When considering the capabilities of fast and smart connected networks of production systems through industry 4.0, predictive maintenance comes to mind. This approach pursues the goal of eliminating system failure before it occurs. This prevents systems and equipment entering unexpected downtimes and saves high costs. When dealing with flexible production lines, flexible solutions are required to plan production, including maintenance. The paper presents a concept and a first case study for a generic, predictive maintenance simulation in AnyLogic™. The concept contains descriptions of the function and the goal of the simulation, interfaces for data transfer to and from the simulation, the generic structure of the simulation, and the indicators used to evaluate the simulated production plans. The first case study demonstrates, how the simulation-model can be used to simulate example data from a testbed production to improve production plans.