Page 310 - The Handbook for Quality Management a Complete Guide to Operational Excellence

296 C o n t i n u o u s I m p r o v e m e n t M e a s u r e S t a g e 297

Once established, these metrics can be used to determine the relative
importance of various factors affecting processes and business units, as
well as comparing the various processes’, business units’, and competitors’
contribu tions to the overall success of the business. They also provide a
baseline to gauge the results of Six Sigma improvement efforts.
In short, metrics provide us with a statement of what is critical to quality,
cost, and/or scheduling; how these will be mea sured and reported; and how
these critical to quality (CTQ), critical to cost (CTC), and critical to schedule
(CTS) metrics correlate with key process vari ables and controls to achieve
system-wide improvements.
At the process level, you may establish a baseline by measuring the cur-
rent process capability. Other common metrics in Six Sigma are the Through-
put Yield, the Normalized Yield, and the Sigma Level for individual process
steps, and the Rolled Throughput Yield for complete processes.

Establishing Process Baselines
The performance of the existing process must be established via a baseline
to ascertain how well the current process meets the customer requirements
and validate the project justification noted in the Define stage. When the
process is in a state of statistical control, then its process capability index
may be calculated, as discussed in Chap. 9. The capability index compares
the calculated process variation with the stated customer requirements.
Process variation may be estimated only after process stability is established
using a control chart. Only stable processes can be predicted; unstable
processes are the combination of multiple processes, each occurring at
separate intervals in time.
When the process is unstable, or not in statistical control, a process
performance index may be used as rough estimate to compare observed
variation within a given sample with the customer requirements.
A control chart is critical at this stage so results of improvement
achieved later in the project can be reliably estimated. When improve-
ment efforts are carried out without the use of a control chart, it’s difficult
to prove that the improvement had any real effort, since any improvement
in result may have just been coincident with the sporadic nature of the
special cause.
When control charts are used to baseline the process, and the process
is out of control, the team may quickly discover the cause. If this special
cause is of the order of magnitude that justified the project, the project is
on the way to rapid conclusion. In any event, this is necessary information
for the Analyze stage, as common cause variation must be treated much
differently than special cause variation.
At the business level, we often seek to measure the customer’s percep-
tions of the current state of the product, product delivery, and/or service

14_Pyzdek_Ch14_p293-304.indd 297 11/20/12 10:32 PM