This work focuses on real-time production planning and scheduling problems that are encountered in modern manufacturing environments. More specifically, a simulation assisted optimization scheme is proposed that seeks to handle dynamic disruption events via robust production re-scheduling. The proposed scheme is tested using data from a real automotive manufacturing plant over a planning period of four months. The examined scheduling problem is formulated as a multi-period multi-model paced assembly line scheduling problem with component availability restrictions. On a daily basis the material replenishment and truck arrival plan at the plant is known; however, delays may occur. Additionally, quality problems may occur at the paint shop and this may affect the availability of specific car bodies. These component availability events may significantly disrupt the current schedule and the cost to fix them can be significant. An advanced meta-heuristic algorithm is proposed for solving the re-scheduling problem. We adopt a hierarchical objective function that seeks to minimize the re-scheduling cost and also to maximize the robustness of the schedule. For this purpose, simulation is used to evaluate the generated schedules and to identify critical components that are prone to delays. On return, this list of components is used by the optimization to measure the robustness of the schedules and to introduce time-buffers. Overall, results demonstrate the applicability, effectiveness and efficiency of the proposed framework in a real manufacturing environment.