Page 198 - Six Sigma Demystified

178 Six SigMa DemystifieD

• Specification. What is the definition of good and bad for this characteristic?
• Measurement technique. How will the characteristic be measured? The
control plan should reference more detailed measurement procedures,
where needed, to instruct personnel on the proper use and interpretation
of the measurement equipment.
• Sample size. How many measurements are required at each point in time?

• Sample frequency. How often should a new set of measurements be taken?
• Analytical tool. How are the measurements evaluated? In this case, audio
alarms are installed to monitor characteristics thought to have high pro-
cess capability. Other characteristics are monitored via control charts to
detect process trends leading to undesirable conditions.
• Reaction rules. What is the proper response to the alarms (audio or from
control chart)? Who should be notified?

Table 8.1 Example Control Plan

Characteristic Specification Measurement Sample Sample Detection Analytical Tool
Technique Size Frequency Method
Finished 1.250 ± Johannson 5 1 subgroup X-bar If other
diameter 0.0002 chart characteristics
with run in control,
rules adjust
Local 70 ± 5 Thermocouple 1 Continuous Alarmed Stop
ambient production;
temperature adjust cooling
control;
resume
production
once
temperature is
okay
Cooling 120 ± 5 Thermocouple 1 1 sample/ Alarmed Stop
temperature min production;
adjust cooling
control;
resume
production
once
temperature is
okay
Coolant 0.88–0.94 Viscometer 1 1 sample/ X chart Purge system
viscosity month