Comprehensive review of shear resistance and web stiffeners design in plate girders

This paper presents a comprehensive review of the role of transverse, longitudinal, and diagonal web stiffeners in enhancing the shear buckling resistance of I-shaped steel plate girders. Over the past few decades, extensive laboratory experiments, analytical formulations, and numerical analyses have been conducted to better understand the shear behaviour of stiffened webs, supporting the development of more efficient and reliable design methods. The review summarises and critically compares the main design models for elastic buckling and ultimate shear resistance, as well as the essential design requirements for web stiffeners. Key investigations are compiled in detailed tables, and existing analytical expressions are examined in terms of their assumptions, boundary conditions, and accuracy. In addition, a finite element (FE) model is developed to evaluate the shear performance of girders with different stiffener configurations – including transverse, longitudinal, or diagonal – and its predictions are benchmarked against available design methods. The FE results complement the literature findings by highlighting general trends, verifying analytical formulations, and assessing the conservatism of current design rules. Overall, the study demonstrates that while existing models predict the shear buckling resistance of stiffened plate girders reasonably well, notable gaps remain regarding the interaction between stiffeners, the treatment of diagonal stiffeners, and the influence of torsional rigidity. The review identifies these challenges and outlines future directions for improving both the accuracy and efficiency of web stiffener design. • Transverse stiffeners boost shear by up to 35%, best in short girder panels. • Longitudinal stiffeners cut deflection, raise shear by 46% in long panels. • Diagonal stiffeners add up to 66% resistance; best when compression-aligned. • Web stiffener impact varies by panel slenderness, stiffness, and orientation. • Current models can be improved; updates needed for stiffener interaction effects.