Transparency in the actual use of complex forming machines is essential for pay-per-x models. However, the use of operating time as the sole indicator of these models is hardly sufficient, as the strongly varying operation with different forming processes has a strong impact on the stress and degradation of the forming machine. This paper presents a digital twin-based framework for load and degradation determination of forming machines as an enabler for pay-per-stress models. It considers especially highly stressed structural components like machine frame in the load determination by using virtual sensors. The degradation is determined based on the fatigue theory and the calculated stress using actual virtual sensor-based load data. With this approach a stress-based indicator for pay-per-stress models for complex forming machines is provided.