Material testing combined with explainable AI predicts elastic properties of rubber-modified asphalt and ranks influential variables. The approach links accuracy with engineering interpretation, but must generalise across feedstocks, temperatures, ageing and field traffic.
Key findings
- Material testing combined with explainable AI predicts elastic properties of rubber-modified asphalt and ranks influential variables. The approach links accuracy with engineering interpretation, but must generalise across feedstocks, temperatures, ageing and field traffic.
Why this matters globally
This work adds internationally comparable evidence in Engineering and defines questions for replication in other populations or systems. Its global value lies in the evidence and transferable reasoning, not in a single impact score.
Thai researcher contribution
Thailand-linked authors and King Mongkut's University of Technology Thonburi contribute to the research network behind this work. Thai participation is identified from bibliographic affiliations and should be checked against the author list and source article.
Limitations to consider
Performance can fall under dataset shift; external validation, leakage checks, calibration and post-deployment monitoring are needed.