Related Experiment Video
Updated: Jun 27, 2026

Surrogate Model Development for Digital Experiments in Welding
Published on: March 28, 2025
Robust Optimization of GMAW Parameters in 6063-T5 Aluminium Welds Using Taguchi Design, ANOVA, and Monte Carlo
José L Meseguer-Valdenebro1, Eusebio José Martínez Conesa2, Diego Vergara3
1Escuela Técnica Superior de Ingenieros Industriales, Universidad Politécnica de Cartagena, Member of European University of Technology (EUT+), c/Dr. Fleming s/n, 30201 Cartagena, Spain.
Abstract:
This study investigates the influence of Gas Metal Arc Welding (GMAW) parameters on the elemental composition and hardness of 6063-T5 aluminum alloy welds. Power, welding speed, and bevel spacing were evaluated using Taguchi's L9 design to quantify their effects on magnesium and silicon contents and on weld hardness. ANOVA revealed that welding power is the only statistically significant factor governing the incorporation of Mg and Si into the weld metal, whereas no welding parameter showed a statistically significant direct effect on hardness within the investigated range. Nevertheless, an inverse relationship was observed between Mg-Si content and weld hardness, indicating that increased precipitation is associated with lower hardness values. Regression models derived for Mg and Si content were further combined with a Monte Carlo simulation comprising 100,000 iterations, enabling a probabilistic assessment of process robustness. The stochastic analysis confirmed the dominant influence of power on elemental composition and identified a robust operational window ensuring high hardness and controlled Mg-Si levels. The optimal conditions predicted by Taguchi (1596 W, 11.5 mm/s, 1.2 mm) were found to lie within the high-probability region obtained through Monte Carlo simulation. The combined use of Taguchi analysis, ANOVA, regression modelling, and Monte Carlo simulation enabled the evaluation of both process performance and robustness under parameter variability.

