Presenting at the Transit Data 2019 – the 5th International Workshop and Symposium on research and applications on the use of massive passive data for public transport on 8th -10th July in Paris, France. Contributions include the following studies:
– Generating network-wide travel diaries using smartcard data
– Enhanced complex network representation of public transport for accessibility assessment based on General Transit Feed Specification data
– Impact analysis of a new metro line in Amsterdam using automated data sources
– Predicting and clustering station vulnerability in urban networks
– Investigating the effects of real-time crowding information (RTCI) systems in urban public transport under different demand conditions
Presented last week some of our research activities together with AMS on the WeMakeTheCity festival in Amsterdam in a session devoted to the future of MaaS and PT. While I do not argue that we should replace all public transport with individual on-demand services, I think that it helps to illuminate the ramifications of hypothetically doing so and hence support the debate with research findings.
New paper on “The underlying effect of public transport reliability on users’ satisfaction”. We show that a a service designed for a satisfaction level of 80% may yield a satisfaction level of less than 30% due to the non-linear relation between service irregularity and related uneven crowdedness, and service satisfaction. Hence the amount of satisfaction loss increases for each additional passenger boarding the vehicle (metro, bus), making things even worse. In technical terms this means a marginally increasing loss in satisfaction with increasing passenger on-board occupancy.
Zooming in and out on maps gives you the impression that a different version of the same network is displayed. However, important topological properties of the network might be distorted in the process.
We develop a method for automatically generating multiscale graph representations without significantly compromising their topological properties.
Our results show that the method is able to successfully reduce the Amsterdam road network by up to 96% of its original size at a computation time of no more than 15 min with a limited loss of information.
This is part of Panchamy Krishnankumari’s PhD thesis which I co-supervise together with Hans Van Lint.