Sampling strategy design for dimensional measurement of geometric features using coordinate measuring machine

Dimensional measurement using a coordinate measuring machine (CMM) has been commonly adapted in advanced manufacturing environments to ensure that manufacturing products have high quality and reliability. To conduct dimensional inspection effectively in a computer-integrated manufacturing (CIM) environment there is an urgent need to derive a sampling strategy which can be used to specify a set of measuring points that lead to accurate sampling while minimizing the sampling time and cost. Owing to the variations in characteristics of geometric features and manufacturing processes, different feature surfaces on a workpiece usually have different variations in their dimensional accuracy and surface finish. The variations may differ considerably from one surface to another, even though those surfaces may share the same feature. Therefore, the variation in dimensional accuracy and surface finish should be considered in determining the proper sampling size for each geometric feature generated by various processes with different production parameters. In this paper, a feature-based methodology which integrates the Hammersley sequence and a stratified sampling method are developed to derive the sampling strategy far various geometric features which have specified measuring points. Case studies are used to compare the effectiveness of Hammersley sequence sampling, uniform (systematic) sampling and random sampling. The results show that the derived sampling strategy based on the Hammersley sequence leads to a nearly quadratic reduction in the number of samples compared with the uniform sampling method, and hence units of time and cost, while maintaining the same level of accuracy. The derived sampling strategy also shows a better performance when compared with the random sampling method. (C) 1997 Elsevier Science Ltd.