In the process planning for pocket machining, selection of the optimal tool sizes and minimizing the number of plunging operations are among the most important factors in minimizing the machining time.This thesis presents a new approach for optimal tool selection of arbitrary shaped pockets based on a polygon subdivision technique. The pocket is subdivided to obtain smaller sub-polygons. The tools are selected separately for each sub-polygon and then the optimal set of the tools for the entire pocket is obtained based on minimizing both the machining time and the number of tools used to machine the pocket. Finally, the subpolygons are sequenced in an optimal order to eliminate the requirement of multiple plunging operations.The approach presented is an improvement over previous work because it makes an effective use of the polygon subdivision strategy to improve the machining time as well as reducing the number of plunges. The implementation examples of this approach suggest that the machining time can be improved as much as 75%.