Machine learning-based estimation of seismic structural damage via an accessible web application
This paper introduces DIGITERRA, a novel web-based platform that enhances accessibility to seismic damage estimation through machine learning techniques. Trained on 120,000 nonlinear dynamic simulations, DIGITERRA provides accurate structural damage assessments without requiring specialized software...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | Alexandria Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1110016825008063 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper introduces DIGITERRA, a novel web-based platform that enhances accessibility to seismic damage estimation through machine learning techniques. Trained on 120,000 nonlinear dynamic simulations, DIGITERRA provides accurate structural damage assessments without requiring specialized software or advanced technical expertise. The platform utilizes gradient boosting, a machine learning algorithm selected as the most effective after evaluating several alternatives. Feature selection is based on sensitivity analysis, SHAP analysis, and input from structural engineering experts to optimize both accuracy and accessibility. By allowing users to input basic building parameters and quickly receive damage state estimations, DIGITERRA democratizes access to advanced seismic analysis tools. This research demonstrates how machine learning can bridge the gap between complex engineering analyses and practical applications, empowering both specialists and non-specialists to make informed decisions about structural resilience in seismic-prone regions. |
|---|---|
| ISSN: | 1110-0168 |