Any machining operation induces significant deformation and associated stress states within the component being machined. Once the component has been finished and is removed from the machining tool, a portion of these stresses remain within the finished component, and are termed residual stresses. These stresses have a significant effect upon the performance of the final component. However, despite their importance there is no accurate and cost effective method for measuring residual stresses. For this reason predicting these stresses without the need for measurement is highly desirable. The focus of this thesis is on advancing the development and implementation of finite element models aimed at predicting residual stresses induced by metal cutting operations. There are three main focus areas within this research, the first of which is concerned with predicting residual stresses when small feed rates are used. It is shown that in the existing cutting models residual stress prediction accuracy suffers when feed rates are small. A sequential cut module is developed, which greatly increases the accuracy of the predicted residual stress depth profiles. A second area of focus concerns the influence of friction models on predicted residual stresses. A detailed set of simulations is used to elucidate the effect of friction not only for sharp tools, but also for tools which have accrued wear. It is shown that whilst friction is not of critical importance for new tools, as tools continue to wear the choice of friction model becomes significantly more important. The third area of focus is on phase transformations, induced by the cutting process. A decoupled phase transformation module is developed in order to predict the depth, if any, of a phase transformed layer beneath the newly machined surface. Furthermore, the effect of this layer on the residual stress depth profile was also studied. All three focus areas present new and novel contributions to the field of metal cutting simulations, and serve to significantly increase the capabilities of predictive models for machining. / Thesis / Doctor of Philosophy (PhD)