Special Issue: Machine Learning Methods for Urban Passenger Mobility
Submission Deadline: Dec 1, 2023; earlier submissions are encouraged
In connection with the recent Sustainable Urban Mobility: Simulation and Optimization Workshop, we are excited to announce a special issue focused on machine learning, simulation, optimization, and game theory methods for the design of more efficient, sustainable, accessible and equitable passenger transportation systems. Problem areas of particular interest include, but not are not limited to, traffic flow dynamics, travel demand management, emerging mobility options, and network optimization accounting for the passenger experience. The SI will focus on passenger-centric mobility approaches, which may go beyond an urban scope to include suburban, rural, and/or inter-urban areas. Submissions for this special issue are not limited to work presented at the above mentioned workshop.
We invite authors to submit research papers that address emerging urban mobility challenges and opportunities with non-conventional, data-driven methods. The Special Issue will feature novel methodological approaches to complex transportation problems arising from:
- the analysis of vehicular or passenger flows in large-scale metropolitan areas, such as city-scale demand modeling, large-scale traffic monitoring and management, and the underlying offline or real-time optimization problems that are high-dimensional, stochastic, and enabled by data;
- the development of cutting-edge technologies in improving accessibility and equity of transportation systems, such as electrified, connected and automated vehicles, and the corresponding control theoretic, game-theoretic and reinforcement learning models;
- the availability of multi-source, high-granular data, enabling enhanced network supply modeling, and enhanced modeling of the demand-supply interactions;
- the need for data-efficient, sample-efficient and computation-efficient methods appropriate for various stakeholders that operate in a small data environment (e.g., Bayesian optimization, sample-efficient machine learning);
- the need for optimization methods that can embed black-box traffic models into optimal decision making to assist various stakeholders (e.g., simulation-based optimization, black-box optimization, derivative-free optimization).
The methodological contributions of the submissions to this special issue may consist in purely data-driven, machine learning based approaches, as well as hybrid approaches that combine machine learning with domain knowledge, simulations, and decision making. However, a direct application of well-known or black-box machine learning models to transportation problems is not encouraged.
The review of papers for this special issue will heavily weigh the innovation and methodological novelty of the work presented, as well as its real-world impacts on society and the transportation community.
Timeline and Process
- Deadline for submission: December 1, 2023; earlier submissions are encouraged
- Two rounds of review, completed on a rolling basis as papers are submitted to meet the deadline for final decisions
- 2025: Publication of special issue
Guest Editors
Sharon Di (Columbia University)
Carolina Osorio (HEC Montreal, Google Research)
Sean Qian (Carnegie Mellon University)
https://pubsonline.informs.org/page/trsc/calls-for-papers
------------------------------
Karen Smilowitz
James N. and Margie M. Krebs Professor in Industrial Engineering and Management Sciences & Operations Department, Kellogg, Northwestern University
Editor-in-Chief, Transportation Science
------------------------------