initial data, which serve as a starting point for the learning Being confronted with a varying set of obstacles,
process, are often referred to as “training data” or “training sets,” such as beams, walls or cliffs positioned in the way, the
and enables the computer to gain some first insight about the algorithm developed the movements needed to tra-
matter at hand. The constant comparison with new incoming data verse the terrain most effectively and efficiently. None

32 QP October 2018 ❘ qualityprogress.com

of these movements had been programmed. Although the organization’s reach in terms of its ability to produce the
the example was conducted in digital space (showing a required product features) certainly will affect level two, where
representation of a running puppet in a virtual land- then most likely the wrong decision regarding the product features
scape), it highlights the enormous power embedded in is taken. But independently from level one, a wrong decision can
machine learning technology. be made on level two when it is not entirely clear to the organiza-
tion which features are expected by the target customers.
Intelligent quality management For example, consider the well-known failure of Volkswagen’s
Can this power also be applied to produce advances in luxury sedan Phaeton, which was aimed at the upper-class car
the fields of quality improvement and quality mainte- buyers segment. Although Volkswagen had all of the technolog-
nance? It’s a shared perception among most quality ical and managerial abilities to aim for that segment and indeed
professionals that achieving high quality is largely designed the Phaeton as a superlative sedan, Volkswagen missed
about learning from data. Hence, there is reason to the point that customers in that segment expected luxury brand-
believe that we can find interesting overlaps between ing as part of the expected product features. Hence, the Phaeton
the two disciplines. should have been sold under a new, luxury brand instead of the
Consider the two main dimensions of quality—design Volkswagen brand, which was considered a producer of reliable
quality and production quality—and where machine and well-designed mid-range cars, not luxury cars.
learning might be useful to find new and innovative Finally, on level three, aiming for the right decision means
solutions. staying constantly informed about potential changes in customer
Intelligent design quality: To start with the first requirements, competitors’ advances and product cycles because
decision a business is confronted with, the design feature compositions considered in vogue last year might have
quality dimension concerns identifying the customer become average this year. These changes aren’t as rapid in other
segment to target and the degree to which the product industries as they are in the IT industry. Nevertheless, it is vital
features match customer requirements in that segment. for all organizations to stay informed about big trends that shape
Excellent design quality is a perfect match of features customer requirements and hence the expectations regarding
to requirements, and poor design quality is a feature their products.
composition that leaves many target customers unsat- Machine learning can help on all three levels by making sense of
isfied with the product. large amounts of unstructured information. This is most obvious
Design quality improvement means improving the on level three when discussing trends. By their very nature, trends
match, and design quality maintenance means main- show early but often weak signals stemming from a variety of data
taining a good match permanently, despite potential sources. Usually they are quite confusing and difficult to grasp until
changes in customer requirements. Data play an import- they (sometimes suddenly) manifest themselves.
ant role in design quality on three sequential levels: The organization that understands trends first and can use them
1. Making the right decision regarding the customer to proactively adapt its product designs has a clear competitive
segment to target. This depends on having sufficient advantage. Trend spotting is very
information about the customer segmentation in the similar to weather forecasting—
market, as well as the technological and manage- and hence, it is a domain in which The main idea of
rial capabilities of the organization (or its ability to machine learning might provide machine learning
acquire them) to fulfill customer requirements in the the organization with vital analysis is enabling the
target segment. and the best forecast. computer to learn
2. Making the right decision on product features for the Using the same logic, the deci- by itself.
targeted customer segment, which depends on having sion about product features on
an in-depth understanding of what the customers in the level two—and even the decision of which customer segment to
target segment expect. target on level one—can greatly be advanced by machine learning.
3. Revising the decision on product features for the Markets are usually quite fuzzy, and it is not entirely clear where
targeted customer segment, which usually evolve the boundaries of segments unfold and what the precise customer
dynamically with time. This depends on having expectations are in randomly defined segments.
sufficient information about trends in the market and Machine learning can not only help define segments better and
potentially changing customer requirements. more precisely (the best method for this would be unsupervised
In reality, levels one and two are often interrelated. learning because it is completely free of any previous assumptions
Making a wrong decision on level one (for example, about the subject), but it also can better analyze the enormous data
aiming for a customer segment that is, in fact, beyond potentially applicable to the definition of customer requirements.

qualityprogress.com ❘ October 2018 QP 33

 F E AT U R E
MACHINE LEARNING

This allows for a better match and an organizations a great deal in achieving This triggers a problem investigation,
optimized feature composition. high design quality by analyzing a richer which usually involves two steps:
Especially in industries in which mass and more complex set of data from the ++ Step one: Finding the quality-
customization is important, machine external environment—whether it be data relevant process parameter (QRP) or
learning can guide organizations in their on customers, trends, competitors or the combination of QRPs that cause
decisions of what product parts to make potential employees—to achieve a much the problem.
customizable and in which form. To use closer match between target customer ++ Step two: Finding the root cause of
the renowned attribute classification of requirements and product features. the parameter deviations.
the Kano model,4 machine learning can Intelligent production quality: The The quality improvement decision
additionally provide an organization with discussion of how machine learning can constitutes the next step:
insights about delight attributes, which help in the other dimension of quality ++ Step three: Making the decision of
can help distinguish the product offer is no less interesting and potentially eliminating the root cause by either
from the competition. game-changing for many industries. optimizing the given technology or
Lastly, on level one, machine learning After the product design decision has adopting a better-suited technology.
also might apply to finding needed techno- been taken, production quality concerns After the defect-causing root cause is
logical or managerial capabilities when—for the question of how well the production cured and the production quality level
strategic reasons, for instance—a certain process is able to turn out products to improves, the follow-up investment is in
customer segment is targeted, but some specifications—in other words, how big quality maintenance. Given the knowl-
of the needed capabilities are unavailable the defect rate is. The lower the defect edge that has been acquired during the
in-house. Finding needed capabilities is no rate, the better the production quality— problem investigation, the logical form of
easy task because they are embedded in with zero defects (or close to zero, such quality maintenance is installing a mon-
human beings. Machine learning, however, as propagated in Six Sigma) meaning itoring system that tracks the identified
might make the search for gifted individ- perfect production quality. Data play an QRPs and indicates looming parameter
uals much more precise because it can important role here. degradations by early warning signals.
infer from a variety of input data (and not Using the quality investment cycle5 To guarantee a permanently high-
only the officially stated facts on résumés for an analytical structure, let’s begin quality level, monitoring must be
or curriculum vitaes) the likelihood that a with recognizing the quality problem—a installed for all QRPs. A main task to
person is able to provide these capabilities. defect rate usually detected by inspec- guarantee high maintenance effec-
In summary, machine learning can help tion or faulty products being returned. tiveness is thus to identify all potential

Machine Learning and Facial Recognition

An example of machine learning is the facial recognition Based on this training set, the algorithm develops a
algorithms used on social media platforms, such as prediction function that minimizes the prediction error
Facebook. Only machine learning can accomplish the on the training set—that is, it looks for the function best
recognition of unidentified faces. matching facial properties to names.
A traditional program would fail because what we In the second phase, the predictive function is tested
humans do with ease cannot easily be broken down into on a set of never-before-seen test examples of selected
simple programs (point No. 1, noted on p. 32), because the individuals from the training set. It should now be able to
algorithm must adapt to changing input data in the form recognize the test faces—that is, infer from the input (test
of different photographic representations of an individual faces) the output (name of the individual).
(point No. 2), and because the billions of unlabeled photos The goal of machine learning is to perfect the predictive
on the platform are a constantly growing set of complex function by refining it on as much data as possible. That
data (point No. 3). explains why large databases such as Facebook’s, in which
As shown in Sidebar Figure 1, the algorithm is the algorithm continuously compares facial properties of
first trained on a set of photos from individuals with individuals and cross-checks it on a vast amount of other
distinguished facial properties, such as hair length, eye data sets, cannot only correctly recognize your face but
color and facial expressions. The facial properties are the also suggest your lookalike or doppelganger celebrity.
inputs and the name of the individual is the output. —J.F.

34 QP October 2018 ❘ qualityprogress.com

QRPs. After the monitoring system has the quality level of the products. What In summary, machine learning can
identified parameters that increase in statistics did before—finding the QRPs in greatly help achieve high production
variability and exceed control limits, the vast number of process parameters—is quality because it increases the effec-
small counteractive investments can be now done by the algorithm. tiveness of problem investigation and
taken before defects are produced, and Whereas this application is straightfor- monitoring systems based on the
a maintenance investment cycle begins. ward, an application connected to step analysis of internal data on process
Therefore, we can add the two steps of three might be scanning diverse data parameters, which lead to a better identi-
quality maintenance as: from the external environment to help fication of QRPs.
++ Step four: Subjecting the known QRPs find the best improvement options for Additionally, it may aid in finding
to a monitoring system to achieve quality problems, similar to finding the better responses to quality problems by
quality control in these QRPs. best match of product features to cus- analyzing external data on technological
++ Step five: Eliminating any deviations in tomer requirements in the design quality advances and checking them for applica-
the monitored QRPs before defects are dimension. bility in the targeted improvement.
produced. Machine learning also can aid in step
Machine learning can aid production four in production quality maintenance. Outlook
quality improvement in various ways. In terms of monitoring effectiveness, the We can think of machine learning as
Its power of making sense of large more QRPs that are known and under being an enabler of what the industrial
volumes of data can directly be applied control (that is, monitored), the more world currently talks about under the
to the problem investigation outlined in effective the maintenance system. Even umbrella term of Industry 4.0—the
steps one and two—that is, finding the without the occurrence of quality prob- automation and data exchange trend
defect-causing QRPs and detecting the lems, machine learning can help detect in industrial production spanning
root causes of the problem. QRPs, which can proactively be sub- cyber-physical systems, the internet
The classical statistical method available jected to monitoring. In fact, we can think and AI. In this context, some flashy new
for this detective’s work—for instance, of machine learning as a logical comple- terms appear, such as predictive main-
fractional factorial design of experiments— ment to a monitoring system because it tenance6—which essentially is nothing
might be replaced by machine learning. can provide the right insights of where more than the combination of monitoring
The computer can be fed a variety of to allocate sensors and how to interpret and machine learning as outlined earlier.
input data from all process parameters variability in sensor data on the basis of And maintenance, as practiced before
that are tracked and the output data from the then-gained monitoring data. the advent of machine learning, was also
predictive in nature.
Therefore, it’s not surprising that
SIDEBAR FIGURE 1 machine learning is hotly debated and

Training set, predictive function and prediction sometimes dismissed as just hype.
However, there is a compelling logic
transcending the hype: its great power
Training
Training images
of making sense of massive amounts of
Training data, which can help organizations in
labels (names)
their quality management endeavors.
Of course, we should not neglect the
details of implementation, which can be
Facial properties Training challenging. Data from the outside are
not always in a consumable format. There
are security issues to be resolved when
an organization connects its internal sys-
Predictive tems to the external environment. And
function
there is a need for capable AI specialists
Testing who understand where best to apply
Testing image machine learning algorithms and how to
Facial properties Prediction develop them further.
Most importantly, we don’t know what
the future holds. There might be other

qualityprogress.com ❘ October 2018 QP 35

 F E AT U R E
MACHINE LEARNING

challenges impeding the rise of machine this will turn out to be a good investment. 6. Relayr, “IoT Enables Predictive Maintenance,”
http://blog.relayr.io/iot-enables-predictive-
learning, such as limits to the growth in com- In the end, there is no reason not to be maintenance-1, 2017.
puting power or political decisions slowing prepared.  7. Johannes Freiesleben, “Can Six Sigma Claim
the further development of the field. to Be a Generic Strategy? Reassessing
REFERENCES the Competitive Implications of Quality
As with any new technology, early 1. Herbert A. Simon, “Why Should Machines Improvement,” International Journal of Six
adopters can exploit the opportunities Learn?” which appeared in R.S. Michalski, Sigma and Competitive Advantage, Vol. 3, No.
arising from the use of machine learning Jaime G. Carbonell and Toni M. Mitchell, eds., 3, 2007, pp. 248-265.
Machine Learning: An Artificial Intelligence
and gain a possibly decisive competitive Approach. Tioga Pub Co., 1983.
advantage. Machine learning applications 2. Shai Ben-David and Shai Shalev-Shwartz,
could furthermore reinforce the positive Understanding Machine Learning: From
Theory to Algorithms, Cambridge University Johannes
effect on the competitiveness a good Press, 2014. Freiesleben
quality management system already 3. DeepMind.com, “Producing Flexible Behaviors is a professor
entails.7 in Simulated Environments, https://deepmind. of business
com/blog/producing-flexible-behaviours- administration at
Even if organizations are cautious, the simulated-environments, 2017. the University of
way forward should be to start exploring 4. N. Kano, N. Seraku and F. Takahashi, Applied Sciences
machine learning options, determining “Attractive Quality and Must-Be Quality,” St. Gallen in
Journal of the Japanese Society for Quality Switzerland, and focuses on quality
whether there is a potential fit with the Control, Vol. 14, No. 2, 1984, pp. 147-156. economics and the application of machine
organization’s business procedures and 5. Johannes Freiesleben and Gabriele Schwarz, learning in production. He holds a
objectives, and preparing for possible “Quality-Triggered Learning Effects,” Total doctorate in business management from
Quality Management & Business Excellence, the University of St. Gallen in Switzerland.
implementation. With some likelihood, Vol. 17, No. 7, 2006. pp. 825-834. He is a member of ASQ.

2018 INSPECTION DIVISION CONFERENCE

October 18 – 19, 2018 | Fogelman Executive Conference Center and Hotel | Memphis, TN

HIGHLIGHTS QUALITY: A BRIDGE TO THE FUTURE

• 16 conference sessions • 2 keynotes The 2018 ASQ Inspection Division Conference
will help individuals and organizations bridge
• 6 preconference courses • 1 fantastic networking reception! the gaps between remembering, understanding,
• 4 tours • Discounted hotel rooms and contributing. Experienced professionals
from a multitude of industries will serve as your
• 4 workshops • Multiple exhibitors
guides as they present relevant and value-added
information during interactive workshops and
Register today at asq.org/conferences/inspection-division. conference sessions.

36 QP October 2018 ❘ qualityprogress.com

F E AT U R E
LEADERSHIP

21 Century
st

Excellence
Updating the system of profound knowledge for
today’s service systems | by Bob Kennedy

In the 1980s, it was common for CEOs and quality profession-

als to attend three-day conferences hosted by W. Edwards
Deming. Many came in search of a magic bullet that would
transform their organizations into the competitive businesses
they’d once been. But they were in for a rude awakening.
Deming didn’t mince words when he scorned their quests for
“instant pudding”—the idea that an organization can solve its
problems with a simple quick fix.
Just the
Instead, he challenged them to adopt his system of profound
Facts
knowledge (SoPK), which is his “theory of related principles
that requires a leader or manager to consider all organizational Introduced in
aspects when making decisions.” 1 For Deming, the SoPK is the the 1980s, W.
Edwards Deming’s
method for transforming management and organizations. system of pro-
It’s timely to update the SoPK for the 21st century, when few found knowledge
(SoPK) provides
people can afford to attend conferences or read tomes on qual- a comprehen-
ity. Telling someone who is looking for a resolution to his or her sive framework
that leaders can
problem to apply the SoPK isn’t helpful and could be met with implement to
scorn and ridicule. transform man-
agement and their
Instead, we, as quality professionals, must develop models organizations.
and terminology suited to today’s service environment that cap-
ture the essence of the SoPK but allow us to engage with As the predomi-
nantely technical
people-based systems more productively. systems of the
past have changed
and evolved, the
SoPK’s models
and terminology
must be adapted
to today’s human-
based service
systems.

qualityprogress.com ❘ October 2018 QP 37

 F E AT U R E
LEADERSHIP

TA B L E   1

Development of SoPK
components over time
SoPK component Component’s development
Appreciation for a system Production viewed as a system, SIPOC, systems thinking
Knowledge of variation SPC, SQC, DoE, Six Sigma
Theory of knowledge PDSA, PDCA
Psychology Teamwork, quality circles, change management, transformation

DoE = design of experiments SIPOC = suppliers, inputs, process, outputs, customers

PDCA = plan, do, check, act SoPK = system of profound knowledge
PDSA = plan, do, study, act SQC = statistical quality control

Deming’s contribution purposeful activity. Think of an HAS as any individual or

Deming is one of the best-known quality gurus. After group engaged in some activity for its own benefit and
his rediscovery in the 1980s, his legacy likely is his the benefit of others.
SoPK. This is comprised of a model and a theory or set HASs appreciate the importance of a quality outcome,
of principles. The model became known as the “pro- which they readily accept is achieved by doing the right
duction viewed as a system model,” and the theory is things right.3, 4 This is a great basis for your initial engage-
comprised of four interrelated principles:2 ment, but these systems don’t respond well to intrusion
1. Appreciation for a system. Understand the com- or an overly technical approach to their issues, and may
plexities and interactions of the system you are resist your attempts to help them.
studying or working in. A good example of this resistance in the healthcare
2. Knowledge of variation. Variation has two causes, sector is outlined in the QP article, “Down With Silos,” in
common and special, which should never be which the authors “continue to struggle with finding the
confused. best way to fully engage physicians with respect to their
3. Theory of knowledge. Use a scientific approach to time and their undivided involvement.”5
testing theories. This is encapsulated in the plan-do-
study-act cycle. SoPK for HASs
4. Psychology. Treat people with dignity and respect. Few people would disagree with the statement that a
They are not cogs in a machine. quality outcome is achieved when the right things are
Table 1 shows how these components have devel- done right. That’s what a quality outcome is. How it is
oped over time. achieved is a different matter and the source of much
In a world that demands instant solutions to problems frustration, stress and confusion. This duality between
more than ever, it can be difficult to meaningfully engage the “what” and “how” of quality has bedevilled profes-
with those who would benefit from your quality profes- sionals, academics and society for decades.
sional skill set because most aren’t interested in applying At various stages in the evolution of quality, there
the SoPK, and many probably have never heard of it. have been glimpses of insight that offer workable, plau-
They, too, want a magic bullet that will easily solve their sible solutions for overcoming this duality in specific
problems. So, what can be done? circumstances. These glimpses include:
Deming’s SoPK must be condensed into terminology ++ Deming’s SoPK.
suited to today’s human activity systems (HAS). An HAS ++ ISO 9001—Quality management systems.
can be an individual, interest group, industry or institu- ++ The Malcolm Baldrige National Quality Award.
tion, and exists when one or more individuals engage in ++ The Shingo Prize.

38 QP October 2018 ❘ qualityprogress.com

 They provide a visual image and model, and a theory poses insurmountable challenges if you don’t appreci-
or set of principles to help organizations transform ate the seismic change that has occurred: The world has
themselves into sustainable entities that consistently moved from technology-centered hierarchical systems
deliver a quality outcome. that are manned by humans to human-based egalitar-
This article builds on Deming’s SoPK and develops ian service systems that merely use technology.
a theory and image that have universal applicability How successful an HAS is at creating and sustaining
in the 21st century service economy. It provides a deep a fit-for-purpose system or delivering a quality outcome
understanding of the HASs that sustain our economy depends on its understanding of and commitment to
and will help you engage productively with them. It also practice excellence. Excellence is all encompassing,
overcomes the duality of “what” and “how” by defining which is why it’s important to address excellence in
excellence as the universal aim for all systems. detail before proceeding with system analysis.
You should engage with an HAS through a simple,
non-challenging and concise message, which can be What is excellence?
encapsulated in the following sentence: “To achieve a Any definition of excellence must encapsulate the
quality outcome, we must have a system that is fit for last three components of the SoPK—knowledge of
purpose and ensure it practices excellence.” This will variation, theory of knowledge and psychology—and
lead to two questions: present them using terminology that is meaningful and
1. What is a fit-for-purpose HAS? relevant to the HASs of today.
2. What is excellence? Excellence is the path by which a system achieves
Guidance can be drawn from a five-year research a quality outcome and sustains itself. A system that
project that attempted to apply quality principles doesn’t embrace and practice excellence will fail or
across a broad range of sectors. Many research par- underperform. Therefore, you must fully understand,
ticipants initially had a negative perception of quality embrace and practice excellence yourself before
and saw it as getting in the way, being bureaucratic engaging with an HAS.
and slowing progress. Yet, all participants wanted their So, what is excellence?
endeavors to achieve a quality outcome. It is an evolving method through which a quality
The SoPK’s images and terminology are rooted in outcome is achieved. It’s based on voluntary, ongoing
production, so the HASs that predominate today have a dialogue and agreement among the creators, consum-
hard time applying them to their circumstances. HASs ers and complementors (legislators, regulators and
are open, complex systems that build relationships society, for example) in the system who define, realize,
internally and externally. They seek to become part deliver and evaluate dynamic emerging expectations in
of their customers’ lifestyle choices by expanding the an enlightened, effective, efficient, ethical, elegant and
services they offer. All of this occurs in an increasingly enjoyable manner.
regulated yet dynamic environment. Notice the definition includes the words “enjoyable”
The complexity of these relationship-based systems and “elegant.” These are important aspects of a quality
outcome. Excellence isn’t
just a core value of a sys-
TA B L E   2
tem’s culture, it’s the only

Capturing SoPK in the definition of excellence value that matters.

Table 2 shows how this
definition of excellence
SoPK component How it’s captured
captures various aspects
Appreciation for a system Not in the definition—human activity system is defined later of the SoPK.
Knowledge of variation Dynamic emerging expectations Excellence must
Theory of knowledge Define, realize, deliver and evaluate become the way the sys-
tem does everything. The
Psychology Voluntary ongoing dialogue and agreement between the
creators, consumers and complementors practice of excellence
should be the universal
Enlightened, effective, efficient, ethical, elegant and enjoyable
aim for all HASs as they
SoPK = system of profound knowledge pursue their purposes.

qualityprogress.com ❘ October 2018 QP 39

 F E AT U R E
LEADERSHIP

FIGURE 1 human attributes—free will, emotions, and positive and

Human activity system

negative capabilities—and aren’t amenable to techni-
cal solutions to their often complex, intangible human
issues.
fractal DNA They are governed by many intangible elements not
identified in Deming’s model, such as loyalty, commit-
ment and disinterest. This poses difficulties because the
Context SoPK doesn’t provide images or vocabulary that make
sense to the 21st century HASs and their problems. You
must be able to see these intangibles and discuss invis-
Clients ible interactions if you’re going to help a modern HAS
achieve its full potential.
Consequently, a new approach is needed to help you
re
tu

capture a shared appreciation of the system the HAS

l
Cu

Climate members are part of and to facilitate resolution of the

many issues they face.
An HAS that has a full appreciation for itself will
operate with a high degree of autonomy, harmony and
Excellence can’t exist in a vacuum; there must purposefulness. It understands its DNA and purpose, and
be a system to put it into practice. The practice of is sustained by its commitment to practice excellence.
excellence is the most difficult thing for any system
to sincerely embrace because, like all good things in A 21st century HAS
life, it is constantly evolving. But through practice, Figure 1 represents an HAS. Its shapes are of no signif-
the system will become better and faster at it. icance but are fuzzy to solicit support from the people
This new interpretation of Deming’s SoPK allows in the system, who often resent being engineered into
you to engage quickly and meaningfully with any neat little boxes. Its deceptively simple DNA is com-
HAS, as shown in the following scenarios: prised of a pattern of four interactive parts: clients,
Scenario one: How do we improve the customer culture, climate and context. This pattern will help you
experience? discuss, design and manage HASs using simple termi-
Solution one: Put a system in place that is fit for nology that is easily remembered, understood and less
purpose and ensure it practices excellence. likely to provoke a dismissive or hostile reaction from
Scenario two: How do we successfully outsource the HAS members.
a specific task? Client. The term “client” is used to capture Deming’s
Solution two: Put a system in place that is fit for assertion that the system serves internal and external
purpose and ensure it practices excellence. members who have a role in the system achieving its
Scenario three: How do I manage this healthcare purpose. ISO 9001:2015 describes them as providers,
facility? customers and relevant interested parties.
Solution three: Put a system in place that is fit for Clients fall into three categories: creators, consum-
purpose and ensure it practices excellence. ers and complementors, as outlined in Table 3 and
Although these scenarios are different, the solu- described in more detail later.
tions are identical. Culture. A culture is the shared beliefs and core values
that underpin all actions and interactions in the system.
Deming’s view of a system Culture can’t be imposed or owned by any one category
Deming provided a visual image of production of client­—it must be an agreed-on amalgam of all three
viewed as a system, which has morphed into today’s and accepted as the system’s modus operandi. Ethical
ubiquitous suppliers, inputs, process, outputs, cus- behavior is an example of a universal value. A commit-
tomers (SIPOC) model. SIPOC is perceived as being ment to practice excellence is another.
applicable to manufacturing environments, but most A lot has been written about the culture of quality,
of today’s HASs exist in service environments. They which really is a subculture. Overall culture is a power-
aren’t machines and display the full spectrum of ful, intangible part of any system. But it remains stale

40 QP October 2018 ❘ qualityprogress.com

rhetoric unless it is translated into something tangible. system as it exists. There are three steps to profiling an
Climate. Climate is tangible—it’s what we see, feel, HAS:
think and experience when we encounter the system as a Step one: Draw a frame and place the HAS pattern,
creator, consumer, complementor or outsider. It’s created shown in Figure 1, inside it. Briefly explain what each
by the HAS as its processes interact with its clients or element is to get some attention and buy-in. Previ-
others. It creates our perceptions of the HAS. ously, the frame was a geographical or physical system
Climate without culture is a deceptive smokescreen— boundary. In this instance, it is a timeframe: past, pres-
like a smile without friendliness. A climate that isn’t ent, future or some continuum. Time is the only system
linked to core values is blind to emerging issues. Take, for boundary and the ultimate constraint.
example, a core value of the Red Cross/Red Crescent, Step two: Agree with the clients in the room on a
which is unconditional aid and service. When the unex- clear definition of the system’s purpose. Deming used
pected happens, this manifests itself in the organization’s the term “aim,” but “purpose” is a better word when
rush to help everyone who is affected. It just knows how dealing with purposeful systems. Why does it exist?
to react. This is a direct link between an intangible cul- What service does it provide? A clear purpose is vital—
ture and a tangible climate that doesn’t require elaborate it’s the thread that holds the system’s DNA together. As
management or procedures to spring into action. a living HAS, the clients decide for themselves, through
Context. An HAS’s context is comprised of the voluntary agreement, what they do and how they do it.6
circumstances that prevail at a point in time and space. Step three: Capture an appreciation for each
It is frighteningly transitory and includes all available element of the system by thoroughly mapping the
resources, prevailing conditions, competitors and system’s DNA elements—clients, culture, climate and
assumptions, for example. context. Stick with the facts at this stage and treat the
The context that exists at the time the system is system as rigid and frozen in the present moment or
created is the initial baseline for clients, but they must be timeframe.
vigilant at monitoring changes, and agile to quickly adapt Clients. Consider the three categories of clients:
the system and its purpose to a new prevailing context. creators, consumers and complementors. Are they
Failure to do so is like failing to update your computer’s all represented in the system? Were they involved in
operating system. The system might become obsolete setting it up? It isn’t unusual for creators to think they
or corrupt and fail to provide the quality service it was know best and exclude buy-in from consumers and
designed to deliver. complementors. Determine whether each category of
client has the same expectations of the system and the
Applying the HAS pattern same appreciation for the system’s purpose.
With a clear picture of what an HAS pattern looks like There is no place for selfish competitiveness. Frus-
and a clear understanding of excellence, you can enthu- tration arises when some clients don’t appreciate that
siastically engage with any individual, interest group, they’re in the system voluntarily for their mutual ben-
industry or institution seeking a quality outcome. efit. Clients’ responsibilities for maintaining the system
Initially, the HAS pattern should be used to profile the are ongoing to ensure that its purpose and processes
remain relevant and in tune with the prevailing
context and their ever-changing needs.
Culture. Despite being intangible, culture
TA B L E   3
affects all of the system’s activities, so it must

Categories of clients be well understood. Is the culture one of inclu-

sion, humility and service, or superiority and
domination? What are the core values? Are they
Client category Who they are
agreed on, accepted and shared by all three
Those who operate the system’s processes and
Creators client groups?
create or provide the service or outcome.
As the term implies, core values are central
The ultimate end users of the system’s output or
Consumers to the HAS’s actions and decision-making. The
outcome.
Those who neither create nor consume, but who
commitment to practice excellence is a core
Complementors are legitimate clients of the system. They include value and universal aim for all successful sys-
financiers, regulators, society and neighbors. tems. The clients will need the help of quality

qualityprogress.com ❘ October 2018 QP 41

 F E AT U R E
LEADERSHIP

FIGURE 2

Appreciation for current and desired systems

Present system appreciation— Transformation Future system desired for
rigid and frozen pathway success—fluid and dynamic
Context Context

Clients Clients

Excellence
re

re
ltu

ltu
Cu

Cu
Climate Climate

Current purpose Desired purpose

professionals to appreciate and practice this. framework.9 Here again, you must engage with the three
Climate. This is the tangible aspect of an otherwise categories of clients and tease out their appreciation for
intangible culture. Do your best to capture descriptive the context that prevails, and the ability and agility of the
terms, such as professional, efficient, friendly, empa- system to engage, adapt and thrive in it.
thetic, sloppy, gruff and indifferent. What are your Helping a system understand itself requires the skills of a
impressions and perceptions of the system? What is quality professional. After profiling the system, you are in a
your emotional response when engaging with its pro- strong position to reach a fact-based, shared appreciation
cesses? Can you determine how the clients perceive it? for the current system and its behavior.
All systems exist to serve their clients. Therefore, it
might be useful to refer to the five SERVQUAL7 dimen- System transformation and obstacles
sions to capture client expectations and perceptions of If the system’s purpose is unclear or no longer relevant, or
service: if any of the elements (context, clients, culture or climate)
1. Reliability: The system’s ability to perform the prom- are deemed suboptimal, corrective transformation is
ised service dependably and accurately over time. needed. This can’t be imposed on the system. Instead,
2. Assurance: The creators’ knowledge and courtesy, you must engage with the clients, practice excellence and
and their ability to convey trust and confidence. repeat the profiling process to capture their desired future
3. Tangibles: The appearance of physical facilities, system DNA. Here, everything is fluid, dynamic and full of
equipment, personnel and communication materials. possibilities.
4. Empathy: The caring, individualized attention the After the desired future DNA and purpose have been
system provides to its consumers, complementors agreed upon, an HAS will naturally want to rush into action
and internal creators.
5. Responsiveness: Clients’
willingness to help creators, FIGURE 3

Obstacles to system transformation

consumers and complemen-
tors provide prompt service.
Mapping the climate helps
the system see itself as others The system boundary is the timeframe
see it. System transformation— System desired for success
System under analysis excellence
Context. Does the system
Communication

have all the resources it needs

Competition

Context Context
Complexity

Constraints
Challenge

to function properly? Is it
Conflicts

Clients Clients

relevant to prevailing trends in

re

re
tu

tu

society or technology? Context

l

l
Cu

Cu

Climate Climate

is now included in ISO 90018

and is similar to the organiza- Vague system purpose Confusion gap Clearly defined purpose
tional profile in the Baldrige (If not acknowledged)

42 QP October 2018 ❘ qualityprogress.com

FIGURE 4 should be embedded in the culture. It’s highlighted here

Obstacles to service realization

to remind systems to encourage, solicit and facilitate
ongoing challenge from all clients and outsiders.
Complexity. Author Adam Kahane describes three
The system boundary is the timeframe forms of complexity:
Transformation of 1. Dynamic: Depends on the closeness in time and
Transformed system ideas and concepts space between cause and effect.

realized, delivered
2. Generative: Depends on the level of uncertainty.
Communication

Quality service

and evaluated
Competition
Context
Complexity

Constraints
3. Social: Depends on clients’ shared assumptions,
Challenge

Clients Conflicts
values, rationales and objectives.12
re

All systems display some level of complexity, but

ltu
Cu

Climate

it should never cause unnecessary complications or

Confusion gap stress. The goal is to remove as much complexity as
Clearly defined purpose (If not acknowledged) possible.
Conflicts. These may arise or exist among clients,
between the current and desired system, or among the
to transform itself and deliver its service. You must system’s interactions with other systems.
resist this pressure and help the HAS proceed sensibly Competition. This includes obvious competitors in
given the timeframe available. Transformation should the field of activity, as well as competition for time and
happen in the shortest timeframe possible to avoid sys- other resources in the system itself.
tem obsolescence, but it can’t be rushed into recklessly. Constraints. Much has been written on the theory
System transformation, as shown in Figure 2, isn’t of constraints.13 To reduce it to a single question: “What
easy. It requires the practice of excellence and all the are the limiting factors on any part of the system?”
skills of a quality professional to overcome the many These can be overcome only by extending the time-
obstacles to transforming to the desired ephemeral frame or enhancing the system’s DNA.
state. As the system embraces the enlightened element None of the existing excellence models or standards
of excellence, it might decide to adopt lean, Six Sigma address these obstacles. But failure to recognize a
or another quality tool, technique or approach. What’s system’s obstacles to transformation and service real-
important is for the system to decide that for itself by ization will result in unnecessary stress on the clients.
practicing excellence. This in turn may result in disillusionment and sometimes
As the system undertakes its own transformation abandonment.
through either intervention or changing context, it must
again commit to practice excellence. It also must rec- Are we there yet?
ognize that obstacles exist and prepare for them. These Now imagine that the desired system is in place, and
obstacles (Figure 3) aren’t encountered in mechanical there is an agreed-on purpose and commitment to
systems. Confusion will occur if they aren’t acknowl- practice excellence. Clients will be eager to let the sys-
edged and systematically addressed by all clients. tem do what it’s supposed to do—namely, serve them
These obstacles include: by realizing a quality outcome.
Communication. Poor communication is the greatest They will be keen to once again rush into action and
obstacle to overcome and the biggest bar in the model. might see you as getting in the way. If their commit-
Excellence is predicated on dialogue. It can’t be based ment to practice excellence is sincere, everything will
on edicts, press releases or emails. Rather, it must be fine. But there are more obstacles to overcome.
engage participants and include reflective and genera- Fortunately, these obstacles are the same as those
tive dialogue.10 Clear, concise, effective and fact-based described earlier (Figure 4) and must be recognized
dialogue channels must be established. A system that and addressed.
fails to communicate will simply fail. Period. Like all quality approaches, this method is a con-
Challenge. As embraced by the Toyota way,11 this tinuum. Self-sustaining, autonomous, dynamic and
should not be viewed as hostile or adversarial. It is the harmonious systems must continually repeat these
system’s clients challenging it to deliver the best system transformation processes by practicing excellence
outcome in the timeframe available. This healthy trait (Figure 5, p. 44).

qualityprogress.com ❘ October 2018 QP 43

 F E AT U R E
LEADERSHIP

FIGURE 5

Self-sustaining system transformation continuum

Timeline

The arrows signify the practice of excellence

Universal aim REFERENCES

Systems exist to serve their clients by providing quality 1. John Schultz, “Out in Front,” Quality Progress, September 2013,
pp. 18-23.
outcomes. The SoPK must be condensed for the 21st cen- 2. W. Edwards Deming, The New Economics for Industry,
tury into a single sentence that is easily understood and Government and Education, Massachusetts Institute of
solicits buy-in from all branches of society: Technology, 1994.
3. Bob Kennedy, “Finding Harmony,” Quality Progress, November
“To achieve a quality outcome, we must have a 2014, pp. 17-20.
system that is fit for purpose and ensure it practices 4. George Labovitz, Y.S. Chang and Victor Rosansky, Making
excellence.” Quality Work: A Leadership Guide for the Results-Driven
Manager, Harper Business, 1993.
To be fit for purpose, a system must have a full 5. Linda Duncan and Sherri Luchs, “Down With Silos,” Quality
appreciation for its own DNA, an accurate understand- Progress, July 2017, pp. 22-29.
ing of its purpose and identify any obstacles if it is to be 6. William J. Latzko and David M. Saunders, Four Days With Dr.
Deming: A Strategy for Modern Methods of Management,
successful in the timeframe available. Excellence must Addison Wesley, 1995.
become the way the system does everything. System 7. Emel Kursunluoglu Yarimoglu, “A Review on Dimensions of
appreciation and commitment to practicing excellence Service Quality Models,” Journal of Marketing Management,
Vol. 2, No. 2, 2014, pp. 79-93.
are essential for harmony in any human endeavor. 8. International Organization for Standardization (ISO), ISO
Through agreement, HASs decide what to do and 9001:2015—Quality management systems—Requirements.
how to do it. The concept of excellence as a practice 9. “Baldrige Performance Excellence Program,” National Institute
of Standards and Technology, www.nist.gov/baldrige.
and core value of the culture recognizes this, and its 10. William Isaacs, Dialogue: The Art of Thinking Together, Double
practice increases the likelihood of success. Excellence Day, 1999.
addresses all of the stages of service: inclusion, agree- 11. Jeffrey K. Liker, The Toyota Way, McGraw Hill, 2004.
12. Adam Kahane, Solving Tough Problems: An Open Way of
ment, definition, realization, delivery and evaluation. It Talking, Listening and Creating New Realities, Berrett-Koehler
also guides clients’ actions, which should be: effective, Publishers, 2004.
efficient, ethical, elegant and—above all—enjoyable. It 13. Eliyahu M. Goldratt, Theory of Constraints, North River Press
Publishing, 1990.
encompasses the “what” and “how” duality of quality.
This approach to system appreciation, transformation,
quality outcomes and excellence isn’t intended to be
rigidly followed as a recipe for success. Rather, it’s an
ongoing, evolving method you can use to analyze and
create self-sustaining, autonomous, dynamic, harmoni- Bob Kennedy is a retired quality
ous systems. management lecturer. He has a
The universal aim for all systems should be a commit- doctorate in community excellence from
the University of Limerick in Ireland.
ment to practice excellence in all its endeavors, and the Kennedy is a senior member of ASQ and
role of the quality professional is to facilitate this.  counselor for members in Ireland.

44 QP October 2018 ❘ qualityprogress.com

Solving quality quandaries through statistics

Statistics Spotlight 1213231315447133415341545721157945495241262

124345464755497541275299949514654237204317
91545154134541213231315447133415341545721157945495241262
124345464755497541275299949514654237204317
91545154134541213231315447133415341545721157945495241262
124345464755497541275299949514654237204317
91545154134541213231315447133415341545721157945495241262
124345464755497541275299949514654237204317
91545154134541213231315447133415341545721157945495241262
124345464755497541275299949514654237204317
91545154134541213231315447133415341545721157945495241262
124345464755497541275299949514654237204317
9154515413454

PATTERNS

Straight Line
Or Not?
Extrapolating patterns beyond their natural are two-plus “down” days on the markets, the
show’s stories adopt the tone of “How should an
range can lead to false conclusions investor weather the oncoming storm?”
Not only are these gyrations in tone exhaust-
by Christine M. Anderson-Cook ing, but they also could induce overreactions
to small changes that would lead to poor
Until this spring, I had never heard of the book Factfulness.1 investment actions. The reporting suggests
Then Bill Gates made a big splash in the news by giving each U.S. that a trend of a couple of points in the same
graduate from college a copy.2 In reading the book, I found it to be direction suggests the pattern will continue on
a remarkable combination of hopefulness and commentary about that trajectory for an extended time, when in
some of our blindspots regarding numeracy, and how we interpret fact these chronologically local patterns often
and internalize information in a media-saturated world. are more a reflection of variability and volatility
If you have not read the book, I recommend it highly—not only than they are a trend.
for how it can help us think about the world in a more informed
and balanced way, but also how it analyzes classic mistakes that When is a straight line
we are all prone to. The book describes our instincts to erroneously appropriate?
divide things into groups when distributions are more accurate Beyond the urge to extrapolate based on a
(the gap instinct), misallocate attention to fear-triggering but low small amount of data, there are nuances to the
probability events (the fear instinct), plus eight more. But beyond straight line instinct that are worth exploring
that, Factfulness also presents helpful strategies to tame these more deeply. Factfulness argues that most rela-
debilitating instincts and prevent them from leading us toward tionships are not sensibly described by straight
false conclusions and perceptions. lines, but with functions that curve or include
In this column, I would like to focus on one of these 10 instincts: asymptotes. But this seems to lie in direct
the straight line instinct, which suggests that we often are tempted contradiction to the response surface method-
to extrapolate patterns with straight lines, when these may not be ology3 (RSM) approach of using straight lines
appropriate. Factfulness gives several examples in which incongru- and main effect models to characterize many
ous results are obtained, including one that projects the growth rate types of relationships.
of newborns unaltered through adulthood and reaches outlandish RSM relies on the principle that complex
heights. relationships often can be estimated with
The premise of the straight line instinct is that in the absence of Taylor series approximations—with a first-order
other information, it is natural to extend the pattern that we see with approximation corresponding to a straight line
the obvious straight line. Even in a less-formal way, these extrap- if a single response, Y, is being described by a
olations of trends surround us every day. Each weekday evening, single explanatory factor, X. As the number of
I watch a business show that summarizes stock market trends. If X’s used to describe the response increases, the
there are two-plus “up” days on the markets, the announcers’ stories approximation is defined by a straight plane
all have a jubilant “We are soaring to new heights” flavor. If there or hyperplane. But how can Factfulness and

qualityprogress.com ❘ October 2018 QP 45

Statistics Spotlight 1213231315447133415341545721157945495241262
124345464755497541275299949514654237204317
91545154134541213231315447133415341545721157945495241262
124345464755497541275299949514654237204317
91545154134541213231315447133415341545721157945495241262
124345464755497541275299949514654237204317
91545154134541213231315447133415341545721157945495241262

RSM both be reasonable? There are three key changes in wealth would lead to wild misses in our predictions.
differences to the scenarios considered: changes Hence an important question to ask when looking to generalize a
to the underlying relationships, human reaction relationship and make predictions for new observations is: Are the
and response to observing the patterns, and the underlying drivers of the relationship stable or are there changes
range over which a curve is being used. that might suggest different patterns are appropriate for different
First, Factfulness is almost exclusively subsets of data? Depending on what the answer is, our level of
focused on relationships between inputs and confidence in what we are predicting (particularly in an extrapolated
responses as they change over time, where region) should change.
prediction involves looking into the future. Second, when a system involving people and their actions is being
Response surface methods are predicated on a observed, the rules driving relationships may change in response to
consistent underlying mechanism for all of the the previous pattern. Consider the stock market example, in which
data. If we are modeling a scientific relationship, several days of gains
it is much more likely that the mechanisms driv- on the market might
ing the observed pattern are going to remain lead to investors
The premise of the straight line
consistent for all of the data, and most impor- taking profits, which
instinct is that in the absence of
tantly, for future data. might translate
other information, it is natural to
When we are looking into the future, the into a drop in stock
extend the pattern that we see
underlying rules on which new data will be prices. Relationships
with the obvious straight line.
based are likely to change. In Factfulness, the governed by science
authors consider predicting future population are unlikely to change
growth and note how changes in the affluence with observation. A ball dropped will fall at the same rate whether it
level of the world population drives changes is watched because the rules driving the pattern are not influenced
in people’s behavior. As more of the world by human response. However, when we examine changes in the
moves from subsistence, the average number of world over time—whether economic, global health or climate—the
children they have is reduced. Hence, projecting roles of humans often are changing the nature of these relationships
historical rates of population using a straight line through their intervention. It is critical that we take these impacts
going forward without adjusting for underlying into account as we think about future changes.

46 QP October 2018 ❘ qualityprogress.com

 Finally, there are big differences between a FIGURE 1

A straight line to approximate a

local approximation and a global one. A straight
line can be a good approximation to many
complicated relationships if we are looking at a
small range of input values. Where we can get
into trouble is extending that relationship for an
curved relationship locally
unrealistic range.
Figure 1 shows an example where the true
relationship is moderating in growth for larger

Response, Y
values of the explanatory variable. Several
black dashed lines have been added to show
that if we consider smaller ranges of X (shown a
bit offset below the blue line for easier view-
ing), the straight line approximation works well.
When we look across the entire range of X in Explanatory variable, X
the plot, however, a straight line does a terrible Note: A straight line can work well to approximate the (blue)
job of capturing the true pattern of change true curved relationship locally (black dashed lines), but may
poorly misrepresent the global relationship (red line).
and could lead to wild predictions if we tried
to extend beyond the range of observed X
values. RSM actively takes this into account as
models of different complexity are considered people: Look at a plot of the data; by examining a plot of data that
and compared for their ability to characterize generated the blue curve in Figure 1, it is easy to see that a straight
relationships. First order models, models with line is inadequate to summarize the entire relationship.
interactions and second order models pro- Next, it is beneficial to think about what underlying attributes
gressively allow more flexibility to capture the could be changing across the different observations. If we think
patterns that we see in our data. through what might be driving the relationship and see the big
Simple models can be helpful. They allow picture, our ability to select an appropriate model to guide our
us to summarize relationships to see patterns predictions and understanding is vastly improved.
quickly and represent them efficiently. Where So the authors of Factfulness had it right: Straight lines can
we can get into trouble is if the model that we be dangerous and misleading, particularly if the ground rules
use is inadequate to capture the true behavior on which the relationship is based are changing. Hopefully, this
of the system. There are several key ways we has helped clarify when we should be particularly concerned
might be led astray: about building our understanding on such a simplification of the
1. The data on which the model is built come observed pattern. 
from several different mechanisms, and
REFERENCES
grouping them together may disguise some 1. Hans Rosling, Anna R. Ronnlund and Ola Rosling, Factfulness: Ten Reasons
of the information we need to understand the We’re Wrong About the World—and Why Things Are Better Than You Think,
patterns we see. Flatiron Books, 2018.
2. Gates Notes on Factfulness, gatesnotes.com/Books/Factfulness.
2. Humans have a way of changing things, and 3. Raymond H. Myers, Douglas C. Montgomery and Christine M. Anderson-
the feedback from seeing or experiencing Cook, Response Surface Methodology: Process and Product Optimization
current patterns often leads to changes in Using Designed Experiments, Wiley-Interscience, 2016.
behavior that will fundamentally change
what future patterns will look like.
3. It is much easier to get a good fit with a
straight line to a local pattern in a relation- Christine M. Anderson-Cook is a research
scientist in the Statistical Sciences Group
ship, compared to looking across the entire at Los Alamos National Laboratory in Los
range of inputs. Alamos, NM. She earned a doctorate in
So how can we keep from being fooled? The statistics from the University of Waterloo
in Ontario, Canada. Anderson-Cook is a
first part of the answer is remarkably simple, fellow of ASQ and the American Statistical
but sometimes eludes even the most sensible of Association. She is the 2018 recipient of the ASQ Shewhart Medal.

qualityprogress.com ❘ October 2018 QP 47

Practical knowledge about Six Sigma

Six Sigma Solutions

SCORECARDS

The Improvement
Of Scorecard
Management
Comparing Deming’s red bead experiment to red-yellow-green scorecards

by Forrest W. Breyfogle III

W. Edwards Deming is well Deming’s red bead experiment

known for helping Japan When conducting the red bead experiment, Deming had audience recruits use a
improve product quality in the paddle to draw 50 beads from a container of red and white beads, as shown in the
1950s. Deming also is acknowl- illustration. White beads were considered satisfactory while red beads were not.
edged for the messages conveyed In this exercise, the number of red beads resulting from a paddle draw was
during his many four-day sem- reported. A count of red beads in a paddle drawn was to be instrumental in monitor-
inars. In these 1980s training ing a fictitious organization’s overall product quality. In this tracking, between-worker
sessions, Deming used a red bead quality of workmanship was highlighted in this exercise’s reporting, in which a lower
experiment to highlight several number of red beads in a paddle draw was better.
poor management practices, In Deming’s workshop experiment, he announced the management goal of no
which underscored aspects of his more than three red beads for each paddle draw. The poorest-performing opera-
14 points of management1 and tors (those drawing beads from the container) relative to not meeting this minimum
seven deadly diseases.2 red bead number objective were “let go” from Deming’s organization company. The

TA B L E   1

Number of red beads in 60 paddle draws

Paddle-draw number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Number of red beads 7 7 9 12 13 13 13 10 8 9 7 9 13 11 10 14 5 6 7 10 10 14 13 13 15 11 6 9 8 7
Paddle-draw number 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Number of red beads 10 5 7 11 9 8 7 10 12 10 5 8 8 9 14 7 5 7 10 10 10 5 15 13 8 7 4 9 11 8

48 QP October 2018 ❘ qualityprogress.com

organization was eventually shut down Online Figure 1, found on this column’s average, 10 red beads are expected from
because none of the operators met the webpage at www.qualityprogress.com, a 50-hole paddle draw. Rather than hav-
minimum number of allowable red beads illustrates an actual organizational RYG ing a goal of three or fewer for any draw
for a paddle draw. scorecard report-out. (as Deming did in his workshops), let’s set
From a management’s scorecard point the goal to be 12 or fewer red beads for
Red-yellow-green scorecards of view, a question that this column paddle draws.
Organizations often use red-yellow-green addresses is whether there are any basic Table 1 shows the number of red beads
(RYG) scorecards to track individual differences from a management point of in 60 example paddle draws. This data
performance metrics relative to goals. At view between Deming’s red bead experi- will be used to compare the management
regular times, an RYG scorecard colored ment and commonly used organizational techniques used in Deming’s red bead
response assesses how well a metric RYG scorecards. experiment and RYG scorecard reporting.
response performs relative to its goal: If there are no fundamental differences For this set of red bead draws, a Dem-
++ Green—satisfactory. between Deming’s workshop exercise ing red bead experiment management
++ Yellow—warning. and RYG scorecards, the same lessons style would be telling operators to do
++ Red—not satisfactory. learned from Deming’s experiment also better for the 12 paddle draws circled in
If a metric’s color is red, action must would apply to the use of RYG scorecards Table 2—that is, whenever the number of
be taken to resolve why performance in businesses. red beads was 13 or more.
at this point is unsatisfactory. With RYG Colors for possible metric draws for a
scorecards, when a green color follows Comparing methods RYG scorecard of Table 2 data are:
a red-colored metric, management using a data set ++ Green: 11 or less.
typically feels good because it believes In the red bead experiment, 20% of the ++ Yellow: 12.
that the red-issue problem was resolved. container’s beads are red. Hence, on ++ Red: 13 or greater.

TA B L E   2

Paddle draws with Deming’s red bead

management style operations feedback
Paddle-draw number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Number of red beads 7 7 9 12 13 13 13 10 8 9 7 9 13 11 10 14 5 6 7 10 10 14 13 13 15 11 6 9 8 7
Paddle-draw number 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Number of red beads 10 5 7 11 9 8 7 10 12 10 5 8 8 9 14 7 5 7 10 10 10 5 15 13 8 7 4 9 11 8

qualityprogress.com ❘ October 2018 QP 49

Six Sigma Solutions

TA B L E   3

Paddle draws with a red-yellow-green scorecard

management style operations feedback
Paddle-draw number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Number of red beads 7 7 9 12 13 13 13 10 8 9 7 9 13 11 10 14 5 6 7 10 10 14 13 13 15 11 6 9 8 7
Paddle-draw number 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Number of red beads 10 5 7 11 9 8 7 10 12 10 5 8 8 9 14 7 5 7 10 10 10 5 15 13 8 7 4 9 11 8

Table 3 shows the results of an RYG be the result of natural variation from not perform satisfactorily).
scorecard management metric reporting the process or a special cause event. If a process has an undesirable com-
style. The container of red and white beads mon cause response, the process must
With this RYG scorecard reporting had 20% red beads. For this simplistic be improved. Talking about individual
approach, management again would be situation, we know that the variability out-of-specification occurrences in isola-
reacting to the 12 paddle draws contain- from a paddle draw in this container tion as though they were a special cause
ing 13 or more red bead paddle draws. In should be common cause. In the red is an ineffective management practice.
addition, management would probably bead experiment and RYG situations, the
think that the red occurrence problem management style was to react to any Metric management
was resolved whenever the metric color paddle draw that contained more than 12 with control charts
changed to green on the next draw. red beads as though the situation were A traditional control charting approach
A process output response can either special cause (that is, the operator did separates common cause from special
cause events with the focus on identifying
special cause events so the issue can be
FIGURE 1 resolved in a timely manner. If we consid-

IEE 30,000-foot-level proportion ered the number of red beads on a paddle

draw as the response, we might select a

nonconforming report-out
c-chart for this common-special cause
event separation. Online Figure 2 shows a
c-chart for the data in Table 1.
IEE scorecard for data: proportion of red beads in the 50 holes But is the c-chart technically the most
0.35 appropriate chart for this number of red
UCL = 0.3231 beads in a paddle situation? Because
0.30 we know from this process that either a
0.25 white or red bead can be present in any
of the 50 holes, a p-chart that tracks
Rate

0.20 over time the proportion of red beads

X̄ = 0.187
0.15 relative to the number of 50-hole paddle
opportunities would technically be more
0.10 appropriate. A p-chart of the data from
0.05 LCL = 0.0509 Table 1 is shown in Online Figure 3.
1 7 13 19 25 31 37 43 49 55 Online Figures 2 and 3 indicate that the
Paddle draw sequence process is in-control or stable. No state-
ments about how a process is performing
The current process is predictable. IEE = Integrated Enterprise Excellence overall relative to specifications or goals
The estimated process nonconformance LCL = lower control limit
rate is 0.187. UCL = upper control limit should be made from a control chart.
X̄ = mean value However, it does appear that the process is

50 QP October 2018 ❘ qualityprogress.com

incapable of achieving the desired results. FIGURE 2
From this control chart analysis,
we also can conclude that the pro-
cess management styles of the red
IEE 30,000-foot-level count of
bead experiment and RYG scorecard
approaches were ineffective in the red beads report-out
achievement of the goal of 12 red beads
or fewer in a paddle draw. IEE scorecard for data: number of defects point of view
From this assessment, we also could 16 UCL = 16.16
conclude that there is no fundamental
14
difference between the red bead exper-
iment and RYG scorecard approaches 12
in that both management styles use an 10
Count

X̄ = 9.35
approach to management described as 8
managing of individual occurrences that
6
are not satisfactory relative to a goal.
4
An alternative approach 2
LCL = 2.54
The data in Table 1 will now be reported 1 7 13 19 25 31 37 43 49 55
Paddle draw sequence
out using an Integrated Enterprise Excel-
lence (IEE) 30,000-foot-level metric
The current process is predictable. IEE = Integrated Enterprise Excellence
reporting method.3-12 This form of metric The estimated average number of LCL = lower control limit
reporting consists of two steps: nonconformances is 9.35 per paddle-draw. UCL = upper control limit
X̄ = mean value
1. The process is assessed for stability
from a high-level (30,000-foot-level)
vantage point. greater than 50 red beads for a given nonconformance of 18.7%. If this common
2. If the process is stable, a predictive draw. This approach will mimic cause frequency of nonconformance
process capability statement is made what would be expected as an IEE estimate for paddle draws is unsatisfac-
and reported at the bottom of the 30,000-foot-level report-out if the tory, the process must be improved.
chart with easily understood wording. data were counts of defects from a Counting defects report-out: Figure 2
To illustrate the versatility of this process. shows an IEE 30,000-foot-level report-out
one-page process stability or capability 3. The number of red beads represents when the data are considered counts of
reporting method, Table 1’s data will be a continuous response in which the defects—that is, the number of red beads
reported using the following approaches: goal is to have 12 or fewer red beads in a paddle. From this individuals chart, you
1. Percentage nonconformance in a for every paddle draw—A response would consider this process stable with
paddle draw—With this method, is unsatisfactory if a draw response an estimated average of counts of defects
you understand that each hole in the is greater than 12. This approach will to be 9.35 for individual paddle draws
50-hole paddle can contain either a mimic what would be expected as a from the container. If this common cause
red or white bead after a paddle draw. 30,000-foot-level report-out if the frequency of nonconformance estimate for
This reporting approach will mimic data were examined as a continu- paddle draws is unsatisfactory, the process
what would be expected as an IEE ous response—that is, from another must be improved.
30,000-foot-level report-out when process in which the data were truly a Continuous response report-out:
the data are analyzed as percentage continuous response. Online Figure 4 shows an IEE 30,000-foot-
defective rate. Percentage nonconformance level report-out if the data from Table 1 are
2. Count of the number of red beads report-out: Figure 1 shows an IEE considered continuous. From the indi-
in a paddle draw—Examine the data 30,000-foot-level report-out when the viduals chart on the left side of the figure,
purely from a number of red beads data are considered a nonconformance you would consider this process stable.
in the paddle point of view, in which rate. From this individuals chart of the The probability plot on the right is used to
we will consider in this analysis that, mean, you would consider this pro- determine process capability/performance
technically, there could be a number cess stable with an estimated rate of from a predictability point of view. For an

qualityprogress.com ❘ October 2018 QP 51

Six Sigma Solutions

upper specification of 12, you would expect on how well a process is performing in
that about 17.276% of the time the paddle which common and special cause events
draw values would be above 12. If this com- are separated.
mon cause frequency of nonconformance This process behavioral characteriza-
estimate for paddle draws is unsatisfactory, tion separation is important because the
the process must be improved. actions for special cause can involve the
Deming’s red bead experiment
assessment of individual or group events,
illustrates how attempting to
IEE 30,000-foot-level manage operators and process
while an unsatisfactory common cause
reporting outputs through individual
response suggests that process enhance-
Critiquing individual responses—like ments are needed.
reported values is ineffective.
that done in the red bead experiment With 30,000-foot-level reporting,
and with RYG scorecards—is ineffec- x-bar and R, p-charts, and c-charts
tive management. From this practice, are not used to make a process sta-
treating common cause variability as A 30,000-foot-level report uses an bility assessment. The referenced
though it were special cause can lead individuals chart when making a process 30,000-foot-level articles show math-
to much firefighting, which can be stability assessment. No timely pro- ematically why the individuals chart
costly, frustrating and detrimental to an cess control efforts are attempted with is appropriate for 30,000-foot-level
organization. 30,000-foot-level reporting. This form reporting and other control charting
IEE 30,000-foot-level charting not of performance reporting provides, from techniques are not for this high-level
only assesses process stability from a high-level viewpoint, an assessment form of performance reporting. 
a high process vantage point, it also
provides a predictive statement when REFERENCES 30,000-Foot-Level,” Quality Diabetes Measurement
the process is stable. If this predictive 1. Forrest W. Breyfogle III, Progress, November 2006, Tracking at the
statement is unsatisfactory, this process Integrated Enterprise pp. 59-62. 30,000-Foot-Level,” Quality
Excellence Volume II— 7. Forrest W. Breyfogle III, “No Progress, January 2017, pp.
measurement response pulls for the Business Deployment: A Specification? No Problem,” 50-53.
creation and timely completion of a pro- Leaders’ Guide for Going Quality Progress, November 12. Forrest W. Breyfogle III,
cess improvement project to enhance Beyond Lean Six Sigma and 2012, pp. 58-61. Integrated Enterprise
the Balanced Scorecard, 8. Forrest W. Breyfogle, Excellence Volume III—
the performance metric’s response. Citius Publishing, 2008. “30,000-Foot-Level Improvement Project
2. Ibid. Performance Metric Execution: A Management
Control charts are for 3. Forrest W. Breyfogle III, Reporting,” Six Sigma Forum and Black Belt Guide for
“Control Charting at the Magazine, February 2014, Going Beyond Lean Six
controlling processes 30,000-Foot-Level,” Quality pp. 18-32. Sigma and the Balanced
Deming’s red bead experiment illustrates Progress, November 2003, 9. Forrest W. Breyfogle III, Scorecard, Citius Publishing,
how attempting to manage operators pp. 67-70. “Integrating Inputs: A 2008.
4. Forrest W. Breyfogle III,” System to Capture and
and process outputs through individual Control Charting at the React to VOC Data Can Pay NOTE
reported values is ineffective, which can 30,000-Foot-Level, Part 2,” Dividends,” Quality Progress, A no-charge Minitab add-in is
extend to the use of RYG scorecards. Quality Progress, November January 2011, pp. 64-66. available for the easy creation
2004, pp. 85-87. 10. Forrest W. Breyfogle III, of IEE 30,000-foot-level charts
Control charts are used to control pro- 5. Forrest W. Breyfogle III, “High Vantage Point Report- at www.smartersolutions.com
cesses. The identification of special cause “Control Charting at the outs to Reduce Risks of /30000-foot-level-add.
events is to trigger process resolution 30,000-Foot-Level, Part 3,” Organizational Problems,” Organizations can benefit
Quality Progress, November Quality Progress, December when 30,000-foot-level
activities. Control charts are not to make 2005, pp. 66-70. 2015, pp. 58-60. reporting is integrated with the
any statement as to how a process is 6. Forrest W. Breyfogle III, 11. Forrest W. Breyfogle III, processes that created them
performing. “Control Charting at the “Monitor and Manage: using an IEE value chain.
IEE 30,000-foot-level report-outs
provide, from a high-level point of view,
a statement about whether a process
Forrest Breyfogle III is CEO of Smarter Solutions
is stable. In addition, if a process is Inc. in Austin, TX, and holds a master of science
stable and its 30,000-foot-level pro- degree in mechanical engineering from the
cess capability/performance statement University of Texas in Austin. An ASQ fellow,
Breyfogle has authored or co-authored more
is unsatisfactory, the process needs than a dozen business management and process
improvement. improvement books.

52 QP October 2018 ❘ qualityprogress.com

Standards-related developments and activities

Standard Issues
INTERPRETATION PROCESS

Open to The formal process to field

ISO 9001:2015 feedback and
questions by Paul Palmes and Dale Isaacs

Interpretation
“The organization shall” is a well-known phrase found throughout Inter-
national Organization for Standardization (ISO) standards. Its purpose
is to introduce a requirement. What follows these three words must be
addressed by an organization, appropriately implemented and capable of
verification. Unfortunately, sometimes the requirements following a “shall”
phrase can be a bit confusing.
There are countless consultants, fellow standards users, blog sites, books
and articles that can—and often do—provide excellent guidance for mys-
tifying and confusing clause content and terminology. But in the working
committee of the standard itself, there is another opportunity to ask questions annually. The same clause might
about the meaning of a clause: the interpretations process. lead a user to ask whether a
planned interval could be monthly
How it works or quarterly, for example. In both
The U.S. Technical Advisory Group (TAG) to ISO Technical Committee (TC) cases, the answer must be stated
176 (quality management systems) and the U.S. TAG to ISO/TC 207 (envi- as a simple yes or no.
ronmental systems) are two national committees that incorporate a formal ++ The TG22 chair and TAG adminis-
interpretations process into their ongoing activities. trator apportion users’ questions
In the case of U.S. TAG to ISO/TC 176, the U.S. task group on interpreta- to a small subgroup of TAG mem-
tions (TG22) process is documented in a published procedure and available bers who previously volunteered
publicly.1 TG22 employs a formal process that normally requires the following to respond to these questions.
approach (see Figure 1): Their job is to reach out to others
++ The user must submit his or her question via the form attached to the pub- in their networks for further input
lished procedure. before formulating their final posi-
The question must be a request for interpretation, not for information tion—yes or no—and providing the
or explanation, and should seek clarification about what the standard answer to the TG22 chair. If there
requires, not how the requirements could or should be fulfilled or applied. is consensus, the user is provided
There must be a brief narrative posed as a yes/no question, reference to with the TAG’s response.
the appropriate clause and the scenario that gave rise to the question. Lack of consensus is subject to
Interpretations must apply broadly to the entire user community, not just an further discussion in the subgroup
individual organization. to achieve a consensus opinion, or
For example, a user could cite Clause 9.2.1—Internal audit (“The organi- to advise the user that a consen-
zation shall conduct internal audits at planned intervals”2) and ask whether sus was not possible given the
it requires the organization to audit its entire quality management system circumstances. Consensus doesn’t

qualityprogress.com ❘ October 2018 QP 53

Standard Issues

necessarily mean unanimous agree- FIGURE 1

ment—it means absence of sustained
opposition to substantial issues.3
++ The results also are sent to the
U.S. TAG 176/SC 2/TG22 interpretations
standard’s international technical
committee as valuable material to process
consider for the next round of revi-
sions. The standard’s writing group is Start
especially interested in the answers
because these interpretations often N = ISO-assigned document number
Request from user SC = subcommittee
identify omissions, unintended or community SOP = standard operating procedure
undesired word choices, and possible TAG = technical advisory group
TC = technical committee
weaknesses in the original text. Advise user to consult TG = U.S. task group
TG22 also contributes to the interna- standards, support documents
tional Interpretations Committee (ISO/ and prior approved interpretations
TC 176, subcommittee (SC) 2, WG22) in
cases where individual member national Provide guideline procedure
standards bodies are unable to reach and forms to requestor
consensus in their country or wish to
post the request for global consensus. Request must meet U.S. TG22
International requests are handled criteria guidelines
in the United States using the same
procedure with one exception: After No Return to requestor;
expert opinions are submitted, the U.S. OK?
instruct to follow SOP
TAG votes by ballot to support the yes/ Yes
no opinion. TG22’s process is fully com-
TAG administrator accepts
pliant with ISO/TC 176/SC 2’s “Guidance and assigns tracking number
for Handling Requests for Interpretation
of the Requirements of ISO 9001.”4
Administrator routes
It’s interesting to note that until
request to TG22 expert No
recently, requests for interpretation
were slow. There is no clear reason
for the uptick, but it may be due to an Experts respond yes/no, providing Opinion
justification citing published documents OK?
increased awareness of the interpre- with clauses supporting position
tations process itself through word of Yes
mouth, blogs and articles.
Administrator sends outcomes
Yes
to experts and requests
Supporting evidence majority/minority
U.S. TG22 No Feedback to
consensus? requestor
The current chair of TG22’s interpre- opinions, as approriate
tations committee is this column’s
co-author, Dale Isaacs, said one of the Administrator will submit
Interpretation posted to secure site and
recent changes made to the interpre- to international process
copied to the TC 176/SC 2 secretariat
tations process was to paraphrase the “TC 176/SC 2/N 949 Doc”
mantra of the U.S. Supreme Court,
appropriating the committee’s tagline to Reply sent to
Stop
requestor
now read, “We don’t write the stan-
dard—we just interpret it.”
Likewise, and as reflected in the
procedure, each yes/no opinion must
be accompanied and supported by ISO

54 QP October 2018 ❘ qualityprogress.com

9001, and its supporting standards which in turn often creates misinter-
and documents. It must cite the Arguments such as, “That’s how pretations of clause content.
specific clause, section or area of the it’s always been done,” or, “It has In addition, ISO/TS 9002 is
document that supports the basis always been interpreted that way,” invaluable to understanding the
for the opinion. with no supporting established intent of some ISO 9001:2015
Arguments such as, “That’s reference are unacceptable. clauses. It was written by many of
how it’s always been done,” or, “It the same people who wrote ISO
has always been interpreted that 9001:2015. But because it’s a guid-
way,” without supporting established ance document, the authors enjoyed a far greater latitude of approach and
reference are unacceptable. Accord- writing style.
ing to TG22, this requirement lends ISO/TS 9002 provides guidance with clause-by-clause correlation to
more credibility to each interpretation clauses 4 through 10 of ISO 9001. It gives evidence of what an organization
and supports the sixth quality man- can do and provides examples, descriptions and options that aid in under-
agement principle: evidence-based standing the requirements of ISO 9001—and often aid TG22 in responding
decision-making. to interpretation requests.
Because consensus means lack of
sustained objection, it’s logical for a Additional resources
majority of TG22 members to have one While no one can argue that it would be wonderful if every user fully
opinion and a minority to have another. understood the intricate interactions, requirements and interpretations of
Like U.S. Supreme Court decisions that ISO 9001:2015, that isn’t—nor has ever been—the case. The interpretations
may shape future case law, the basis process managed by TG22 performs a valuable service in helping users
of those majority or minority opinions understand the “shalls” crucial to implementing these important standards.
may influence how future standards are An additional resource for questions involving other standards is ASQ’s Ask
written. the Standard Experts blog, which can be found at asqasktheexperts.com. 
Also, like U.S. Supreme Court opin-
REFERENCES AND NOTE
ions, sometimes the minority opinion is 1. “ISO 9001 Resources,” ASQ, asq.org/quality-resources/iso-9001.
used to establish future case law when 2. International Organization for Standardization (ISO), ISO 9001:2015—Quality
circumstances or context changes. management systems—Requirements, clause 9.2.1.
3. ISO and International Electrotechnical Commission (IEC), ISO/IEC Directives and
“We thought that, if done correctly, this Policies.
could help in writing future versions of 4. Information on the international process can be found at https://committee.iso.org/
standards,” home/tc176sc2.
Isaacs said.

For reference
Of the many resources used by TG22 Paul Palmes is principal consultant with Business
when rendering interpretations, among Systems Architects Inc. in Fargo, ND, and Prescott,
WI. He is chair of the U.S. Technical Advisory Group
the most helpful are ISO 9000:2015 and to the International Organization for Standardization
ISO/TS 9002:2016. Many of the inter- Technical Committee 176 (ISO/TC 176), and chair
pretation requests submitted to TG22 of ISO/TC 176, subcommittee 1, responsible for ISO
9000:2015.
can be explained by referencing the
definition of terms in ISO 9000:2015—
knowing the proper definition of a key Dale Isaacs is the owner of Brentwood Management
Consulting in Franklin, TN, an industrial and systems
word enables the understanding of a engineer and an ISO 9001 lead auditor. He is the
phrase. Any number of auditors can Secretary of the U.S. Technical Advisory Group to
attest to the difficulties of communi- the International Organization for Standardization
Technical Committee 176 (ISO/TC 176) and chair of
cating with people who simply don’t ISO/TC 176, U.S. TG22, responsible for interpretations
understand a word’s technical definition, for ISO 9001.

qualityprogress.com ❘ October 2018 QP 55

 SPECIAL ADVERTISING SECTION

Standards and Auditing Guide

A Direct Route
Are you confused about where to begin
with implementing standards within your
organization? Don’t get lost in the woods.
ASQ’s special section can help you start
down the right path. Here you will find:

• Accreditation
• Auditors
• Certification
• Education
• Tools
• Training

And there’s much more to see!

If you’re interested in appearing in next year’s guide, email mediasales@asq.org.

A2LA – American ASQ DQS Inc.

Association for Laboratory 600 N. Plankinton Ave. 1130 W. Lake Cook Road, Suite 340
Accreditation Milwaukee, WI 53201-3005 Buffalo Grove, IL 60089
5202 Presidents Court, Suite 220 Phone: 414-272-8575 Phone: 800-285-4476
Frederick, MD 21703 www.asq.org www.dqsus.com
Phone: 240-575-7499
ASQ’s training and education gives you
www.a2la.org
and your organization the best tools to
A2LA offers accreditation to and training on succeed. Its diverse range of learning topics
ISO/IEC 17025, ISO/IEC 17020, ISO 17034, and delivery methods are second to none.
ISO/IEC 17065, ISO/IEC 17043, ISO 15189 and Continue your path toward success with ASQ
CLIA requirements. training. Discover a broad range of course
CONTINUING EDUCATION • E-LEARNING • offerings from classroom-based, virtual,
METROLOGY • ON-SITE • QUALITY blended and web-based learning.
AUDITS • QUALITY TRAINING • AUDITING • BLENDED • CERTIFICATION
STANDARDS TRAINING PREPARATION• CERTIFICATION
REFRESHERS • CONTINUOUS
IMPROVEMENT • CORPORATE • DELIVERY
METHODS • E-LEARNING (WEB-BASED) •
LEAN/SIX SIGMA • PUBLIC • QUALITY
FUNDAMENTALS • STANDARDS/
COMPLIANCE • VIRTUAL

56 QP October 2018 ❘ qualityprogress.com

 THE VALUE OF SIX SIGMA
FORUM MEMBERSHIP
When you become a valued member of the ASQ Six Sigma Forum, you join a distinguished
group of problem solvers and analytical thinkers driven by a desire to achieve breakthrough
improvements. Your membership not only exposes you to a remarkable network of individuals,
but also links you to the Six Sigma Body of Knowledge and supporting resources and case
studies. Lastly, you will stay plugged in to the Six Sigma community through monthly newsletters
and members-only online communities.

Call ASQ Customer Care at 1-800-248-1946 and ask to

have the Six Sigma Forum added to your membership.

For more information, visit asq.org/sixsigma.

 Newly released products and tools

Marketplace
ACCELEROMETER MODULES CONTOUR CONTROL

Samples Machine capable

available of milling
through and turning
Amazon operations
Silicon Designs now offers Heidenhain’s TNC 640 features a touch screen that
samples of their micro elec- allows operators to navigate through long lists,
tro-mechanical systems variable programs, tables and other content through what is
capacitive (MEMS VC) acceler- known as “kinetic scrolling.” The kinetic scroll converts
ometer modules, in ranges from the impulse of hand contact into a scrolling speed. A
±2 g to ±100 g, through Amazon. short, slow swipe scrolls the screen over a few lines,
com. while a long, fast swipe scrolls it over many lines.
The MEMS VC accelerometer Users can stop dynamic scrolling at any time with a short tap. Users move the pro-
modules’ on-board voltage reg- gramming and 3-D test graphics directly on the screen. The desired graphic view can be
ulation and an internal reference adjusted conveniently by touch. Users also can use touch gestures to zoom in and out for
minimize temperature and details, as well as move or rotate graphics.
voltage changes while eliminat- www.heidenhain.us | 847-884-4761
ing additional external power
regulation requirements. Series FOOD ADDITIVES
units are relatively insensitive to
wider temperature changes and Additives prevent
gradients. Carefully regulated
manufacturing processes ensure plastic-contaminated food
Eriez’s PolyMag additives are used by molders for imparting metal detectability, X-ray con-
trast and magnetic susceptibility into plastic tools used in the food manufacturing process,
which have the potential to make their way into food products. Growers, packers and food
processors use their existing metal detectors and X-ray inspection systems with detect-
able plastics to prevent plastic-contaminated foods from reaching consumers, effectively
preventing costly recalls and ensuring compliance with ever-increasing government safety
standards.
Major brand owners, retailers and restaurant chains specify plastic and rubber arti-
cles that are detectable and make this part of their hazard analysis critical control point
programs. Additionally, the Food Safety Modernization Act is spurring the implementation
of new hazard analysis and risk-based
preventative controls to prevent plastic-
contaminated food products from reaching
that each accelerometer is man- consumers.
ufactured to virtually identical By incorporating PolyMag additives,
performance standards, allowing molders can satisfy varying levels of
users to swap out parts in the detectability at a cost substantially less than
same g-range with few-to-no specialty compounds. This eliminates the
testing modifications. need to buy minimum quantities of special
www.silicondesigns.com | resins and carry excess material inventory.
425-391-8329 www.eriez.com | 814-835-6000

58 QP October 2018 ❘ qualityprogress.com

 STEP ACCENT LIGHT

LED lights offer

90-degree protection
POWER TAKE OFF

PTO
ideal for
severe
applications
Twin Disc has released the HP500 power take
off (PTO) that is hydraulically actuated that
has a power rating of 373 kW at 1,800 rpm.
The HP500 has a modular design that allows
TecNiq’s step accent LED light, the E02, is waterproof and offers for side-load applications. A spring-applied,
90-degree projection. The rugged design of the E02 uses an ultra violet hydraulically released brake has been added
and impact resistant polycarbonate. It’s waterproof to IP68 standards as a standard feature. Pads and splines are
with a urethane potting and offers a 100,000-hour service life. interchangeable to accommodate varying
The E02 is available in white, blue, red and green LED colors. It can pump sizes.
be used to illuminate passageways, steps, stairs and as accent or task The HP500 is ideal for driving pumps,
lighting. It is ideal for work trucks, heavy machinery and emergency grinders, crushers, dredgers, chippers,
response vehicles. shredders, heavy-duty drills and other severe
www.tecniqinc.com | 269-629-4440 applications.
www.twindisc.com | sales@twindisc.com

LED EMITTER

Emitter offers uniform light for

high precision applications
TT Electronics has introduced a com- security detectors and other equipment off transitions. The integrated collimat-
pact, infrared LED emitter with a spot requiring dependable position, proxim- ing lens creates a tight beam profile for
diameter of 7mm with optical sensing ity or motion sensing. use with accuracy-dependent devices
and position encoder applications. The With up to 10mW total radiated such as radial or linear encoders for
OP207CL has an operating temperature power at the maximum drive current of absolute or incremental measurement,
range of -40°C to 105°C, which can 100mA, and 2.25° angle of half inten- as well as long-range light curtains,
withstand environments in a wide range sity, the OP207CL couples optical flux edge detectors, scanners, and general
of applications including industrial efficiently onto the receiving photo- optical sensing and switching.
automation, safety systems, robotics, sensor to ensure clearly detectable on/ www.ttelectronics.com | 972-323-2315

qualityprogress.com ❘ October 2018 QP 59

Managing Organizational Risk Using the Supplier Audit Program
LANCE B. COLEMAN SR. ❚ ASQ QUALITY PRESS ❚ 2018 ❚ 154 PP. ❚ $33 MEMBERS, $55 LIST (BOOK).

A practical handbook to a robust supplier audit program, soft skills. Readers are given prac-
this book guides readers through the steps to plan and exe- tical advice on several fronts. The
cute an effective risk-based supplier audit. There are several book also contains 25 questions to
handy tools and utilities that the author provides as job guide a risk-based audit from start
aides. Additionally, the many case studies provide practical, to finish.
first-hand scenarios of best practices and lessons learned Effective risk management is at
on topics related to audits, including several of the author’s the heart of any successful business
colleagues across the globe and across multiple industries. environment. This is even more perti-
Part one of the book focuses on end-to-end logistics to nent to global organizations that see
holistically plan, execute, report and close out risk-based more than half of their value-creation occur outside of their
supplier audits. Elements of risk analysis are incorporated in walls. Applying risk-based thinking to supplier management
each of these steps through pointed questions and intro- life cycles is at the heart of this book. This book presents a
spection. For instance, explicitly documenting risks and well-curated collection of methods to perform a supplier audit
hazards in the input, performance and output of a process while keeping risk-management front and center.
being audited. —Anuradha Rangarajan, Harvard, IL
Part two focuses on equipping auditors with a balance of

Additions to your quality library

Footnotes
Mass Customized Manufacturing: Theoretical Concepts and Practical Approaches
VLADIMIR MODRAK ❚ EDITOR ❚ CRC PRESS ❚ 2017 ❚ 332 PP. ❚ $93.95 (BOOK).

In many markets, customers are The first section on trends and success factors provides an
seeking a customized product that overview of designing production lines along with the role of
satisfies their needs. The manu- information systems in mass customization.
facturer who can quickly provide Section two examines the drivers of complexity in sourcing,
customers with the desired product manufacturing and distributing a customized product. This
will continue to capture and expand section is technical in nature and provides several theoretical
market share. This book can be used models that show how complex the manufacturing system and
as a guide for organizations that supply chain may be based on the given inputs.
desire to provide customized prod- The last section addresses sustainability—namely, sustain-
ucts to their customers. ability of the business by understanding consumer trends. The
This book contains 12 chapters divided into three sections section also covers the topic of environmental sustainability.
covering trends and success factors, complexity drivers, and The 12 individual chapters cover a wide range of topics.
management and sustainability of mass customization. Each Readers who are interested in any aspect of mass customiza-
chapter is written by an expert or experts on the given topic. tion will find chapters that address that topic. Readers seeking
Overall, 21 international experts on specific aspects of mass an overview of mass customization also will find many topics of
customization contributed to this book. Each chapter is a interest.
self-contained work on a specific topic, containing a presenta- —Gene Placzkowski, Racine, WI
tion of the material, examples and references.

60 QP October 2018 ❘ qualityprogress.com

Linear Regression Analysis With JMP and R
independent errors, regressor assump-
RACHEL SILVESTRINI AND SARAH BURKE ❚ ASQ QUALITY PRESS ❚
tions, model assumptions and model
2018 ❚ 468 PP. ❚ $30 MEMBERS, $30 LIST (BOOK).
diagnostics summary are discussed.
This book is all about linear regression analysis using JMP and R. Chapter seven discusses remedial
In the authors’ words, it is a “comprehensive low-cost textbook measures such as transformations to
intended for use in an undergraduate level regression course, as the response, transformations to the
well as for use by practitioners.” regressors and analytical approaches
The book is divided into 10 chapters. The first chapter to transformation. Chapter eight dis-
discusses the regression model-building process. There are cusses some advanced topics relating
brief, informative tutorials on using JMP and R. A reader who to model selection techniques. Chapter nine discusses model
has never used JMP and R may see the need to refer to other validation, which is an important aspect of applying regression
books on JMP and R to become more familiar using them. The models, and readers will appreciate the discussion of metrics
second chapter deals with data collection and cleaning. This for model validation. Chapter 10 presents a detailed case study.
book discusses how to clean data in terms of formatting data, There also is an interesting and informative discussion about
dealing with missing data, and dealing with new and combined model presentation methods.
variables. Each chapter has JMP and R software tutorials pertaining to
The third chapter discusses data visualization using histograms, the topics presented. The book is written in simple language,
box plots, scatter plots, bar charts, contingency tables and mosaic which allows readers with any level of statistical knowledge
plots. In chapter four, the linear regression equation is discussed to understand and apply the concepts presented. There are
along with tests for significance of regression, coefficient of numerous graphs and explanations of the concepts throughout
determination and statistical inference on the individual regression the book, which are useful in comprehending some of the top-
coefficient. Chapter five discusses model prediction and inference ics. This book is a valuable addition to the library of students,
along with confidence intervals, prediction intervals, extrapolation professors, industry practitioners and other users of linear
and influential observations. Chapter six deals with model diagnos- regression.
tics. In this chapter, topics such as errors being normally distributed, —Rangarajan Parthasarathy, Harvard, IL

Senior Management and Quality: How to Leverage Quality for Profit creativity, collaborating with internal and
FIN ROONEY ❚ ASQ QUALITY PRESS ❚ 2018 ❚ 210 PP. ❚ $24 MEMBERS, $40 LIST (BOOK). external partners to drive improvement,
and allowing subordinates and mid-level
What does the word “quality” mean? quality plan is keeping things simple. This employees to own and showcase their
The word might have many definitions strategy of simplicity is one of the ways successes. Owning success means that
to diverse types of people, and they may senior leadership can establish processes those employees must own their pro-
all be correct. What we know to be true, that provide support to the quality of the cesses, any changes to those processes
though, is that there are a few absolutes organization and protect its bottom line. that drive continuous improvement, all
that let us standardize the concept. Good leadership performance metrics and any knowledge
In Rooney’s work, you see connec- also should believe required to understand why the product
tions between the customer experience, in the product as becomes known for its specific level of
production lines, R&D, and leadership much as the team. quality. This total quality approach allows
and management. These workflows can To create a sus- leaders to justify rewarding performance
be continuous, opposing, connected or tainable business that promotes that quality.
simultaneous, but in any case, they all model, leaders In addition to determining quality to
lead to a product that can be rated with must perform or be a consistent strength throughout any
a specific tier of quality. solicit mar- product line, the author reminds readers
Initially the text is as much of a critic ket research and that all departments involved in the value
of the quality profession as a proponent analysis. In a way, stream—which allow the product to be
of it. In the field or in the factory, quality the customers or created—are responsible. From HR and
teams find themselves using a multitude clients within a given market determine talent-sourcing agents for management,
of three-letter acronyms to determine what the word “quality” means. Another to the janitorial staff, which keeps infec-
the quality of a product or service. This market’s customers might have different tious material away from patients at a
alphabet soup can be confusing and expectations and select another product hospital, all team members have a role to
demeaning, which can lead to impasse based upon those norms. play in keeping the customer experience
on the road to knowledge or discovery. Leaders must strategize, support and at the forefront of the quality profession.
One of the keys to executing a solid sell quality. They can do this by fostering —Trevor Jordan, Orlando, FL

qualityprogress.com ❘ October 2018 QP 61

PROFESSIONAL SERVICES For information on placing an ad, contact Media Sales at 866-277-5666.

TQM ASSOCIATES INC.

“we’re here to support you”
Leadership
• Established in 1994
Customer • Women-Owned
Strategy Experience • Quality Assurance Professionals
• Across the U.S. and Worldwide
C3 Here’s a great way to improve • Temporary or Permanent
Inside
quality or implement a • Source Inspection
Culture Innovation
Quality Management System • Surveys
and it’s FREE! • Audits
Measures • Expediting
of Success
qmscapa.net 800-424-4729 3990 Old Town Ave. #C109
Fax 619-297-3251 San Diego, CA 92110
Align, integrate and advance with
tqmassociates.com email: stephk@tqminc.net
Customer-Centered Culture (C3)
workshops, books and tools
“Old ways simply do not work any more.
We have had no ways of measuring
human knowledge until now.”
A. Sheyner, IT Director, Healthcare

“Using Rob’s system, we added $48 million to our

bottom line and virtually all our key measures of
success were significantly better a year later.”
K. Markland, CEO, Financial Institution

“The Baldrige provides an excellent lens to

diagnose and prioritize; C3 provides a useful
approach to significantly transform
an organization’s culture.”
B. Lassiter, Executive Director, Nonprofit

www.C3Excellence.com 941-704-9888

IMDS Data
Looking • We can create your data and submit within
72 hours
for Quality • We do the work and invoice you
• References found on our website
Professionals? • Based in the United States
www.imdsdata.org • 734-205-8874

Place a
Recruitment Ad in Classroom Training Aids…
G Quincunx Boards

Quality Progress G Sampling Bowls

G Catapults
G Deming Funnels
and reach more G And lots more …
Visit us at: www.qualitytng.com

than 100,000 Email sales@qualitytng.com for brochure

Ph: 248-641-7030 Fax 248-641-7031

readers from all PO Box 611 Troy, MI 48099-0611

over the world!

Advertisers Index
ADVERTISER PAGE PHONE WEB
Contact Media Sales
CyberMetrics Corp. OBC 800-777-7020 www.cybermetrics.com
at 866-277-5666. EtQ Inc. IFC 516-293-0949 www.etq.com

62 QP October 2018 ❘ qualityprogress.com

 THANKS, VIEWERS
AND CONTRIBUTORS*

videos.asq.org
*3M views and counting!
 H w
The w w w
A refresher on a quality
tool, concept or method
Right H
w
Questions
Make the most of problem
solving by asking effective
questions
by Ondrej Ďurej

All problem solving begins with a These are excellent questions, but of the problem (symptoms) to its
problem description. One of the best aren’t part of the problem description. essence (primary defect). After the pri-
and most-used methods for describ- They’re part of the problem analysis mary defect is known, the point of origin
ing a problem is five W’s and two H’s, and should be asked later. The right can be determined.
where the questions who, what, where,
when, why, how and how many/how
descriptive questions are:
++ Who detected the problem?
Bullet The point of origin is the answer
to the question, “Where (in which
much are asked. These questions also ++ How is the problem manifested? process) did the problem arise?” It’s
can be used in problem analysis. ++ Why is this a problem? Bullet important to know that quality prob-
People rush to conclusions when The problem was detected by the lems arise only in processes. Therefore,
they intertwine descriptive and ana- problem finder. His or her answer to the point of origin and the root cause
lytical questions. Instead, descriptive the question, “How is the problem can only be found in a process. To find
questions must be asked first, followed manifested?” is called the voice of the the root cause, you also must know
by analytical questions. When the finder. Because operators and quality when the problem arose.
questions are answered, action can be technicians are the best people to Asking who caused the problem
taken to solve the problem. (A sample identify the symptoms of a problem, is common, but the real question is,
action is shown in Online Table 1, which typically they are the problem finders. “Who is accountable?” or “Who is the
can be found on this article’s webpage The answer to the question, “Why process owner?” Although an operator
at www.qualityprogress.com.) is this a problem?” tells you why the can be responsible for the problem,
Answers to the descriptive questions product is unacceptable to the cus- the person accountable always is the
what, where, when and how many/ tomer and thus is called the voice of process owner, who also is responsible
how much typically are easy. You can the customer. Typically, it’s provided by for the solution to the problem.
ask, “What is the problem?” or, “What a quality engineer who, in addition to When looking for the root cause, it’s
product is defective?” identifying the respective symptoms, useful to know the causal mechanism
“Where” questions ask for the point also can express the importance of the or mechanism of origin: “How did the
of detection: “Where was the problem problem. Obviously, there’s a big dif- problem arise?”
detected?” Logically, “when” questions ference between minor symptoms and To improve problem solving, every
ask for the date and time of detection: symptoms that stop production. employee must know: a description,
“When was the problem detected?” When looking for root cause, the analysis and action, and the root cause
The final descriptive questions are, key analytical question is, “Why did is found in a process.
“How many products are defective?” the problem arise?” When you know
or “How much of production (in per- the root cause, you can take corrective
centage) is defective?” action; no other questions are necessary.
These questions and answers are But that’s easier said than done, so
straightforward, but things get tricky problems must be investigated sys-
Ondrej Ďurej is a
when asking who, how and why: tematically, especially if features can freelance consultant for
++ Who caused the problem? be examined for only a limited time. Odecon in Slovakia. He
++ Who is responsible? The essence of a problem often is earned a master’s degree
in electrical engineering
++ How did the problem arise? hidden under several layers, so you must from Slovak Technical
++ Why did the problem arise? proceed from the outward manifestation University in Bratislava.

64 QP October 2018 ❘ qualityprogress.com

DO YOU MISS YOUR
PRINTED JOURNAL?
In 2018, the Journal of Quality Technology (JQT)
and Quality Management Journal (QMJ) launched
a new subscription model, allowing ASQ members
Full and above to access these valuable articles
and issues online. Find them here: asq.org/pub/.
But we didn’t do away with print entirely. Full and
above members may now purchase a hard-copy
version of these journals, containing a full year’s
articles in one volume. Order by October 31 to
reserve your copy for $45 each. The annual hard
copies of QMJ and JQT for 2018 will be mailed at
the end of the year.

Place your order today at http://asq.org/annual-print-qmj

or http://asq.org/annual-print-jqt.
 Reliable software
fosters reliable equipment.
For over 30 years, GAGEtrak calibration management software has been at the core of the world’s best quality
management programs, assuring unwavering accuracy, efficiency and standards compliance. And now, with
the addition of GAGEtrak Lite to our product line, we have a software solution to meet every need and budget.
Contact us today to discuss which one is right for your company or download both for free and see for yourself.

GAGEtrak and GAGEtrak Lite

Two levels of calibration management. One auditor-recommended name.

Download for free at: GAGEtrak.com/downloads 1-800-777-7020 +1 (480) 922-7300