Measurement resolution and a specified requirement

A fictional design team receives a component drawing with a stated acceptance interval and an instrument report. One instrument displays more decimal places than another, but its calibration record is missing. Repeated readings from the first instrument cluster tightly around one value, while a reference check suggests a possible offset. The team is tempted to select the instrument with the finest display and release the component immediately. A reviewer asks whether repeatability, resolution, bias and uncertainty have been confused. The exercise gives no actual machine, calibration tolerance or industry acceptance rule. Students must explain what each observation can establish and propose a traceable decision process. The aim is not to prescribe a universal rejection threshold, but to relate evidence to a supplied requirement and identify conditions that prevent a confident acceptance decision.