A practical methodology for analysing and improving the measurement system

Measurement systems are used widely in manufacturing organisations in order to make many important decisions. These decisions range from the acceptability of a given product against tolerance requirements to the level of statistical control of a process or its capability to consistently perform a given task. In most situations, however, little thought is given to the quality of the data generated by such measurement processes. By using potentially flawed data in making fundamental manufacturing decisions, organisations undermine the quality of the decision-making process itself. This paper reviews the approaches presently available from both a technical and a practical point of view, based on the priorities of process improvement practitioners. The conclusion of the paper is that the Evaluation of the Measurement Process proposed by Wheeler and Lyday offers the best balance of accuracy and utility.