The Load distribution factor equations of concrete I-Girder and super-T girder bridges in Australia

Bridges are one of the most important transport infrastructures. They play essential roles in a country's economy as they create links between cities or a country scale. The failure of bridges does not only disrupts the transportation system but also leads to a major impact on the economy, society and the safety of road users. Due to the growth of the economy, industry, logistics, and advanced technology, the demand for transportation has increased resulting in a larger number of Australian heavy vehicle loads travelling on bridges. The increase in vehicle loads is pressuring the bridge asset management team to ensure the safety of existing bridges so that they can accommodate the new heavier vehicles safely. Generally, bridges are assessed with field tests and analytical analyses such as grillage analysis (GA) and finite element analysis (FEA). However, field testing is not always achievable for many reasons, such as budget limitations and impact on road users. Moreover, using analytical analyses to assess the individual bridge will require high computational affords and a period of time to model and analyse. Super-T and I-girder bridges represent 30% of Victorian bridges, approximately 900 bridges. Due to the drawbacks of field tests and analytical analyses, it would not be feasible to use those methods in the preliminary stage. Load distribution factor (LDF) has been widely used for bridge design and assessment in the preliminary stage in many countries around the globe. In Australia, LDF is not specified in the current Australian bridge standard code (AS 5100). The development of LDF would bring enormous benefits to the asset management team. Therefore, this study aims to investigate the accuracy of GA compared with FEA and develop LDF empirical equations for Super-T and I-girder bridges for preliminary bridge design and assessment. In order to achieve the aims of this study, the load effects and LDF from GA and FEA were compared and derived the load effects ratios to identify the accuracy of GA under the typical bridge cross-section and vehicle loads. The LDFs of Australian bridges were also compared to  the different standard codes to find the correlation and utilisation of the other standard codes in Australian bridges. Moreover, the LDF empirical equations for Super-T and I-girder bridges were developed according to bridge design space in Australia. It is concluded that there are some discrepancies between grillage and finite element analyses in Super-T and I-girder bridges. The developed load effect ratios could be utilised to optimise the accuracy of load effects under typical bridge cross-sections and vehicle loads. However, the discrepancy tends to ease with the increase in span length. The comparison of the LDF of Australian bridges with the different standard codes shows that the American Association of State Highway and Transportation Officials (AASHTO) Load and Resistance Factor Design (LRFD) 2017 provides the closet LDF estimation of Australian bridges compared to other standard codes. However, developing the LDF equations for Australian bridges is recommended because the conservative results of AASHTO LRFD could impact the accuracy of bridge assessment. Later, the LDF empirical equations of Super-T and I-girder were developed with nonlinear regression analysis. Sixteen developed equations provide good accuracy with slight conservative compared to the LDFs from FE models and existing bridges in Victoria, Australia.