In this paper, a procedure to evaluate the wear state of non-return valves based on process and machine data of injection molding machines is introduced. In industry, non-return valves of injection molding machines are often replaced following specified maintenance intervals. Due to conservative lifetime estimations, the replacement of non-return valves is often carried out too early, since the current state of wear of the non-return valves is not known without elaborate disassembly and in case of failure other adjacent components can also be damaged. This paper shows an approach for an automated condition monitoring of non-return valves. By this, the corresponding wear state can be automatically evaluated during production without the use of additional sensors. By analyzing the curves of pressure, position signals etc. from experimental process data with injection molding process monitoring tools, specific key figures were defined. These key figures allow wear-specific conclusions about the different signs of wear and thereby allow an estimation of the current wear state of the non-return valve. With regard to predictive maintenance, the remaining service life of the non-return valve can be predicted based on the current state of wear. The knowledge of the remaining service life leads to increased machine uptime while reducing storage and maintenance costs for plastic processing companies.