Data mining techniques applied to predictive modeling of the knurling process

Abstract

Knurls are designed into a product to provide the correct frictional force for easy assembly and maintenance and sometimes for decorative purposes. The literature to date has merely studied how to realize a good and consistent knurl, but no predictive models of the knurling process have been presented. This paper applies two competing data mining techniques, regression analysis and artificial neural networks, to develop a predictive model of the knurling process. Fractional factorial design of experiments is used to plan the experiments. Four criteria, namely the PRESS statistic, the adjusted R ², the C _p statistic, and the residual mean square s ², are employed to select the best regression model. Hypothesis testing is conducted to test the effectiveness of each model, and to compare the two data mining schemes. This study demonstrates that for a reasonably large set of data from structurally designed experiments, the two methods produce comparable results in both model construction (or training) and model validation. Due to the explicit nature of a regression model, it is preferred to a neural network model to investigate the process.

Acknowledgements

This research has been partially supported by the Bradley University GRASP Award, Research Excellence Award number 13-26-233 and Hueser Research Award number 25-13-757 granted to Dr. Jack Feng. The authors are grateful to the reviewers for their valuable comments. Mr. Chinh Tran, Senior Manufacturing Engineer at Applied Materials in Dallas (formerly a Manufacturing Engineer at ElecSys Inc. located in Peoria, Illinois and a graduate student of Dr. Feng), developed the NC codes and collected the data.