On the 17th of September Danique Ton has defended her thesis entitled ‘Unravelling mode and route choice behaviour of active mode users’ at the Delft University of Technology. Her thesis will soon be available and can be found in the repository of the TU Delft. We are happy to see that she passed this final hurdle with flying colours and wish the young new doctor all the best in her new position as a postdoc in our department.
In the morning, TRAIL organised a nice seminar to highlight this occasion, which was attended by more than 40 participants from TU Delft and other knowledge institutes in the Netherlands. Four experts in the field of travel behaviour modelling presented recent developments and applications. Prof. Patricia Mokhtarian presented a latent-class regression approach to determine the role of attitudes in perceptions of bicycle facilities. Prof. Elisabetta Cherchi studied latent constructs to model travel choice behaviour. Prof. Shlomo Bekhor discussed a new frequency-based transit assignment model which can consider online information and capacity restrictions. Prof. Bert van Wee closed the workshop with an introduction of a new concept in the field of policy-making for travel behaviour, namely that of ‘substitutability’.
During Kingsday (the 27th of April) we performed a second pilot test in Amsterdam regarding the crowd monitoring dashboard that is being developed by the Amsterdam Institute for Advanced Metropolitan Solutions in collaboration with the Delft University of Technology. This dashboard is developed within the research project ALLEGRO: an ERC project lead by Serge Hoogendoorn on ‘unrAveLLing active modE travellinG and tRaffic’.
During the day, the live dashboard displayed the in- and outflow and the route split information of pedestrian flows near Amsterdam Central Station and station Amsterdam Zuid-WTC. The input for the dashboard was formed by several data feeds 11 counting cameras and 22 WiFi sensors being combined to create the output of the dashboard. Besides that, an analysis performed on social media streams (e.g Instagram and Twitter) which enabled the identification of relevant events (e.g. parties) taking place during the day; the combination of social data and data about pedestrian flows and routes allows a rich characterisation of activities in the city.
Overall the pilot has been a great success! During the day, the dashboard has been available in the control room, and its outputs have been available for municipality, police and public transport operators. Phenomena shown on the dashboard have been confirmed by crowd managers on the spot. For now we are studying the results of the day, and continue to work towards adding and fusing these and other data sources in order to create rich insights into the development of the flow of active modes through urban environments.