With the ongoing digitalization in the industrial sector, also called Industry 4.0, automation techniques offer huge potential for the rubber industry. Advanced automation methods allow for precise control of the extrusion process and increase the consistency in product quality. Since product quality is affected by process parameters in a complex way, predicting the flow conditions, the thermal conditions and the mixing in an extruder is a prerequisite for more sophisticated Industry 4.0 applications. These Industry 4.0 applications require short computing times and thus sufficiently fast models for online implementation in production systems. A useful measure for the flow inside the extruder is the residence time distribution (RTD). It enables an analysis of the residence time of the material particle in the process and subsequently a statement about the mixing characteristics. In order to avoid cross-linking reactions of the compound during the extrusion process, the production limits are determined by the residence time coupled with the thermal states within the associated residence time. For the investigation of the RTD and the effect of process parameters on residence time properties, a single-screw rubber extruder equipped with a gear pump was used. In this contribution, we present a method to identify the RTD for a certain compound using experiments. We describe an experimental procedure with a special tracer material and a modified image processing method to identify the RTD. To validate the method, the experiment was repeated and compared to a predefined reference state. The experimental data and analyses, in particular, the input and output signals of the experiments, are used to determine the residence time distribution (transfer function) of the system. Based on previous findings from literature and experimental studies of the process, the influences of selected parameters on the transfer function were investigated and used to parameterize a general transfer function for the chosen compound. Consequently, the derived transfer function is used to predict the RTD of the process and the results are validated with experimental data. We discuss the potential of the method for an online implementation in a production system and outline how this method can be used within predictions for advanced process control.