Beruflich Dokumente
Kultur Dokumente
A number of factors affect the ability of a measurement system to discriminate among the units it measures. These factors can be categorized generally into those that affect central location and those that affect the variability (spread) of the measurements. Variability factors measured by repeatability and reproducibility are the more familiar, while factors related to the central location of the measurements (stability, bias, and linearity) are relatively new approaches. Both approaches may need clarification. In addition, methods of measurement must be developed along with standards for indicating their acceptability. In a series of articles, these concepts will be addressed. The first of these will deal with repeatability and reproducibility and combining them into the R&R component. The next article will take this R&R component and calculate R&R percentages based on study variation, process variation, and tolerance.The Measurement Study (classic) typically utilizes one to three appraisers for one measuring instrument that is measuring a single characteristic. Each appraiser measures five to ten units selected from a process two or three times (replications). Before proceeding with the analysis of the study, the ranges for the replications of the measurements made by each appraiser on each part are determined and used to calculate control limits for the range chart. Then each range is checked to determine if it falls inside the limits. Those measurements that result in a range outside the limits should be excluded from further analysis or should be redone. Operative assumptions include: 1. 2. 3. The measuring instrument stays in calibration (central location does not change); Appraisers use the same method of measurement; Parts are measured in the same place. (If the assumption that the parts are measured in the same place is incorrect, the possibility of within-part variation will need to be considered.)
Repeatability refers to the variation in measurements for one characteristic made with one measuring instrument by one appraiser on the same part. An estimate of repeatability is obtained by first determining the average range ( ) of the repeated measurements of the same characteristic, using the same measuring instrument for several parts. Note: if more than one appraiser is used in the study, the average range is the combination for all appraisers, e.g., [If there are three appraisers, a, b, and c, you would determine the average range using ( = a + b + c )/3]. Next, the standard deviation for the repeatability (se) is estimated by dividing by d*2. A 99% (-2.575 < z < +2.575) interval for repeatability is determined by multiplying 5.15 by (se).Reproducibility refers to the difference in the average of the measurements on one characteristic made by different appraisers using the same measuring instrument on the same part(s). Note: if there is only one appraiser using the gage, there will be no reproducibility (appraiser variation). Again, the assumptions are that the instrument stays in calibration, the appraisers use the same method of measurement, and the part is measured in the same place. An estimate of reproducibility is obtained by determining the mean of all the measurements made by each appraiser e.g., [If there are three appraisers, a, b, and c, you would determine the average range using ( = a + b + c )]. The range estimate for the operators (Ro) is obtained by subtracting the minimum i from the Maximum i.Next, the standard deviation for reproducibility (so) is estimated by dividing (Ro) by d*2. Again, a 99% (-2.575 < z < +2.575) interval for repeatability is determined by multiplying 5.15 by (so).Since the measuring instrument is used in making the measurements, it is a contributing factor to the calculation of reproducibility. Therefore, the calculation of reproducibility needs to be adjusted by subtracting a portion of repeatability. The adjusted appraiser variation is given by:
Where: n = number of parts used in the study r = number of times each part is used
R&R is the combination of repeatability and reproducibility variation and frequently is considered as the total measurement variation excluding within part variation and variation in central location. R&R studies can be done easily and accurately using software products like GAGEpack. Gordon Constable
The two items that are specifically varied are the number of appraisers and the number of units measured. The selection of these two factors is critical to meeting the objective of an R&R study. The first one addressed is the number of appraisers. The appraisers should be selected from those individuals who are currently measuring the output. If there is only one appraiser, then only one should be used in the study. If there are two or three, use two or three. If there are more than 4 or 5, use two or three "typical" appraisers. The second is the number of parts, generally between 5 and 10. The major factor to consider in selecting the parts is that they need to represent the total variation for the characteristic being studied for the process producing them. This means that going down to the line and pulling 5 to 10 consecutive parts or going to a bin and selecting 5 to 10 parts, is a not acceptable. The selection of the parts must be spread out over time so that the parts represent the "typical" variation in the process. The final parameter is the number of replications, generally 2 or 3. A single replication will not allow for separate estimates for equipment variation and appraiser variation. Too many replications complicate the study without adding much value to the analysis. To be statistically sound, take measurements randomly (see note below). This confounds the factor of time, different in these repeated measurements, from appraisers and parts. As a practical matter, it often comes down to making sure that the appraiser does not know which part is being measured (randomizing the order of measurement) and no appraiser completes two replications before the others finish one replication. For example: Appraiser 1 measures all five parts once. Appraiser 2 measures all five parts once. Appraiser 3 measures all five parts once. Appraiser 2 measures all five parts a second time. Appraiser 3 measures all five parts a second time. Appraiser 1 measures all five parts a second time. Note: Do not always measure parts in the order of 1-2-3-4-5-6-7-8-9-10 This results in something like Table 1 below once the data are organized. Appraiser John John Mary Mary Gerry Gerry Part 1 2.112 2.114 2.112 2.112 2.108 2.110 Part 2 2.112 2.106 2.117 2.107 2.116 2.124 Part 3 2.071 2.073 2.084 2.082 2.102 2.086 Part 4 2.101 2.097 2.098 2.099 2.110 2.104 Part 5 2.114 2.130 2.110 2.108 2.108 2.093
If you enter this type of information in GAGEpack as an R&R event, GAGEpack will calculate R&R percentages. A free 30-day trial of GAGEpack is available at pqsystems.com. Gordon K. Constable Ph.D
For percent of study, the process variation is based on the spread of the parts (P) determined by . This is considered a range and using the /d2 relationship, a sigma for the process is estimated. This is then used to calculate the percentages. A second method is to use the spread of the specs (USL - LSL). Now this must be compared to the estimate of the measurement error (R&R). However, one needs to multiply the sigma of the measurement by 5.15 (old method) or by 6.0 (new method) to compare the total measurement spread with the spec spread. (An alternative method is to divide the spec range by the respective numbers given above.) The third method uses the information from an the , the chart on the process and characteristic being studied. In this case, enter
, and the sample size. This is used to estimate the process spread.
Ideally the measurement error (R&R%) is less than 10% of whatever method is used (process spread or spec spread). It is usable in some cases when the R&R percentage is between 10 and 30%. More than 30% suggests that one should not be using it. [page 60 of MSA Manual 2nd edition or page 77 of MSA Manual 3rd edition] If the number of distinct categories is 5 or more, it can be considered a capable measurement system. Wheeler and Lyday use a concept closely aligned with distinct categories called discrimination ratio, for which greater than four is satisfactory. The differences, in a nutshell, are that the distinct categories is a truncated number (no rounding or fraction used) and the discrimination Ratio assumes that appraiser variation has been reduced to zero and carries the fractional part as well. Gordon K. Constable Ph.D
range of the instrument. The accuracy of those parts is determined by the difference between the master measurement and the observed average measurement. The accuracy of these parts can be determined by plotting the accuracy values from the smallest size (closed position) to the largest size (open position). The linearity of the equipment is represented by the slope of a "best fit" line through these points. This best fit line is determined by using least squares regression. If equipment demonstrates non-linearity, one or more of these conditions may exist 1. 2. 3. 4. Equipment not calibrated at the upper and lower end of the operating range; Error in the minimum or maximum master; Worn equipment; Possible review of internal equipment design characters.
Product and process conformance are determined by measurements that are taken by a measurement system. Errors in these measurements have a direct bearing on conformance as defined within the system. A clear understanding of the results of the measurement system requires an understanding of the possible error within the system. To understand this error, one needs to understand the terminology, and in particular the concepts of stability and linearity. Both stability charts and linearity plots can easily be accomplished using software products like GAGEpack. Gordon Constable