Reliability analysis has long been used to understand and predict system behaviors in various industries, including manufacturing, aerospace, and energy. However, the increasing complexity and dynamics of modern systems can quickly outpace manually developed, expert-based models. Conversely, the increasing availability of data from industrial Internet of Things (iIoT) sensors and advanced control systems enables a more data-driven approach to reliability modeling, coping with the aforementioned issues. In this paper, we introduce a framework for data-driven reliability assessment of manufacturing systems using process mining. With our framework, we aim to provide a systematic approach to extract, simulate, validate, and exploit reliability models to support decisions within manufacturing systems. We demonstrate our framework using two case studies based on a flow line commonly found in today’s shop floors.