Connection Design of Precast Concrete Structures Using Machine Learning Techniques - Research Focuses on Dowel Number Estimation

Aug 3, 2026

Upload by Himbo

Research determines the number of dowels in horizontal precast concrete connections using support vector machines, gradient boosting, and artificial neural networks based on multilayer perceptron. Wall geometry and demand variables—building height, length, thickness, maximum shear, maximum compressive force, and maximum tension—serve as inputs, while dowel count is the predicted output. From 1140 models, SVM with radial basis kernel performs best with correlation 0.9264, RMSE 0.3677, and scattering index 4.75% versus resilient ANN-Multilayer perceptron. Results indicate strong suitability of machine learning for improving civil connection design.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Dimensions and profile of precast members

Joints & connections

Literature

|
Contents lists available at SCCE | | | --- | --- | |
Journal of Soft Computing in Civil Engineering | | |
Journal homepage: [http://www.jsoftcivil.com/](http://www.jsoftcivil.com/) | | | Connection Design of Precast Concrete Structures Using Machine Learning Techniques
Nitin Dahiya 1 * , Babita Saini 2 , H.D. Chalak 3
1 Ph.D. Student, Faculty of Civil Engineering, National Institute of Technology Kurukshetra, Haryana, India
2. Associate Professor, Faculty of Civil Engineering, National Institute of Technology Kurukshetra , Haryana, India
3. Assistant Professor, Faculty of Civil Engineering, National Institute of Technology Kurukshetra, Haryana, India
Corresponding author: [nitindahiya17@gmail.com](nitindahiya17@gmail.com)
[https://doi.org/10.22115/SCCE.2023.356547.1506](https://doi.org/10.22115/SCCE.2023.356547.1506) | | | ARTICLE INFO
Article history:
Received: 17 August 2022
Revised: 25 January 2023
Accepted: 04 April 2023
Keywords:
Machine learning;
Gradient boosting;
Support vector machines; Precast concrete structures; Computer programming. | ABSTRACT
In this research, the number of dowels (horizontal connection) has been determined using support vector machines (SVM), gradient boosting and artificial neural networks (ANN-Multilayer perceptron) . Building height, length and thickness of the wall, maximum shear, maximum compressive force and maximum tension were the input parameters while the output parameter was the number of dowels. 1140 machine learning models were used, out of which 814 were used as training datasets and 326 as test datasets. A coefficient of correlation of 0.9264, root mean square error of 0.3677 and scattering Index of 4.75 % was achieved by SVM radial basis kernel function (SVM-RBF) as compared to a coefficient of correlation of 0.9232, root mean square error of 0.3743 and scattering Index of 4.83 % by resilient ANN-Multilayer perceptron, suggesting that SVMRBF is more accurate in estimating the number of dowels. The study's encouraging findings highlight the need for additional research into the use of machine learning in civil engineering. | 1. Introduction The precast structure includes the analysis and design of members, the method of fabrication, and the various aspects including the members lifting, transporting and erecting the members onsite. The analysis and design require an examination of the strength and stability of the structure. Apart from design, the casting of members, transportation, and temporary stability of the structure during construction has been also considered. The following aspects have been considered during the design stage for achieving good-quality precast concrete elements: 􀁸 Dimension and profile of precast members. 􀁸 Member joints and connections. 1.1. Dimensions and profile of precast members The size of precast members mainly depends on the lifting capability of the lifting crane at the casting area and site. The designer attempts to provide the largest size of elements to minimize joints and handling at the site. Casting and construction quality has been achieved by the combination of different members, such as shear wall system, 3D units and beam-column system. The design has been regulated in such a way allowing for maximum replication in the dimension and profile of the members. This is an advantageous factor for achieving economic and quality assurance as fewer steel mould changes will make possible the construction agenda, thereby reducing production time. 1.2. Joints & connections The design of connections is to effectively transfer the loads from members to the structure or a neighbouring member. Strength, Stability, ductility, durability and fire resistance are important considerations considered in the design of joints. Design criteria not only consider the production of elements but also consider the erection process of the element as well. The joints and connections are de

FAQs

Which input parameters are used to predict the number of dowels in the study?

The study uses building height, wall length, wall thickness, maximum shear, maximum compressive force, and maximum tension as input parameters.

How is the number of dowels determined, and which models are compared?

Support vector machines, gradient boosting, and an ANN multilayer perceptron are used. The comparison highlights SVM with radial basis kernel against resilient ANN-Multilayer perceptron.

What performance metrics show that SVM-RBF estimates dowel number more accurately?

SVM-RBF achieves a coefficient of correlation of 0.9264, RMSE of 0.3677, and scattering index of 4.75%, which are slightly better than the resilient ANN-Multilayer perceptron results.

Aug 3, 2026

Upload by Himbo