This thesis proposes a framework for generating operating curves for semiconductor manufacturing facilities using a modular flexible discrete event simulation (DES) model embedded in an application that automates the design of experiments for the simulations. Typically, operating curves are generated using analytical queueing models that are difficult to implement and hence, can only be used for benchmarking purposes. Alternatively, DES models are more capable of capturing the complexities of a semiconductor manufacturing facility such as re-entrancy, rework and non-identical toolsets. However, traditional craft-based simulations require much time and resources. The proposed methodology aims to reduce this time by automatically calculating the parameters for experimentation and generating the simulation model. It proposes a novel method to more appropriately allocate simulation effort by selecting design points more relevant to the operating curve. The methodology was initially applied to a single toolset model and tested as a pilot case study using actual factory data. Overall, the resulting operating curves matched that of the actual data. Subsequently, the methodology was applied to a full semiconductor manufacturing facility, using datasets from the Semiconductor Wafer Manufacturing Data Format Specification. The automated framework was shown to generate the curves rapidly and comparisons against a number of queueing model equivalents showed that the DES curves were more accurate. The implications of this work mean that on deployment of the application, semiconductor manufacturers can quickly obtain an accurate operating curve of their factory that could be used to aid in capacity planning and enable better decision-making regarding allocation of resources.