improve-control (DMAIC) method. The elements of each
step of the DMAIC method are shown below:
(1) Define phase: The define phase establishes the
rationale for a Six Sigma project. This requires
understanding the relationships between suppliers-
inputs-process-outputs-Customers (SIPOC), gathering
and analyzing “Voice of the Customer” data (that is,
identifying the issues or concerns important to
customers, called critical-to-quality (CTQ) variables,
and preparing a business case (rationale for doing the
project) with a project charter.
(2) Measure phase: The measure phase involves
studying and understanding the CTQs. This requires
developing operational definitions for each CTQ variable
(develop definitions for each CTQ that have the same
meaning to all users of the definition), performing a
Gauge Repeatability and Reproducibility (R&R) study
for each CTQ (determine if the measurement system is
appropriate for the needs of the study), and establishing
baseline capabilities for each CTQ. Additionally, the
measure phase involves determining key measures for
upstream suppliers, inputs, and processes, and collecting
baseline data for those measures, if they exist.
(3) Analyze phase: The analyze phase involves
identifying the upstream Xs for each CTQ, operationally
defining each X, performing a Gauge R&R analysis for
each X, establishing a baseline for each X, controlling the
Xs for each CTQ, identifying the major noise variables
for each CTQ, and understanding the effect of the Xs
on each CTQ. Data mining and screening experimental
designs help determine the “vital few” Xs for each CTQ.
(4) Improve phase: The improve phase optimizes the
relationship between the CTQs and “vital few” Xs. This
requires designing experiments to understand the relation-
ship between CTQs and high risk Xs and major noise
variables, generating the actions needed to implement the
optimal levels of the “vital few” Xs that optimize the
spread, shape, and center of the CTQs, developing action
plans, and conducting pilot tests of the actions.
(5) Control phase: The control phase involves
locking-in the improvements from a Six Sigma project
and transferring them to the process owner. This requires
avoiding potential problems in Xs with risk management
and mistake proofing, standardizing successful actions in
respect to the Xs and CTQs by developing, documenting,
and implementing process control plans for all high risk
Xs and CTQs. Additionally, the control phase involves
institutionalizing and leveraging successful pilot tests
with other areas in the organization, and transferring
ownership of the improved process, products, or services
to the process owner.
It is a common practice to teach Six Sigma Black Belt
training in 4 sessions of 5 days each separated by
3 weeks. The first week session covers the Define and
Measure phases of the DMAIC model. The second week
session covers the Analyze phase of the DMAIC model.
The third week session covers more of the Analyze phase
and the Improve phase of the DMAIC model. The fourth
week session covers the Control phase of the DMAIC
model and future steps. Green Belt training is covered in
2 sessions of 5 days each, separated by 3 weeks. This
article focuses on the Define and Measure phases of a Six
Sigma case study appropriate for the first week of Six
Sigma Black Belt training.
This article distinguishes between Black Belt and
Green Belt Six Sigma projects on the basis of five
criteria. Green Belt projects tend to be less involved
(e.g., they have one CTQ and few Xs), do not deal
with political issues, do not require many organiz-
ational resources, do not require significant capital
investment to realize the gains identified during the
project, and utilize only basic statistical methods. On
the other hand, Black Belt projects tend to deal with
more complex situations that may involve two or more
CTQs and many Xs, may involve substantial political
issues, or are cross-functional in nature, require
substantial organizational resources, may need sub-
stantial capital investment to realize the gains made
during the project, and utilize sophisticated statistical
methods. Candidates for Green Belt training are
individuals who are able to dedicate approximately
25% of their time to project work. Often the project
work is focused on processes within or related to the
area in which they currently work. Ideally Black Belt
candidates are those who will be able to dedicate 100%
of their time to one or more Six Sigma projects.
This article assumes that the reader is familiar with all
the tools and methods discussed in the paper, for
example, Quality Function Deployment (QFD), Kano
surveys, Gauge R&R studies, capability analysis, control
charts, probability distributions, to name a few. Readers
unfamiliar with the tools and methods discussed are
referred to Breyfogle,
[1]
Gitlow,
[2]
and Gitlow, Oppen-
heim, and Oppenheim.
[3]
BACKGROUND OF PAPER
ORGANIZERS INTERNATIONAL
Purpose
The purpose of this article is to present the define and
measures phases of a fictitious application of the DMAIC
Rasis, Gitlow, and Popovich128