The rapid growth of modular construction has increased the demand for accurate and computationally efficient methods for predicting the shear performance of cold-formed steel members. Modular construction-optimised beams, characterised by a mono-symmetric triangular hollow flange geometry, exhibit shear behaviour that is not well represented by existing analytical formulations. This study proposes an explainable machine learning framework to predict the ultimate shear capacity of cold-formed steel modular construction-optimised beams using a validated finite-element dataset comprising 105 parametric models. Six supervised machine learning algorithms are trained and evaluated using resampling-based validation and statistical performance metrics. Categorical boosting achieved the best predictive performance, with a coefficient of determination of 95.9% and a mean absolute percentage error of 6.49% under 50 repeated train and test splits. Model transparency is supported using Shapley Additive Explanations, which confirm thickness and yield strength as the most influential inputs within the investigated domain. In addition, prediction uncertainty was quantified using empirical 95% prediction intervals, and the modelling workflow was strengthened by explicitly defining reproducibility and no-leakage conditions. Overall, the proposed framework provides an efficient and interpretable finite element surrogate tool for rapid design-oriented estimation of modular construction-optimised beam shear capacity within the defined parameter ranges and loading configuration.

Modular construction is gaining prominence for its sustainability, speed of assembly, reduced material waste, and cost-effectiveness. Cold-formed steel (CFS) beams, such as the Modular Construction Optimised (MCO) beam, play a vital role in these structures due to their lightweight characteristics, high strength-to-weight ratio, and ease of fabrication. However, the thin-walled geometry of CFS beams introduces challenges in structural design, particularly due to complex buckling and failure modes. The structural behaviour of the MCO beam remains insufficiently explored, with no prior research focusing on its web crippling performance under interior two-flange (ITF) loading. Existing design codes provide equations for estimating web crippling capacity. However, these provisions have been shown to underestimate the actual capacity of complex CFS sections, resulting in overly conservative designs and inefficient material use. To address these limitations, this study investigates the web crippling behaviour of the MCO beam using finite element analysis (FEA). Numerical models were developed and validated against experimental web crippling data from similar beam types. A parametric study involving 162 FE models was conducted to assess the influence of key geometric parameters displaying an average reduction of 27% due to corner radius effects. All models assumed unfastened flanges, reflecting common modular construction practices. Based on the results, new design equations were proposed to improve the accuracy of web crippling capacity predictions, providing a mean value of 1.00 and COV value of 0.08 and 0.07. These findings support the development of more efficient design practices, reduce material overuse, and contribute to the optimisation of lightweight modular steel structures.

© Copyright 2004 IEEEThe use of multiple antennas on a synthetic aperture radar (SAR), i.e., multichannel SAR, offers benefits such as improved ambiguity suppression for moving target indication (MTI) and imaging large swaths (Goodman, N.A. et al. 2002; Ender, J.H.G., 2000), improved signal-to-noise ratio (SNR) (Younis, M. et al., 2003) and the potential to suppress spatial jammers by use of space time adaptive processing (Ender, 1998). The multichannel matched filter (MF) interpolation imaging scheme presented by L. Rosenberg and D. Gray (see Int. Radar Symp. Proc., 2004) offers a good tradeoff between imaging accuracy and computational complexity. However, there are a number of potential problems which affect this algorithm in practice. Backprojection offers a solution to these problems and instead offers a direct tradeoff between accuracy and computation time. We extend the single channel backprojection algorithm to include multiple transmit and receive antennas. An analysis of the performance of the algorithm with varying levels of accuracy is shown, as well as a comparison with conventional MF algorithms