Footbridge response: Improved understanding through probabilistic interpretation

Design of footbridges is dominated by the structural response under their most common pedestrian loading scenarios. Failures of several structures under these serviceability conditions have highlighted that fundamental assumptions considered when developing load models applied in their design were not realistic. Following these failures, a lot of recent research has focused on the topic of improving the definition of both vertical and lateral loading, included its nonlinear character, as well as on flow movement on platforms. However, these improvements have not as yet resulted in improved load models for design purposes. Building upon some of these recent advances, this paper introduces a new methodology for evaluating the response of footbridges under serviceability conditions. This method arises from a probabilistic characterisation of the parameters involved. The main aim of this work is to highlight the advantages of the probabilistic approach as opposed to the deterministic approaches, currently included in design codes and guidelines, when the intention is to obtain the best possible reproduction of reality. With this purpose, the paper presents the framework and underlying assumptions of the probabilistic method, the typical assumptions and methods in deterministic approaches, and the comparison of response of an example structure under diverse loading scenarios obtained from both approaches. Comparison of the results highlights the advantages of the probabilistic method over the other as well as the key parameters to be taken into account in order to obtain the best possible reproduction of real service scenarios and, therefore, responses.