Robust extreme gradient boosting model for predicting the behavior of RC slabs under impact loading: key influencing factors and performance insights
Abstract This study presents an advanced approach to analyzing the impact behavior of reinforced concrete (RC) slabs, utilizing an optimized extreme gradient boosting (XGB) machine learning algorithm. Supported by a comprehensive dataset of 143 records drawn from diverse sources, the methodology ef...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Instituto Brasileiro do Concreto (IBRACON)
2025-04-01
|
| Series: | Revista IBRACON de Estruturas e Materiais |
| Subjects: | |
| Online Access: | http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1983-41952025000200210&lng=en&tlng=en |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849315336781824000 |
|---|---|
| author | Ammar Babiker |
| author_facet | Ammar Babiker |
| author_sort | Ammar Babiker |
| collection | DOAJ |
| description | Abstract This study presents an advanced approach to analyzing the impact behavior of reinforced concrete (RC) slabs, utilizing an optimized extreme gradient boosting (XGB) machine learning algorithm. Supported by a comprehensive dataset of 143 records drawn from diverse sources, the methodology effectively pinpoints and evaluates critical parameters affecting the model's predictions. The XGB algorithm showed superior predictive performance, with high coefficients of determination, underscoring the model’s robustness and accuracy across both training and testing datasets. The study identifies concrete compressive strength, slab thickness, and projectile velocity as key determinants in the impact behavior of RC slabs. The outcomes reveal that optimal concrete compressive strength is found to be between 45 and 60 MPa, whereas the optimal projectile mass lies in the range between 300 and 750 kg. Moreover, the ideal slab thickness is determined to be between 200 and 300 mm. These results provide valuable insights into the design and optimization performance of RC slabs subjected to projectile impact, offering practical applications in structural engineering. The study highlights the potential of ML algorithms in structural analysis and encourages future research to validate these outcomes with experimental data and explore the influence of different projectile shapes on slab performance. |
| format | Article |
| id | doaj-art-1bcc2de804e042ea887f1de8f13c0808 |
| institution | Kabale University |
| issn | 1983-4195 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Instituto Brasileiro do Concreto (IBRACON) |
| record_format | Article |
| series | Revista IBRACON de Estruturas e Materiais |
| spelling | doaj-art-1bcc2de804e042ea887f1de8f13c08082025-08-20T03:52:10ZengInstituto Brasileiro do Concreto (IBRACON)Revista IBRACON de Estruturas e Materiais1983-41952025-04-0118210.1590/s1983-41952025000200013Robust extreme gradient boosting model for predicting the behavior of RC slabs under impact loading: key influencing factors and performance insightsAmmar Babikerhttps://orcid.org/0000-0002-4081-5015 Abstract This study presents an advanced approach to analyzing the impact behavior of reinforced concrete (RC) slabs, utilizing an optimized extreme gradient boosting (XGB) machine learning algorithm. Supported by a comprehensive dataset of 143 records drawn from diverse sources, the methodology effectively pinpoints and evaluates critical parameters affecting the model's predictions. The XGB algorithm showed superior predictive performance, with high coefficients of determination, underscoring the model’s robustness and accuracy across both training and testing datasets. The study identifies concrete compressive strength, slab thickness, and projectile velocity as key determinants in the impact behavior of RC slabs. The outcomes reveal that optimal concrete compressive strength is found to be between 45 and 60 MPa, whereas the optimal projectile mass lies in the range between 300 and 750 kg. Moreover, the ideal slab thickness is determined to be between 200 and 300 mm. These results provide valuable insights into the design and optimization performance of RC slabs subjected to projectile impact, offering practical applications in structural engineering. The study highlights the potential of ML algorithms in structural analysis and encourages future research to validate these outcomes with experimental data and explore the influence of different projectile shapes on slab performance.http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1983-41952025000200210&lng=en&tlng=enprojectile impactreaction forcereinforced concretemachine learningdependent plots |
| spellingShingle | Ammar Babiker Robust extreme gradient boosting model for predicting the behavior of RC slabs under impact loading: key influencing factors and performance insights Revista IBRACON de Estruturas e Materiais projectile impact reaction force reinforced concrete machine learning dependent plots |
| title | Robust extreme gradient boosting model for predicting the behavior of RC slabs under impact loading: key influencing factors and performance insights |
| title_full | Robust extreme gradient boosting model for predicting the behavior of RC slabs under impact loading: key influencing factors and performance insights |
| title_fullStr | Robust extreme gradient boosting model for predicting the behavior of RC slabs under impact loading: key influencing factors and performance insights |
| title_full_unstemmed | Robust extreme gradient boosting model for predicting the behavior of RC slabs under impact loading: key influencing factors and performance insights |
| title_short | Robust extreme gradient boosting model for predicting the behavior of RC slabs under impact loading: key influencing factors and performance insights |
| title_sort | robust extreme gradient boosting model for predicting the behavior of rc slabs under impact loading key influencing factors and performance insights |
| topic | projectile impact reaction force reinforced concrete machine learning dependent plots |
| url | http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1983-41952025000200210&lng=en&tlng=en |
| work_keys_str_mv | AT ammarbabiker robustextremegradientboostingmodelforpredictingthebehaviorofrcslabsunderimpactloadingkeyinfluencingfactorsandperformanceinsights |