Situation Edward is the quality manager for a small company. He has just completed a training session at which he learned several approaches for analyzing data from a measurement study. Because his company is anxious to verify that the measurement techniques they use are sound, Edward studied the one that is used frequently. The technique is used by four operators— two per shift. Edward hopes to assure company management and its customers that these measurement methods are reliable. Available data Edward selected three parts and a height dimension to measure. All four operators measured the height of each part twice at different times. The measurements are recorded in the table, “Measurement Study.” Specifications for this height are 20 ± 10 thousandths of a unit. The data are recorded in tenths of a thousandth of a unit. Questions 1. What is the variability (test retest errors or repeatability) because of the measurement technique? 2. Do the operators report comparable answers? If not, which operators differ from the others? 3. Is this measurement technique adequate for measuring the characteristic under study? 4. What actions should Edward take based on the information from the measurement study? How Good is my Measurement Process: Part II Situation As part of his job as quality manager, Edward evaluates the measurement techniques used by his company. In his first study of these methods, presented in the August issue of Quality, Edward discovered that two of four operators were making the measurements for the height of a part in a manner that gave different results from the others. Edward instituted a review of the measurement technique and an ongoing training session for all operators. Several weeks later, he repeated the measurement study to see if the operator differences had been eliminated. Available data For the follow-up study, Edward used the same three types of parts and the same height dimension as the first study. All four operators measured the height of each part twice at different times. The measurements are recorded in the table, “Measurement Study, Part II.” Specifications for this height are 20 ±10 thousandths of a unit. The data are recorded in tenths of a thousandth of a unit. Questions 1. Is there evidence in these data that the training program for operators was successful? 2. What is the current test-retest error or repeatability of the measurement process? 3. Is this measurement process still adequate for detecting differences in the parts?
Published by QualityMagazine. View All Articles.