This study presents an interpretable machine-learning (ML) approach to evaluate the impact of heat treatment (HT) processes on the creep performance of superalloys and predict creep rupture life with high accuracy. The approach integrates classification and regression models with domain-specific knowledge, enhancing prediction accuracy by incorporating physical constraints. The heat treatment process is identified as the most significant descriptor through the integration of ML and superalloy creep theory. The study uses the Waspaloy alloy as an experimental validation, demonstrating a significant improvement in creep performance (5.5 times higher than previous studies) by optimizing the HT process. The research provides novel insights into enhancing the precision of predicting creep rupture life in superalloys and broadens its applicability to other properties. Additionally, it supports the use of ML in designing HT processes for superalloys. The study highlights the importance of physical features and domain-specific knowledge in improving model interpretability and predictive accuracy.This study presents an interpretable machine-learning (ML) approach to evaluate the impact of heat treatment (HT) processes on the creep performance of superalloys and predict creep rupture life with high accuracy. The approach integrates classification and regression models with domain-specific knowledge, enhancing prediction accuracy by incorporating physical constraints. The heat treatment process is identified as the most significant descriptor through the integration of ML and superalloy creep theory. The study uses the Waspaloy alloy as an experimental validation, demonstrating a significant improvement in creep performance (5.5 times higher than previous studies) by optimizing the HT process. The research provides novel insights into enhancing the precision of predicting creep rupture life in superalloys and broadens its applicability to other properties. Additionally, it supports the use of ML in designing HT processes for superalloys. The study highlights the importance of physical features and domain-specific knowledge in improving model interpretability and predictive accuracy.