Operational data collection and analysis for a smart 3D-printed footbridge

Purpose This paper presents the structural monitoring results of the world's first 3D-printed steel footbridge, using a fixed sensor network built into the bridge, to better understand both the behaviour of this novel structure and the way it is used. Design/methodology/approach The bridge was publicly exhibited and then installed for two years in central Amsterdam. The main features of the sensor network installed to monitor its behaviour are described. The bridge's behaviour was studied using a combination of labelled data collected in controlled conditions at the University of Twente and long-term monitoring during normal use in Amsterdam. Findings The data collected show that thermal behaviour can be effectively decoupled from the response of the bridge due to pedestrian loading and that the pedestrian movements captured by camera can be anonymized as coordinates, which can be correlated with the loads and strains produced by those pedestrians. The Pearson correlation condition is used to identify the type of movement on the bridge, effectively distinguishing between heel-drops, running and walking movements. Originality/value The richness of such a dataset is demonstrated, measured using embedded sensors and Internet of Things technology. Analysis of these measurements gives insights into the behaviour of a unique large 3D-printed steel structure and the use of a busy piece of urban infrastructure more generally.