As part of her visit to DTU, Bianca Pascariu presented her work on Integrating passenger demand prediction in real-time rail traffic management to the Division for Transportation Science.

Below you can find the abstract and more information about Bianca’s research interests.

Abstract: Recent transport policies promote a major shift towards rail travel to facilitate green mass mobility. This shift is hampered by widespread unexpected perturbations and disruptions in operations, resulting in perceived poor punctuality and reliability. When prevention of such perturbations and disruptions is not feasible, traffic management must mitigate their effects, resolving arising conflicts to restore regular train operations and minimize delay. Existing policies include the assessment of railway performance in terms of train delays, and the quality of service to passengers is not explicitly accounted for.
In this work, we propose a railway traffic management policy that accounts for both passenger and train delays. To do so, we propose a demand prediction module. It dynamically predicts future origin-destination passenger flows using real-time observed smart card data. Then, an assignment module links predicted passengers to specific train paths, given a railway schedule. Finally, the traffic management module RECIFE-MILP, a mixed integer linear programming based heuristic, optimizes train scheduling and routing in real-time, under the combined objective of minimizing both train and passenger delays. We validate and benchmark our methodology against equivalent passenger agnostic traffic management on a case study of the Copenhagen suburban railway network. The results show that it is possible to take into account passenger perspective in railway traffic management, without reducing the system efficiency compared to classic approaches.

Bio: Bianca’s research interest is in rail traffic optimization. Her academic journey led her to a PhD in Computer Science and Automation at Roma Tre University, Italy . During her doctoral studies, she focused on solving real-time railway traffic control problems by developing and applying state-of-the-art techniques for practical-size instances.

She is involved in the SORTEDMOBILITY project, which seeks to self-organize public transport operations in urban and interurban areas, with a focus on rail transport as the backbone of mobility. Her research aims to integrate rail traffic management by taking into account the perspectives of passengers and trains.