Publications / 2024 Proceedings of the 41st ISARC, Lille, France

Predicting Indicators of Design Quality for Cast-in-Place Reinforced Concrete Structures Using Machine Learning

Leonardo Garcia-Bottia, Daniel Castro-Lacouture
Pages 1255-1262 (2024 Proceedings of the 41st ISARC, Lille, France, ISBN 978-0-6458322-1-1, ISSN 2413-5844)
Abstract:

This paper presents the application of the information contained in exchange standards to predict indicators of design quality for concrete-in-place reinforced concrete (CIP RC) structures early in the design process. A logistic regression model is applied to each node type of a frame structure: beam-column, slab-column, beam-slab, and beam-beam. All model results present the significance of the variable chosen, as well as the classification table with very high values of prediction accuracy. The results show how well the obtained models fit the data, and therefore may be used to estimate potential construction issues early in the process, based on the parameters of the design intent standard exchanges.

Keywords: Design computing, Reinforced Concrete, Constructability, Logistic regression