There is an increased interest among both private and public urban
transportation stakeholders to develop and use urban mobility models to inform the design and the operation of their services. This talk presents computationally efficient algorithms for high-dimensional, static and dynamic, demand calibration problems. To tackle these problems, we formulate analytical metamodels with a complexity that scales linearly with network size, making them suitable for large-scale networks. We benchmark the approach versus standard calibration algorithms, and discuss Berlin and Singapore case studies. We will discuss ongoing work on real-time calibration algorithms.

Carolina Osorio is an Associate Professor of Civil & Environmental Engineering at Massachusetts Institute of Technology.