Musa Aaron Esekhagbona

This study aims to optimize reinforced concrete (RC) bridge decks using the Leonhardt and Makowski method. A MATLAB based GUI for live load distribution analysis was developed, allowing users to observe the impact of varying design parameters like span, slab thickness, and deck width. The analysis will follow BS5400 provisions, focusing on HA and HB loading combination. The methodology involved creating a MATLAB program that integrates the Leonhardt method into an interactive GUI and MATLAB scripts to run batch inputs. This tool validates user inputs, apply load cases (UDL, KEL, HB vehicle loads), compute section properties, and generate both tabular and graphical outputs (e.g., bending moment diagrams). To verify accuracy, results from manual calculations for bridge of span 25m, deck width of 11m and slab thickness 230mm was compared. The MATLAB tool is showed strong agreement with both manual calculations with 0.002% difference while the percentage difference compared to the STAAD.Pro analysis was 2.97% when computing the maximum longitudinal bending moments. The parametric study showed that the maximum moments appeared on the first support. The tool created will be able to provide engineers and students a flexible environment to explore design alternatives and understand how the inputs influence bridge behavior. Although the MATLAB GUI developed in this study performed excellently when compared to manual calculation, more comparative testing with other load distribution methods and with finite element based tool to further test the accuracy of the study.

This study focused on the analysis of load distribution in reinforced concrete (RC) bridge decks using the Guyon–Massonnet–Bares (GMB) method, enhanced through artificial intelligence (AI) and MATLAB integration. The primary aim was to simplify and automate the lengthy manual calculations typically associated with the GMB method by employing AI-assisted computation and visualization tools. Bridge deck parameters were obtained for a 25 m span bridge, and traditional analytical procedures were performed to determine the composite moment of inertia, centroidal properties, bending moments, and required reinforcement area. To improve efficiency, ChatGPT was utilized to generate MATLAB scripts based on defined parameters, enabling automated computation, graphical validation, and comparison of results with manual calculations. The generated MATLAB program successfully reproduced the analytical outcomes, verified bending moment distributions, and produced visual outputs such as load distribution profiles, bending moment diagrams, and influence lines. The integration of AI in bridge analysis effectively reduced human error, saved computational time by approximately 75%, and served as a dynamic learning platform for engineers and students. Benchmarked against classical GMB results, the AI-enhanced system achieved over 95% predictive accuracy, confirming its reliability and ability to simplify complex structural analysis without compromising computational precision