A project is a temporary endeavour with a specific goal (e.g. the construction of a railway station) and consists of several components (i.e. activities, a project network and resources). Often, projects do not meet their goals or suffer from delays or cost overruns. Therefore, the project manager should control his/her projects by monitoring the progress and, if the observed progress is not acceptable, take corrective actions on the components to get the project back on track. Due to uncertainty during project execution (e.g. bad weather conditions or employee illness), deciding on the right corrective actions is not an easy task. However, selecting the most effective corrective actions is crucial to avoid unnecessary additional costs and to maximise the likelihood of project success.

The aim of this proposal is developing improved methods to select the most effective corrective actions on the different components by explicitly considering the uncertain outcome of these actions. To that purpose, a well-balanced mix of artificial and empirical data will be used. Artificial data and Monte Carlo simulation will be used to evaluate the novel methods in a controlled environment, while empirical data (i.e. progress data of more than 100 real-life projects completed in recent years) will be used to validate the applicability and effectiveness of the approaches in a real-life context. Ultimately, the proposed methods should aid the project manager to deliver projects successfully.