Abstract This study investigates the structural and sustainable performance of wood‐infilled steel tubular columns, comparing circular (CWCC) and square (SWCC) cross‐sectional geometries under axial compression. The experimental program examined columns with varying slenderness ratios ( L / D or L / B = 6, 7, and 8), evaluating load–deflection and load–strain responses while monitoring deformation using strain indicators. Circular columns demonstrated superior axial strength due to enhanced confinement. Finite element models developed in ANSYS accurately captured nonlinear behavior, stress distribution, and progressive failure, with deviations of 5%–8% from experimental results. A total of 46 numerical models were employed to train a Decision Tree regression model, achieving excellent prediction accuracy ( R 2 = 0.987 for ultimate load and R 2 = 0.977 for axial deflection), demonstrating strong capability in capturing geometric and material nonlinearities. The failure mechanism was governed by local steel buckling and wood core crushing, while wood–steel interaction improved load redistribution, ductility, and energy absorption. The integration of wood waste–resin infill promotes sustainability by reusing industrial by‐products, offering a lightweight, eco‐friendly, and structurally reliable solution for modern construction. The study underscores the influence of cross‐sectional shape and slenderness ratio on column performance and validates the effectiveness of combined experimental, numerical, and machine learning approaches in predicting their behavior.
Sangeetha et al. (Tue,) studied this question.