Autor/innen
Til Lux, Chair of Structural Concrete, Technical University of Dortmund, 44227 Dortmund, Germany; Johannes Sundheim, Chair of Structural Concrete, Technical University of Dortmund, 44227 Dortmund, Germany; Tania Feiri, Chair of Structural Concrete, Technical University of Dortmund, 44227 Dortmund, Germany; Udo Wiens, Chair of Structural Concrete, Technical University of Dortmund, 44227 Dortmund, Germany; Marcus Ricker, Chair of Structural Concrete, Technical University of Dortmund, 44227 Dortmund, Germany
Über dieses Buch
The construction sector has a significant share of global CO₂ emissions due to its high material and resource consumption. Since material and resource consumption can be significantly influenced by the structural design of components, strategies to promote resource-efficient and optimised designs are required providing that the target reliability levels of structures prescribed in building design codes (e.g., DIN EN 1990+NA) are strictly satisfied. Structural components are designed in accordance with building design codes, such as DIN EN 1992-1-1+NA. Since the mechanical behaviour of some components can hardly be analytically described—e.g., punching shear of flat slabs without shear reinforcement—ultimate limit state evaluations partly rely on empirical equations. These equations exhibit significant model uncertainty, meaning that the ratio of actual to predicted load-bearing capacity is subject to statistical scatter. In punching shear verifications, model uncertainty has a great influence on the probability of failure. Previous studies have suggested that the model uncertainty of machine learning models can be lower than that of the empirical design equations specified in international codes. In this work, a machine-learning model (Extreme Gradient Boosting – XGBoost) was trained and utilised to predict the punching shear resistance of flat slabs without punching shear reinforcement. This model was employed in a reliability analysis with decoupled load and resistance side. Monte Carlo simulations were used to determine the probability that the actual load-bearing capacity, determined by the machine-learning model, falls below the design resistance according to DIN EN 1992-1-1+NA. The results indicate that predicting the actual load-bearing capacity with a trained model of lower uncertainty leads to a higher reliability level. This higher reliability can enable reductions in safety margins, for example, by lowering partial safety factors, thereby increasing the potential for resource-efficient designs.
Veröffentlicht
21.08.2026
Copyright (c) 2026 Matthias Faes, Michael Beer, Julian Behrend, Marco Broccardo, Guohai Chen, Hanshu Chen, Jieyu Chen, Fang Cong, Henrik Ebel, Peter Eberhard, Andrea Franchini, Zhuojia Fu, Yanzhao Gu, Michael Hanss, Hui Huang, Md Maidul Islam, Jafar Jafari-Asl, Balsa Jovanovic, Tom Könecke, Olga Kosheleva, Vladik Kreinovich, Hao-Ming Li, Pei-Pei Li, Qinghe Li, Ruikun Li, Stefano Marelli, Rafi L. Muhanna, Robert L. Mullen, Chiara Nardin, Lukáš Novák, Shenghao Piao, Mario Rosenfelder, Styfen Schär, Jan Schneider, Wei Shi, Panagiotis Spyridis, Bruno Sudret, Ruben Van Coile, Bo-Yu Wang, Dixiong Yang, Xuan-Yi Zhang, Zi-Xin Zhang, Bo-Qi Zhao, Yan-Gang Zhao, Zhibao Zheng, Baihong Zhong, Zhaowei Zhou, Marcos A. Valdebenito, Chao Dang, Lan-Xin Li, Marius Bittner, Parth Tambat, Marco Behrendt, Zhao Liu, Ping Zhu, Zlatan Dimitrov, Petar O. Hristov, Meng-Ze Lyu, Yang-Yi Liu, Jian-Bing Chen, Qitian Lu, Miroslav Vořechovský, Ekaterina Auer, Lorenz Gillner, Julien Alexandre dit Sandretto, Binod K. Yadav, Ghifari Adam Faza, Hans Hallez, David Moens, Reza Naeimaei, Steffen Schön, Elisabeth Ötsch, Hans Neuner, Zhanhua Liang, Jingwen Song, Michael Desch, Mehdi Modares, Valentin Nikolov, Paulo Lucas Figueiredo, José Torres Farinha, Hugo Raposo, António J. Marques Cardoso, Alice do Carmo Duarte Rodrigues, Paula Gonçalves, Peihan Chen, Xinyu Zhu, Pengfei Wei, Kaiwen Li, Mansoureh Shahabi, Y. F. Cui, Johannes O. Royset, Salvatore Russotto, Mario Di Paola, Antonina Pirrotta, Til Lux, Johannes Sundheim, Tania Feiri, Udo Wiens, Marcus Ricker, F. Niklas Schietzold, Felix Harazin, Yoshi Diepelt, Wolfgang Graf, Michael Kaliske, Rongyao Song, Changcong Zhou, Xuanyi Zhao, Weiping Zhang, Juntao Hu, Yu Chen, Edoardo Patelli, Tingting Sun, Jianbing Chen, Bouwe Verkens, L. Bogaerts, Patrick Van Rymenant, Maximilian Schweizer, Marc Fina, Werner Wagner, Steffen Freitag, George Stefanou, Dimitros Savvas, Panagiotis Gavalas, Luigi Schiano, Razhan S. Ahmed, Oleksandr Al-Shboul, David Ringeloth, Vladislav Gudžulić, Stefanie Schoen, Jelena Bijeljić, Ernst Niederleithinger, Iurie Curoşu, Gerrit E. Neu, Jia-Hui Fu, Yi Luo, De-Cheng Feng, Xiao-Qiu Ai, Luyi Li, Hao Wang, Tairan Wang, Sifeng Bi, Emily Zeller (Kapitelautor/in)
Zitationsvorschlag
Application of machine learning to reduce model uncertainty and enable resource-efficient design of flat slabs. (2026). In
REC 2026 - 11th International Workshop on Reliable Engineering Computing: Reliability Computations in a Data and Model-Driven World (pp. 83-98). TUDObooks.
https://doi.org/10.17877/tudobooks-11.182
