Human induced vibrations on footbridges

In structural design there is a tendency towards more slender and challenging architectural structures. Footbridges become more slender with an increasing high ratio of live load to dead weight. A consequence of the increased slenderness of footbridges is an increased susceptibility to human-induced vibrations. The evaluation of the serviceability of footbridges therefore becomes more important. Many studies have been carried out to cover the human-induced bridge dynamics in a series of regulations in the Eurocode, guidelines and standards, but there is still limited clarity for the users. The first aim of this study was therefore to provide a good basis and overview of the critical aspects in the evaluation of humaninduced vibrations in footbridges. This study has been performed by means of a literature study, discussions with the engineering companies ARUP, RoyalHaskoningDHV and GemeentewerkenRotterdam and an impact study on three simply supported and two cantilever structures. The focus has been on vibrations in the vertical direction. From the literature study, the discussions and the impact study it became clear that there are four critical aspects particularly relevant for further study at the moment. First of all the structural safety of short footbridges and especially the safety of the more sensitive footbridges designed with new lightweight materials, should be examined for vandal loading. Secondly, the impact of structural damping has been shown to be a significant large percentage of the bridge response, which makes it an interesting and important topic for further study. Furthermore, the possibility of pedestrians to add mass and damping to the structure is of importance in the design of footbridges. The final bottleneck that has been indicated as an important issue for further research is the probability of occurrence of accelerations in footbridges in relation to the probability of discomfort of individual pedestrians. As a result of the investigation of the critical aspects, the second part of this thesis was focused on the question whether a probability-based approach can demonstrate that the Eurocode is conservative in the evaluation of human-induced vibrations in footbridges. Hereto, a probability-based analysis has been performed for the vibration serviceability of footbridges and implemented in a case study with three simply supported footbridges. Instead of looking at the maximum acceleration that is expected for the bridge deck, the accelerations which individual pedestrians experience when crossing the bridge have been investigated. Four scenarios have been compared to the approach prescribed by the Eurocode. The pedestrian loading has been modelled based on two assumptions for the scenarios: scenarios based on densities of the pedestrian flow and scenarios based on group formation. The conducted research based on assumptions for the pedestrian traffic and a fixed criterion for human comfort, has given valuable insight in the use of a more realistic evaluation of human-induced vibrations in footbridges. This insight has been obtained for the incoming pedestrian traffic and the exposure to vibrations for individual pedestrians. The adopted probability-based approach contributes to demonstrate potential conservatism in the Eurocode regarding the evaluation of human-induced vibrations in footbridges. It was not possible to give a final answer to the question if the adopted probabilistic-approach can demonstrate whether the Eurocode is conservative or not in the evaluation of human-induced vibrations in footbridges. This is due to a lack of data about the actual incoming pedestrian traffic on footbridges and due to outstanding questions regarding the type of accelerations to be considered for measuring human comfort on footbridges. However, regarding some key aspects in the evaluation of human-induced vibrations in footbridges, this research has shown four important aspects. The first important aspect this study showed is a distinction between two main types of pedestrian traffic: normal traffic comprising the commuter traffic, shopping area and park; and special locations with large groups of pedestrians walking in a high density, such as the train station. Whereas the Eurocode prescribes pedestrian traffic with a density of 0.5 P/m2, this study suggests a density for normal traffic of about 0.3 P/m2 based on a 95% non-exceedance level. For special locations on the other hand this density is expected to exceed 1.0 P/m2 considerably.