OmegaBuckling

Overview

Section-based buckling capacity prediction of rack uprights using machine learning.

OmegaBuckling is an advanced machine learning-based tool for predicting the axial buckling capacity of cold-formed steel uprights used in storage rack systems. Developed from a rigorously validated dataset derived from experimental and numerical analyses, OmegaBuckling provides accurate estimations of critical buckling load based on geometric and material properties of rack sections.

The underlying model is based on a peer-reviewed scientific study published in a high-reputation academic journal, ensuring transparency, credibility, and engineering-grade accuracy for structural professionals and researchers alike.

Academic Publication:
C. Yazici et al. (2025). Buckling capacity prediction of storage rack uprights using machine learning and feature importance techniques. Resources in Engineering. (In press) DOI: https://doi.org/10.1016/j.jcsr.2025.109458

Core Features

• Axial buckling load prediction for cold-formed steel uprights
• Geometric and material parameter integration (e.g., thickness, inertia, length)
• XGBoost regression model with high prediction accuracy
• Real-time estimation without need for finite element modeling
• Model explainability using feature importance analysis

Input Parameters

Output Parameters

• Predicted axial buckling load (kN)
• Normalized feature contributions
• Exportable performance report (PDF/CSV)
• Visual chart of load vs. section geometry

Target Users

• Rack system designers and engineers
• Structural consultants
• Cold-formed steel researchers
• Warehouse infrastructure specialists

Disclaimer

This software predicts the buckling capacity of structural rack uprights based on data reflecting collapse-stage loading. Estimated capacities represent critical loads under idealized conditions.

All design decisions, safety factors, and regulatory checks must be independently evaluated and applied by the practicing engineer. The authors and developers assume no liability for misuse or misinterpretation.

The underlying model is academically validated and published in a reputable scientific journal:
DOI: https://doi.org/10.1016/j.jcsr.2025.109458