Parameter selection in non-traditional machining processes using a data mining approach

With ever-increasing demands for high surface finish and complex shape geometries on various difficult-to-machine materials, conventional metal removal methods are now being replaced by non-traditional machining (NTM) processes. These NTM processes use energy in its direct form to remove material from the workpiece surface. They are also cost effective for a wide range of micro- and nano-level applications. For effective utilization of different NTM processes, it is quite important to study their characteristics and material removal mechanisms in order to identify the most significant control parameters affecting the process responses. In this paper, a data mining approach using classification and regression tree algorithm is employed to identify the most important input parameters of three NTM processes, i.e. micro electro discharge milling process, wire electrical discharge machining process and laser beam machining process. The derived observations are also validated using the analysis of variance results.