Abstract: The computational efficiency and real-time performance of creep fatigue life prediction for aerospace engine turbine disks are low, making it difficult to meet the requirement of life management in practical engineering applications, therefore, a digital twin model for predicting the creep fatigue life of turbine disks is proposed. The model employs reduced order model (ROM) methods to rapidly derive structural stress fields under varying operating conditions, followed by life prediction based on stress-driven fatigue-creep damage models. A parametric ROM is first constructed to replace high-fidelity FEA for stress field computation. By projecting the full-order governing equations onto a low-dimensional subspace using techniques such as singular value decomposition (SVD), the ROM achieves a 99.74% reduction in computational time compared to conventional FEA, enabling stress field predictions within seconds under varying operational conditions. Leveraging the real-time stress results, the model enables instantaneous prediction of creep-fatigue life across diverse operational scenarios, effectively resolving latency issues in conventional methods. To validate reliability of the model, six groups of creep-fatigue tests were conducted on the test specimens for the bottom of the turbine disk under varying conditions. Results demonstrate that the relative errors between predicted and experimental lifespans consistently fall within a 1.5-fold error band, confirming the model's high prediction accuracy. The proposed framework significantly enhances computational efficiency while maintaining precision, offering a practical solution for real-time lifespan monitoring and management of turbine disks in engineering applications. This advancement bridges the gap between theoretical models and industrial requirements, providing a robust foundation for proactive maintenance strategies in aviation systems.