means Define, Measure, Analize,
Improve and Control.
It is
structured, disciplined, rigorous approach
to process improvements consisting of the
five phases mentioned, where each phase
is linked logically to the previous phase
as well as to the next phase. The reason
to follow this method is to achieve the
Six Sigma level.
The project purpose and scope are defined.
The goal for this phase is to set project
goals and boundaries based on your knowledge
of your organization's business goals, customer
needs, and the process that needs to be
improved to get you to a higher sigma level.
In this phase the effort is gathering information
for the current situation, so you can pinpoint
the location or source of problems by stratified
the data in the baseline performance, as
precisely as you can by building a factual
understanding of existing process conditions
and problems. That knowledge will help you
narrow the range of potential causes you
need to investigate in the Analyze phase.
You should be able to have your baseline
for capability level.
The focus of the analyze phase is develop
theories of root causes, confirm them with
data by testing using sophistic statistics
tools as hypothesis, correlation, regression
and others, and finally identify the root
cause(s) of the problem. The verified causes(s)
will form the basis for the solutions in
the next phase. The main question is “What
vital few input variables affect critical-to-quality
(CTQ’s) process performance or output
measures?”
Now
the activities are around develop, implement
and evaluate solutions targeted at your
verified causes. The goal is to demonstrate,
with data, which solution solve the problem
and lead to improvement. The most common
tool used in this phase is the design of
experiments (DOE) that are running as test
or pilots on the process. The result is
a detailed plan implementation ongoing.
Putting
a solution in place can fix a problem for
the moment, but in this phase the goal is
designed a monitoring system to help the
process owner maintain the gains by standardizing
the process and show it stable and that
the new methods can be further improved
over time. Control Charts are one of the
most usefully tools for that, and a complete
documentation of results, learning and recommendation
is delivering.
|