The concept of computer integrated manufacturing has been the primary focus of research in the manufacturing community, to achieve high product quality and productivity. The need to continually improve on manufacturing processes necessitates the identification and modeling of all factors that affect product quality in the manufacturing process. The sources of machine tool vibration during the machining process is one such set of significant factors that is of interest.

This thesis work attempts to explain the effect of the non-homogeneous distribution of workpiece material microstructure on machining quality. The main contributions of this thesis are (1) that two stochastic models, using the concepts of sample variance and Markov chains, are used to simulate the microhardness distribution, (2) that an algorithm is developed for the calculation of the sample shape function, critical to the sample variance model, and (3) that a computer simulator has been built to simulate the turning process, which takes the cutting conditions and the microstructural image of the material as its inputs, and evaluates the machining performance. The validity of the models in describing random excitation has been confirmed by comparison of dynamic force values obtained from experiments and from simulation.

This integrated approach of analyzing the machining process provides the production engineer with a tool for estimating the optimal cutting conditions for a given material. It also provides the design engineer a method of selection of materials based on their ability to be machined to a required surface finish.