Experimental verification of manufacturing error pattern and its utilization in form tolerance sampling

This paper presents an adaptive sampling procedure, which uses manufacturing surface error patterns and optimization search methods for reducing sample size, while maintaining high accuracy. The methodology is demonstrated with straightness and flatness evaluation. Two manufacturing processes, end and face milling are used to fabricate plates used for modeling and subsequent testing. Respective surface errors for each process are quantified and validated using previously published models. The initial points for sampling are identified through such a characterization of the process and its affect on the workpiece. These sampled points are fit using the least-squares method to complete the form evaluation. Points are added to the initial set using optimization search heuristics. The final tolerance value obtained is compared with that obtained from a large population sample to check the accuracy. With such an adaptive approach, it is proposed that the number of points sampled is potentially less than that which would be expected to achieve the same level of accuracy using traditional sampling methods. This paper demonstrates the error modeling and validation aspects of this adaptive sampling procedure.