Given the changing dynamics of mobility patterns and rapid growth of cities, transport agencies seek to respond more rapidly to needs of the public with the goal of offering an effective and competitive public transport system. A more data-centric approach for transport planning is part of the evolution of this process. In particular, the vast penetration of mobile phones provides an opportunity to monitor and derive insights on transport usage. Real time and historical analyses of such data can give a detailed understanding of mobility patterns of people and also suggest improvements to current transit systems. On its own, however, mobile geolocation data has a number of limitations. We thus propose a joint telco-and-farecard-based learning approach to understanding urban mobility. The approach enhances telecommunications data by leveraging it jointly with other sources of real-time data. The approach is illustrated on the First- and last-mile problem as well as route choice estimation within a densely-connected train network.

We consider the problem of trajectory-based road pricing with the objective of reducing congestion on a road network. It is well-known that traffic conditions resulting from typical non-cooperative behavior of selfish drivers do not minimize total travel time spent on the road network. In the context of real-time GPS data collection from all vehicles, drivers can be charged differently based on their origin and destination, and according to the path they take from that origin to that destination. In this work, we propose a new formulation of the set of multi-commodity prices based on a price potential, and describe an efficient algorithm to construct such multi-commodity prices. We provide an analysis of the subset of valid prices satisfying several specific user-driven constraints. The numerical performances of the method proposed are assessed on a benchmark network, and the social benefits resulting from the commodity-based potential pricing scheme introduced in this article are discussed.

We consider the stochastic on-time arrival (SOTA) routing problem of finding a routing policy that maximizes the probability of reaching a given destination within a pre-specified time budget in a road network with probabilistic link travel-times. The goal of this work is to provide a theoretical understanding of the SOTA problem and present efficient computational techniques to enable the development of practical applications for stochastic routing. We present multiple speedup techniques that include a label-setting algorithm based on the existence of a minimal link travel-time on each road link, advanced convolution methods centered on the Fast Fourier Transform and the idea of zero-delay convolution, and localization techniques for determining an optimal order of policy computation. We describe the algorithms for each speedup technique and analyze their impact on computation time. We also analyze the behavior of the algorithms as a function of the network topology and present numerical results to demonstrate this. Finally, experimental results are provided for the San Francisco Bay Area arterial road network to show how the algorithms would work in an operational setting.