With an increasing proportion of elderly and an increasing demand for healthcare, managerial efforts are needed in order make the best use of resources and to keep cost under control. One of the most critical and expensive resources in a hospital is the operating theatre. This thesis aims to investigate the potential of computer-based modelling for supporting healthcare decision makers to improve management policies related to the hospital operating theatre. In a study conducted at a medium sized Swedish hospital we identify important prioritisations and decisions made in relation to patient scheduling and resource allocation when planning for surgery. Patient scheduling and operating room planning are complex tasks with a number of influencing factors to consider like, e.g., uncertainty in patient arrival, uncertainty in surgery procedure time and medical prioritisations and diagnosis. Further, several intersected dependencies, e.g. pre- and post operative care, have to be considered as to prevent occlusion and obtain a maximum patient through-put. With an optimisation-based approach we demonstrate how different criteria in patient scheduling and resource allocations can affect various objectives in terms of patient perspectives, staff perspectives and costs. For instance, we show that the current policy for resource allocation does not handle the variability generated by the patient diagnosis very well. In Sweden a law has recently been introduced, which advocates restrictions in elective patient waiting times. We extend the optimisation-based approach to include post-operative care and simulate a scenario based on patient data from a Swedish hospital to be able to predict the possible impact of the new law. The results indicate that the law causes an unsuitable increase in the waiting times for medium prioritised patients. Furthermore, we propose a combination of discreteevent simulation and optimisation to examine what impact different resource allocations of emergency and elective resources have on both utilisation rate and disturbance consequences, i.e. surgery cancellation and overtime work, due to emergency cases and other unexpected events. We show that both utilisation rate and cancellation frequencies can be improved significantly by the application of some minor changes in the resource allocation. Finally, we explore some future possibilities of using agent technology for modelling health care management decisions.