STOCHASTIC APPROACH TO EXPERIMENTAL-ANALYSIS OF CYLINDRICAL LAPPING PROCESS

Lapping is a very complicated and random process resulting from the va riation of abrasive grains by its sizes and shapes and from the numero us variables which have an effect on the process quality. Thus it need s to be analyzed by experiment rather than by theory to obtain the rel ative effects of variables quantitatively. In this study, the cylindri cal lapping experiment designed by Taguchi's L. orthogonal array was p erformed and analyzed by Yates' ANOVA table. As a result, effective va riables and interaction effects were identified and discussed. Also th e optimal variable combination to obtain the largest percentage improv ement of surface roughness was selected and confirmatory experiments w ere performed.