Meaning
Collection of data in small clusters of items produced under nearly identical conditions helps in identifying variations that occur over time. Using subgroup sampling allows a quality technician to separate the variation within a single batch from the variation between different batches. This technique is a fundamental part of creating accurate control charts for a production line.
The data collection stops once a sufficient number of clusters have been analyzed to establish a baseline.
Rational Selection
Choosing items that are made close together in time ensures that any differences found are due to long-term process shifts. Proper subgroup sampling avoids mixing parts from different machines or operators into the same data set. This isolation makes it easier to find the root cause of a sudden change in product dimensions.
Internal Variation
Measuring the range within a small group of parts provides an estimate of the short-term capability of the equipment. If the subgroup sampling reveals high internal variation, the problem is likely located within the machine itself, such as a loose fixture or a worn tool. Low internal variation suggests that the machine is stable even if the overall process is drifting.
The consistency of these small groups defines the lower boundary of process improvement.
Monitoring Interval
Timing the frequency of checks involves a trade-off between the cost of inspection and the risk of producing scrap. Regular subgroup sampling provides a continuous stream of information that alerts the team to gradual changes in material quality or environmental conditions. This oversight is necessary for high-volume automated assembly.