Quality Progress - January 2016 - 38
BACK TO BASICS
BY PETER J. SHERMAN
Guidelines to manage processes effectively
CONTROL CHARTS are at the founda-
control limits. For attribute charts (p-chart),
you're interested in improving monthly
tion of Six Sigma. Invented by Walter A.
the suggested sample size is at least 50.
customer satisfaction scores. You collect
Naturally, the more data you can collect
Shewhart while working for Bell Labs in the
scores from August to November 2009 and
1920s, control charts serve as the primary
during an extended period of time, the
plot the data in time series using an I-chart.
tool to filter out the probable noise (inher-
better it can be used to see how the process
The chart measures the process mean, and
ent variation or common cause) from the
behaves. But the time and cost to collect a
the control limits are calculated:
potential signals (nonrandom variation or
sample must be balanced with the amount
special cause).1 From this, you know when
of information needed.
Consider the process being observed.
and where to take action on a process.
When first learning about control charts,
a few basic questions must be addressed:
UCLx = X + 3MR
LCLx = X - 3MR
UCL represents the upper control limit,
Does it operate in a fairly steady state? If so,
LCL is the lower control limit, and X repre-
a few weeks of historical data is sufficient.
sents the process mean. MR is the average
Meanwhile, pharmaceutical companies
* How many data points do you need?
* What length of time should be examined?
often must collect tens of thousands of
* When should you recalculate control
historical data points because they're
moving range, and the constant d2 is from
the Shewhart table (n = 2).
You confirm the process is stable.
dealing with human lives. It's important
Improvements are made in December
Applying basic guidelines and common
to understand the context of the process
2009, and you continue plotting the data
when deciding how much data to collect.
during the next year (see Online Table 1 at
sense can help answer these questions.
www.qualityprogress.com) using the same
The number of data points needed in a
Revising control chart limits should be
Notice the upward shift in the mean dur-
control chart varies. For variable data used
handled with similar care. Assume you
ing the next 14 months. The process does not
in an X-bar and range (R) chart, a minimum
collect data, and the control chart shows
appear to be in statistical control given sever-
of three to five data points per sample and
evidence of special causes (a point on or
al points outside the UCL, as shown in Figure
20-25 groups of samples is appropriate.
outside the limits). You know the process
1. But the improvements made in December
is not stable or predictable.
2009 changed the process. Recalculating the
With an individuals (I) and moving range
(MR) chart, in which the sample size is one
After you identify and remove the as-
limits on the first control chart.
limits depicts that change (see Online Figure
because data occur much less frequently,
signable cause(s), collect additional data
1 at www.qualityprogress.com). The process
12-24 values are reasonable to compute the
and recalculate the mean and control lim-
is actually in statistical control.
scores / FIGURE 1
its. Observe the control
Control charts are powerful tools for oper-
chart to confirm it is in
ational excellence professionals. Being aware
statistical control. Con-
of these guidelines and using common sense
tinue plotting new data
can ensure good decision making and allow
but do not recalculate the
control charts to be used effectively. QP
If the process does not
change, the limits should
not change. If you make
changes to the process,
recalculate the control
CSAT = customer satisfaction
LCL = lower control limit
UCL = upper control limit
38 QP * www.qualityprogress.com
limits to observe shifts in
variation after the change.
For example, imagine
1. Donald J. Wheeler, Making Sense of Data, SPC Press, 2003.
PETER J. SHERMAN is director of process excellence at Cbeyond Communications in Atlanta and lead instructor in
Emory University's Six Sigma certificate
program in Atlanta. He has a master's
degree in civil engineering from the
Massachusetts Institute of Technology
in Cambridge, MA, and an MBA from
Georgia State University in Atlanta. A senior member of ASQ,
Sherman is an ASQ-certified quality engineer and a certified
lean Six Sigma Master Black Belt by Smarter Solutions Inc.
Table of Contents for the Digital Edition of Quality Progress - January 2016
According to Plan
Use Your Head
Stakeholder Management 101
All About Data
Eight Simple Steps
Which Six Sigma Metric Should I Use?
Turning ‘Who’ Into ‘How’
In the Beginning
Outputs and Outcomes
That’s So Random—Or Is It?
Improving a System
Putting It All on the Table
Know the Drill
It’s Fun To Work With an F-M-E-A
Solve Problems With Open Communication
Tell Me About It
Separate the Vital Few From the Trivial Many
To DMAIC or Not to DMAIC?
Breaking It Down
1 + 1 = Zero Defects
Curve Your Enthusiasm
Make a Choice
What Is a Fault Tree Analysis?
Successful Relationship Diagrams
The Benefits of PDCA
Return on Investment
The Art of Root Cause Analysis
Why Ask Why?
Get to the Root of It
Checks and Balances
Clearing SPC Hurdles
Supplier Selection and Maintenance
Building a Quality Team
Plan Experiments to Prevent Problems
Quality Progress - January 2016