The thesis examines the integration of high level process planning and optimization activities with the normal processes of interpolation and control. The normal approach to part manufacture starts with process planning and proceeds to the programming of tool paths and the actual cutting on the machine tool. The process planning system itself is hierarchical in nature, and many cost or process parameters must be estimated at the higher levels. It is usual, because of the lack of appropriate values for all process parameters, to be somewhat conservative in the planning process. It is also unfortunately usual to discard much of the technological data derived during the planning in the process of producing the standard input format required for CNC machine tools.

The work carried out has shown that there are considerable advantages to the integration of these levels of the manufacturing process. Most importantly, information regarding part geometry and estimated process constants may be passed to monitoring tasks within the CNC, and calibration tasks within the CNC system can pass updated values of actual part parameters back to the planning system for further analysis. Clearly, however, such integration allows much further advantage, and in this regard the possibility of dynamic process planning appears most promising. The concept of dynamic process planning is the provision of the capacity for the machine to reprogram itself in response to feedback from the monitoring tasks. Typical examples of dynamic planning would include the recovery from tool breakage or, more typically, the replanning of paths within a feature when more or less material than expected is found. (Recovery from tool breakage requires replanning of the remaining feature without the current tool).

The integration of activities on a machine tool requires the presence of an adequate monitoring system, reliable sensors, and suitable models to relate the measured parameters to part geometry and tool condition. The author has developed basic models of cutting force and methods for identifying runout or breakage in milling. The work directed toward runout has mainly considered radial runout;​ however, simplified models have also been proposed to identify axial runouts. Finally, in order to allow the measurement of force on practical workpieces, the author has designed a simple robust and low cost table dynamometer.