Machine learning for optimal design of circular hollow section stainless steel stub columns - A comparative analysis with Eurocode 3 predictions

Aug 3, 2026

Stainless steel offers corrosion resistance, long service life, and recyclability, yet its higher initial cost makes reliable, accurate design methods essential for material optimization. This study develops machine learning models to predict the cross-section resistance of circular hollow section stainless steel stub columns using a parametric database covering diameter, thickness, length, and mechanical properties. Models trained and evaluated include ANN, DTR, GEP, and SVMR, with accuracy assessed against finite element results and Eurocode 3.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages20

Reading Time50 min

AccessFree to download

Introduction

Material background and design motivation

Constitutive behavior and code limitations

Machine learning models and comparative evaluation

Parametric study and dataset construction

Model training: ANN, DTR, GEP, SVMR

Comparison with finite element analysis and Eurocode 3

Engineering Applications of Artificial Intelligence 132 (2024) 107952 Contents lists available at ScienceDirect Engineering Applications of Artificial Intelligence journal [homepage: www.elsevier.com/locate/engappai](homepage: www.elsevier.com/locate/engappai) | Machine learning for optimal design of circular hollow section stainless steel stub columns: A comparative analysis with Eurocode 3 predictions | | | | | --- | --- | --- | --- | | Ikram Abarkana, Musab Rabib, Felipe Piana Vendramell Ferreira c, Rabee Shamassd, *, Vireen Limbachiyae, Yazeed S. Jweihan f, Luis Fernando Pinho Santos e
a Department of Physics, Faculty of Sciences, Abdelmalek Essaˆadi University, 93002, Tetouan, Morocco b Dept of Civil Engineering, Jerash University, Jordan
c Faculty of Civil Engineering, Federal University of Uberlandia – Campus Santa Monica, Uberlandia, Minas Gerais, Brazil d Department of Civil and Environmental Engineering, Brunel University London, London, UK
e Division of Civil and Building Services Engineering, School of the Built Environment and Architecture, London South Bank University, UK f Civil and Environmental Engineering Department, College of Engineering, Mutah University, Mutah, Karak, 61710, P.O. BOX 7, Jordan | | | | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords:
Circular hollow sections Stainless steel
Finite element model
Artificial neural network Support vector machine regression Gene Expression Programming Decision Trees for Regression | | Stainless steel has many advantages when used in structures, however, the initial cost is high. Hence, it is essential to develop reliable and accurate design methods that can optimize the material. As novel, reliable soft computation methods, machine learning provided more accurate predictions than analytical formulae and solved highly complex problems. The present study aims to develop machine learning models to predict the crosssection resistance of circular hollow section stainless steel stub column. A parametric study is conducted by varying the diameter, thickness, length, and mechanical properties of the column. This database is used to train, validate, and test machine learning models, Artificial Neural Network (ANN), Decision Trees for Regression (DTR), Gene Expression Programming (GEP) and Support Vector Machine Regression (SVMR). Thereafter, results are compared with finite element models and Eurocode 3 (EC3) to assess their accuracy. It was concluded that the EC3 models provided conservative predictions with an average Predicted-to-Actual ratio of 0.698 and Root Mean Square Error (RMSE) of 437.3. The machine learning models presented the highest level of accuracy. However, the SVMR model based on RBF kernel presented a better performance than the ANN, GEP and DTR machine learning models, and RMSE value for SVMR, ANN, GEP and DTR is 22.6, 31.6, 152.84 and 29.07, respectively. The GEP leads to the lowest level of accuracy among the other three machine learning models, yet, it is more accurate than EC3. The machine learning models were implemented in a user-friendly tool, which can be used for design purposes. | | 1. Introduction There are ever-increasing demands to improve the durability and the entire life performance of steel structures. In this context, stainless steel has recently emerged as a highly desirable option for structural applications owing to its distinctive characteristics which include outstanding corrosion resistance, long lifespan and recyclability (Rabi et al., 2022a). Moreover, stainless steel offers an attractive aesthetic appearance with excellent mechanical strength, significant strain hardening and great ductility. Although stainless steel has a relatively higher initial cost, it becomes a more competitive and efficient alternative design option than carbon steel, given the reduction in costs associated with regular inspections and maintenance and rehabilitation works (Rabi et al., 2022b). The

FAQs

What is the main objective of the study?

To build machine learning models that predict the cross-section resistance of stainless steel stub columns with circular hollow sections, and to compare their accuracy with finite element results and Eurocode 3 predictions.

Which machine learning methods are used for the predictions?

The study trains and tests Artificial Neural Networks (ANN), Decision Trees for Regression (DTR), Gene Expression Programming (GEP), and Support Vector Machine Regression (SVMR).

How do Eurocode 3 predictions compare with the proposed machine learning models?

Eurocode 3 provides conservative estimates, with an average Predicted-to-Actual ratio of 0.698 and RMSE of 437.3, while the machine learning models show the highest accuracy and lower RMSE values.

Aug 3, 2026

Upload by Emma Wilson