Gordon P. Constable And Eric Gasper 2014-06-21 02:47:02
Simply buying a program that keeps records of gage calibration does not constitute a robust measurement system. Moving from a 3 x 5 card system to more sophisticated measurement system analysis requires more than a purchase order. Simply buying a program that keeps records of gage calibration, for example, does not constitute a robust measurement system, since such a system demands an analysis of analytical methods that will garner genuine information rather than only generating more data. One methodology that supports this generation of information is the use of the gage R&R study, which will indicate whether the measurement system in use can adequately distinguish between or among units. The first “R” stands for repeatability, the variation seen in multiple measurements by one operator with the same gage on the same part or sample characteristic. It is linked to the equipment variation of the measurement system. Reproducibility is the second “R” and it is the variation between the averages of different operators when using the same gage, measuring the same part(s), and using the same measurement procedure.This represents the operator or appraiser variation. A number of factors affect the ability of a measurement system to discriminate among the units it measures. These may be categorized as those that typically contribute to any process variation: • Machine (the gage) • Manpower (appraiser) • Method (method of measurement followed) • Material (units being measured) • Environment In conducting a simple study, an attempt is made to minimize or hold most of these factors constant. (Note: the study should replicate as much as possible the conditions under which the process measurement is occurring. This means that if production operators are making the measurements, the study should use production operators under the same conditions that exist when they are “normally” measuring the output.The study should not be conducted using quality technicians in a lab.) Therefore: • One gage is selected for the study.(No inter-gage variation.) • One method of measurement or procedure is employed by all appraisers.(No variation generated by appraisers using different methods.) • The same dimension on each part is measured each time. A further Assumption is that each part is measured in the same place to eliminate the possibility of within-part variation.(No variation introduced because of different characteristics or because the measurements have been taken at different locations on the part.) • The study is conducted under the same conditions that exist when the parts are normally measured. (No variation introduced by the location or environmental concerns of where the measurements are taken.) Also, a coordinator or administrator (usually a member of the organization’s quality department) will prepare, organize and conduct the study. This employee will often be responsible for analysis of the results and creation of action plans to improve the measurement system if necessary. 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. 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 four or five, use two or three “typical” appraisers. The more appraisers added to a study, the greater the time, cost and complexity of the results. Generally, the number of parts is between five 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 neither going down to the line and pulling five to 10 consecutive parts nor going to a bin and selecting 5 to 10 parts is 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 two or three.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, measurements should be taken randomly (see note below). This separates 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 that 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 Below is an example of data for a study with three appraisers, two replications, and five parts. After the measurements have been taken and gathered, basic calculations can be performed. These calculations will include averages and ranges for each part, replication and appraiser, as well as an overall part average and overall range average for the appraisers.These results will be base numbers used to calculate the repeatability and reproducibility. Several techniques are available for computing the study results. Each organization must determine which method to use. In many cases this is determined by outside influences, such as customer requirements or a set of governing standards that must be adhered to for compliance. One method is the Measurement Systems Analysis (MSA) which is governed by the Automotive Industry Action Group (AIAG). Their method can calculate results three ways: • Percentages of total variation based on the study parameters • Percentages of the specification range • Percentages of total variation based on process parameters identified by an X-bar and range chart maintained on the production process of the parts used in the study The EMP III (Evaluating the Measurement Process, 3rd edition) can also calculate repeatability and reproducibility, but this method does not rely on R&R percentages to determine a measurement system’s capabilities. Instead, it places more emphasis on the interclass correlation component to determine whether a system can be used for critical measurements. Lastly, ANOVA (analysis of variance) can also be used to compare averages. In the context of evaluation of measurement systems, this relates to measurement of dispersion or spread while identifying any statistically significant background variation. Each technique uses the same base numbers gathered during a study but each method’s results can highlight different elements of variation in a measurement system. Conducting gage R&R studies can help transition an organization from a place that just gathers data to one that embraces their quality goals and continuously strives for improvement. And is actually a whole lot more efficient that those old 3 x 5 cards. Gordon P. Constable, Ph.D., recently retired as a PQ Systems trainer/consultant in MSA and GAGEpack.Eric Gasper is lead trainer, measurement systems analysis for PQ Systems. For more information, call(800) 777-3020, email firstname.lastname@example.org or email@example.com, or visit www.pqsystems.com.
Published by QualityMagazine. View All Articles.