An important part of capability studies when measuring the total variability in manufacturing is to account for gauge or test process variability. Variability is not limited only to the manufacturing process; the variability of the measurement system needed to test the manufactured parts should also be considered. Figure 5.6 shows typical sources of variation and error in a process and its measurement system. The majority of measurement errors, including those due to the operator (appraiser) or the equipment (gauge), can be measured and quantified through gauge reliability and reproducibility (GR&R) methodology. The use of GR&R to evaluate measurement systems quality is mandatory in achieving six sigma quality.
The following is an explanation of the terms used in Figure 5.6.
• Short and long variations in the manufacturing process are due to time-dependent parameters, such as incoming part changes, age of equipment, and methods for maintaining equipment, They will be discussed in the next section.
• Precision the relative amount of the variability of the measuring system; hence，it is an indicator of the variability of the equipment(gauges).
• Accuracy is a relative measure of achieving the measurement target. It is the difference between the true and measured values, although the true value may not be known in many cases. Accuracy is usually referred according to some standard of measurement. Hence, it is an indicator of the measurement error average of μ of the equipment (gauges). Accuracy and precision are shown in Figure 5.7, using a target analogy.
• Repeatability is a measure of the consistency of readings of the same part for a single operator (appraiser). Poor repeatability indicates measurement system problems related to equipment. Repeatability is derived from the same operator measuring different parts repeatedly using the same measurement equipment. Sometimes it is called precision or equipment variation (EV).
• Reproducibility is a measure of variation in average measurements when different operators are taking many measurements of the same part. It can used as measure of the relative amount of training or skills for the operators. Sometimes it is called appraiser variation (AV) using the same parts and gauges and different operators.
• GR&R is the root sum of the squares (RSS) value of repeatability and reproducibility. It should be noted that the average or the process and measurement errors add up algebraically, whereas the standard deviations add up in the squares, as shown in Figure 5.8