Overall Equipment Effectiveness (OEE) is a productivity performance metric widely used in industry to support production control decisions. However, there is still a gap in organisational procedures to systematically identify and address the most promising opportunities to improve the production setup. In this study, we propose and demonstrate a data-driven approach for increasing OEE by combining the strengths of discrete-event simulation with data analytics tools and methods, which provides a risk-free test environment that forms the basis for data-driven decisions and supports revealing production interdependencies. Therefore, this approach eases the process for practitioners to proactively identify production losses and forecast the outcome of the most promising selected improvement measures. A case study is performed to illustrate the potentialities of the proposed approach, demonstrating the interdependence between the processes and the improvement measures, and the knock-on effect both upstream and downstream. The results yield substantial insights and facilitate operational decisionmaking for managers.