28
||
QP
||
February 2021
LEAN SIX SIGMA
Incorporating
agile methods
in lean Six
Sigma to get
results—fast
by Jay Arthur
Nastco via Getty Images/Tara Moore
JUST THE FACTS
In some business
scenarios, full-scale
lean Six Sigma
training isn’t always
necessary to
dissect and solve
problems. Smaller
projects often can
be shortened and
handled using a
select number of
quality tools.
In some instances,
that means
overtraining teams
can be avoided. This
agile approach to
lean Six Sigma can
save on training
costs.
More complicated,
cross-functional
problems within
an organization
may require more
advanced training
for teams, but many
problems can be
tackled more quickly
by those with less
training.
qualityprogress.com
||
QP
||
29
LEAN SIX SIGMA
ver the past 20 years, many
organizations have taken
the traditional path of Six
Sigma implementation—train
Green Belts (GB) and Black Belts
(BB), start teams and hope for the
best.
Unfortunately, after two or three years
of marginal return on investment, many leadership
teams shut down their quality departments. Recently,
organizations have been rediscovering how to mini-
mize the cost of lean Six Sigma and get results. I call
this agile lean Six Sigma.
At recent ASQ Lean and Six Sigma Conferences, ASQ
World Conferences on Quality and Improvement, and
the Lean Six Sigma World Conferences,
1
organizations as
diverse as Christus Health,
2
Novartis,
3
Crayola and Under-
writers Laboratories (UL) have presented their approaches
to getting results. All outline the same key steps:
1. Choose a problem that has preexisting data.
2. Allocate a one to two-day Yellow Belt (YB) training
for teams focused on solving that problem.
3. Develop and implement an improvement project
with countermeasures for implementation.
This is the nuts and bolts of agile lean Six Sigma:
focus on results.
One- or two-day training
Having attempted the traditional approach to Six Sigma,
Christus Health wasn’t gaining any traction. Leadership
asked whether the quality staff could develop a one-day
course to train teams and deliver results. Like good sol-
diers, the quality staff took a whack at it and discovered
it worked. Key takeaways included:
No team could be trained unless it had a problem
to solve and preexisting data about the problem. No
training took place just to boost individuals’ résumés.
One-day training focused on using data and a few
key tools to solve the problem.
Implementation of countermeasures followed. Much
to the staff’s surprise, results started flowing from
the teams focused on real problems and results.
These projects helped identify individuals with
an aptitude for quality improvement. These team
members were selected for additional GB or
BB training.
This agile approach to lean Six Sigma proved
especially useful in healthcare, which could ben-
efit from rapid improvement. While healthcare
organizations have tried performance improve-
ment, as Six Sigma often is known in healthcare,
it often has failed to deliver results. Doctors,
nurses and other healthcare workers cannot be
pulled off the line for weeks of training. They
need a way to learn a few key tools, apply them
and implement them to achieve results.
Similar to Christus Health, Novartis used a
two-day training session to accelerate the avail-
ability of sample drugs to their sales reps. Crayola
used a two-day training focused on results.
Because UL has 14,000 employees spread all over
the world, it used a blended method: one day of
online training followed by one day of on-site
team meetings to solve specific problems.
Agile lean Six Sigma
“Six Sigma is too complex and time-consuming to
fit into a regular workday. We need tools that don’t
require the entire organization to undergo weeks-
long training programs.”
—Thomas Wedell-Wedellsborg
4
The 20
th
century quality improvement
approach tried to prepare people to work in
manufacturing, where you needed measurement
systems analysis, design of experiments and other
advanced tools to optimize production. In the 21
st
century, most people work in service industries
(Figure 1) in which these tools are not useful and
often confusing.
Traditional lean Six Sigma required two to four
weeks of training over several months, and four
to 16 months to complete a project (see Online
Figure 1, which can be found on this article’s
webpage at qualityprogress.com). While this is
great for training and consulting companies to fill
billable hours, it slowed the adoption of lean Six
Sigma and put it on the list of endangered corpo-
rate programs.
1 3 52 4
30
||
QP
||
February 2021
LEAN SIX SIGMA
iStock.com/MrsWilkins
A bit of hacking
“The spirit of hacking can be adapted and
applied to general business management,
not just technical innovation.”
—Scott Brinker
5
The solution to this problem of
sometimes overtraining is to hack lean
Six Sigma to deliver results in hours or
days, not months and years. So, how do
we hack lean Six Sigma? Stop teaching
people things they don’t need to know
to solve problems they don’t have to
impress people they don’t like.
Hacking lean: In service indus-
tries, 95% of delays are unnecessary
and preventable. I call this the 3-57
rule (Online Figure 2). People are
only working on the service for three
minutes out of every hour. The other 57
minutes are unnecessary delays.
Value-stream mapping and spaghetti
diagramming will help pinpoint how to
eliminate these delays. Virtually all of
U.S. employment 1900–2014
FIGURE
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1900 1950 2014
SERVICE, 80%
M
ANUFACTURING, 10%
ASQ
1946
F
ARM, 2%
Source: www.bls.gov
Raw PivotTable data
FIGURE
80%
50%
20%
4%
1 step out of 25
qualityprogress.com
||
QP
||
31
LEAN SIX SIGMA
‘Magnificent Seven
Power Tools’
the delays are in the arrows between steps. You don’t have to
make people faster or more efficient. You just have to make
the product or service faster by eliminating delays.
Often, this can be done in an hour or two, and it doesn’t
require a week-long kaizen event. All you need are Post-it
notes and people who work in the area that’s under scrutiny.
I’ve done this with nurses, and they always find solutions.
For example, one group found ways to cut travel time by
50% or more in a nursing unit. The exercise rarely takes
more than two hours.
Hacking Six Sigma: You’re probably familiar with Pare-
to’s 80-20 rule, which states that 20% of what you do causes
80% of your waste, rework, and lost profits. Pareto’s rule is
actually a power law, which means that 4% of what you do
produces more than half the waste, rework, and lost profit.
I call this the 4-50 rule (see Figure 2, p. 31).
The same is true of knowledge. You don’t need to
know everything to solve most of these problems. A few
tools—I call them the magnificent seven (see Figure 3)—will
solve more than 90% of the problems facing businesses today.
6
So, you don’t need to fix everything or know every tool
in the toolbox. These two key concepts are crucial to under-
standing agile.
If 4% of what you’re doing is producing more than half
the waste and rework, that means one out of every 25 steps
is causing 50% of the waste rework and lost profit. It might
be as few as one out of 100 steps. So, you just need to find the
$1 million misstep.
Using PivotTables
The magnificent seven includes value-stream mapping and
spaghetti diagrams, control charts, Pareto charts or histograms,
fishbone diagrams, countermeasures matrixes, and action plans.
There’s one more tool that’s often missing from Six Sigma.
Consultants often talk about low-hanging fruit, but what
they don’t tell you is that most of it has been picked already.
There is, however, what I call invisible low-hanging fruit:
Raw data about these problems reside in Excel files and
corporate information systems.
To find the invisible low-hanging fruit, you need Excel
PivotTables to summarize this raw data and chart the results.
Every multimillion-dollar improvement project I’ve ever
worked on started with a PivotTable of raw data. It used to
take me hours—and sometimes days—to fully analyze the
data to create improvement projects. I can now complete
analysis in a matter of seconds.
FIGURE
1. Value stream mapping and spaghetti diagramming—to map and analyze
process flow.
2. Excel PivotTables—to summarize raw data about defects, mistakes, errors,
and money.
3. Control charts—to identify the most frequent types of defects, mistakes,
errors, money, and time.
4. Pareto chart—to identify the most frequent types of defects, mistakes,
and errors.
5. Histogram—to evaluate deviation in process performance, especially time.
6. Fishbone—to identify the root causes of the problem.
7. Countermeasures and action plan matrixes—to prioritize corrective actions.
4-hour lean hack—simplify and streamline
5S—Sort,straighten,
shine, standardize
and sustain
Eliminate delays and movement
Value
Map
Stream
Spaghetti
Diagram
4-hour Six Sigma hack—optimize
Analyze
4-hour Six Sigma hack
Pivot table Control chart Pareto chart Cause-eect
diagram
Improve
Cause eect
diagram
Countermeasures
What Who When HowHow
Action plan
Before
After
Good
Before
After
Control chart Pareto chart
Comparing Control Charts Comparing Pareto Chart
Before and After
Histogram
LSL USL LSL USL
Control
Control Chart
Honor
Recognize, reward, refocus, and repeat
iStock.com/curraheeshutter
32
||
QP
||
February 2021
LEAN SIX SIGMA
PivotTable case study
Most hospitals have too many problems
with rejected, appealed, and denied
claims costing millions. Lean Six Sigma
can help reduce billing problems among
other operational problems—and the
process is simple.
Denied claims drive up the cost of
healthcare and can push many hospitals
toward bankruptcy. For one hospital
system in the Southwest, monthly deni-
als were more than $1 million. Using raw
data shown in Table 1, PivotTables were
used to summarize the data and turn it
into control charts and Pareto charts.
For instance, the team:
Summarized the data by date and
defect (Figure 4), which yielded
a control chart of defects by day
(Figure 5, p. 34).
Summarized denied charges by type
of denial, which yielded a Pareto
chart of the costliest defect (Figure 6,
p. 35).
Drilled down into the timely filing
data (double-click on any cell in a
PivotTable and it will reveal the raw
data used to create it). From there, we
created another PivotTable to discover
that one insurance company caused
67% of denials (Online Figure 3).
This enabled the team to pinpoint
the problem: One insurance company
was causing 67% of the denied claims
due to timely filing. In a matter of
hours, the team changed manual pro-
cesses to prevent the problem, saving
$5 million a year.
Looking at the data regarding
rejected claims, the team found a way
to eliminate $24 million in rejected
claims. While rejected claims are
ultimately paid, rejection causes
delays that can result in denials.
7
Hacking lean Six
Sigma training
How do you train people in a day and
deliver results? Restrict the training
to the seven tools and use software to
automate data analysis.
Don’t teach people formulas or how
to calculate limits and draw control
Pivot table of denied charges
FIGURE
GET TRAINED
ASQ oers training extensive training on lean, Six Sigma and many other
quality-related practices and tools. Courses are designed and developed by the
best subject matter experts and taught by the best instructors. Explore ASQ’s
extensive selection of courses and find the right level of training for you and
your sta by visiting asq.org/training.
Raw PivotTable data
TABLE
PivotTable of denied charges
FIGURE
qualityprogress.com
||
QP
||
33
LEAN SIX SIGMA
YB training, the investment is much less and the results
more forthcoming.
Surprisingly enough, some of these participants will
continue to develop more projects. Consider them for more
advanced training. They’ve earned it.
Ready to delight
So, that’s agile lean Six Sigma:
Only train teams with real problems to solve and preexist-
ing data. They will learn faster and avoid the one-and-done
problem of most lean Six Sigma trainings.
Train teams on the seven key tools needed for problem solv-
ing. Anything else is overproduction. It causes confusion.
Focus on analyzing the data and developing
countermeasures.
Implement and measure results.
Use control charts to monitor and sustain the
improvement.
Charges coded as denials
FIGURE
$3 million
$2.5 million
$2 million
$1.5 million
$1 million
$500,000
$
Charges
UCL
CL
$1,071,509
$2,552,122
CL = central line
UCL = upper control limit
charts by hand. Don’t teach them how to manually draw
Pareto charts or histograms. Give them software that does
this automatically. Without Six Sigma software, it will be
impossible to achieve rapid results using agile methods or
train people in a single day. Obviously, the software must be
easy to learn, or training will take longer than a day.
When teams focus on learning tools to solve a pressing
problem, they learn much more quickly than when trying
to learn from randomly designed exercises using imaginary
organizations and data. Too many healthcare workers have
been trained using manufacturing data. It doesn’t stick.
Using their own data makes the training stick.
Many presentations at conferences point to one-and-
done problems with lean Six Sigma training. People go to
class for two to four weeks, do one project and stop. That’s
a terrible waste of resources. Your organization will spend
a lot of money to complete just one project. What if you
get two or three projects done in one day? With one-day
Editor’s note: Chart recreated but
originally generated using QI Macros software
iStock.com/muangsatun
34
||
QP
||
February 2021
LEAN SIX SIGMA
Denial—no appeal charges by memo code
FIGURE
Editor’s note: Chart recreated but originally generated using QI Macros software
Repeat as needed.
Identify people for more advanced training to tackle
more complicated cross-functional problems.
This agile approach will delight leadership and
customers. It also prevents the extinction of the quality
department and lean Six Sigma.
8
QP
EDITOR’S NOTE
References listed in this article can be found on the article’s webpage at
qualityprogress.com.
Jay Arthur is president of Knowware International Inc. in Denver. He holds a bachelor’s degree in systems engineering from the University
of Arizona in Tucson. Arthur, an ASQ member, is the author of several books, including Agile Process Innovation: Hacking Lean Six Sigma to
Maximize Results (Lifestar Publishing, 2019), Lean Six Sigma Demystified, second edition (McGraw-Hill Education, 2010) and Lean Six Sigma
for Hospitals, second edition (McGraw-Hill Education, 2016).
12 million
10 million
8 million
6 million
4 million
2 million
0
Claim denied (no appeal)
Categories
Timely filing Medical necessity No authorization Partial authorization
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Claim denied (no appeal)
Invalid auth for
patient status
91.5%
97.4%
61.6%
81.3%
WATCH AND LEARN
Watch ASQTV video interviews of Azizeh Elias Constan-
tinescu and Newton Moore of Underwriters Laboratories
as they discuss the importance and benefits of applying the
agile approach to training design and development in adult
learning. Visit videos.asq.org/applying-agile-methodology
to watch these interviews.
qualityprogress.com
||
QP
||
35
LEAN SIX SIGMA
