Home

Emerging Mobility Systems and Services

A weekly online seminar on emerging topics of new mobility systems and services

Every Thursday 10 a.m. EST. on Zoom.

morning (North America); afternoon (Europe); evening (Asia)


Previous presentations

Nikolas Geroliminis

Associate Professor, Institute of Transportation Studies

Ecole polytechnique fédérale de Lausanne (EPFL)

Title: On the new era of traffic management for large urban networks – Recent advances in MFD research

Video recording: Youtube; Bilibili

Abstract: Human mobility in congested city centers is a complex dynamical system with high density of population, many transport modes to compete for limited available space and many operators that try to efficiently manage different parts of this system. New emerging modes of transportation, such as ride-hailing and on-demand services, and new technologies, such as autonomous vehicles, create additional opportunities, but also more complexity. The new era of sharing information and ‘big data world’ has raised our expectation to make mobility more predictable and controllable through a better utilization of existing resources and capacity. The primary motivation of this talk is to study the spatiotemporal relation of congested links in large networks, develop new advancements in the Macroscopic Fundamental Diagram (MFD), observe congestion propagation from a macroscopic perspective, identify the effect of multimodal interactions in network capacity and finally design network-level control strategies to improve multimodal mobility. Investigating the clustering problem over time help us reveal the hidden information during the process of congestion formation and dissolution. In this framework, we will be able to chase where congestion originates and how traffic management systems affect its formation and the time it finishes. Different control strategies are developed based on principles of optimization and control theory.

Kara Kockelman

Dewitt Greer Centennial Professor of Transportation Engineering, Civil, Architectural and Environmental Engineering

University of Texas at Austin

Title: Shifting toward Shared Fleets and Shared Rides, via Autonomous Vehicles and Congestion Pricing

Video recording: Youtube; Bilibili

Abstract: Connected and (fully-) automated vehicles (CAVs) are set to disrupt the ways in which we travel, and result in more motorized trips and longer trips. Shared AVs (SAVs) will offer many people access to such technologies at relatively low cost (e.g., $1 per mile), with empty-vehicle travel on the order of 10 to 15 percent of fleet VMT. If SAVs are smaller and/or electric, and dynamic ride-sharing is enabled and regularly used, emissions and energy demand may fall. If road tolls are thoughtfully applied, using GPS-based systems along all congested network segments, total VMT may not rise: instead, travel times – and their unreliability – may fall. If credit-based congestion pricing is used, traveler welfare can rise and transportation systems may operate near-optimally. This presentation will present research relating to all these topics, including the benefits of SAV stop aggregation.

Kay W. Axhausen

Professor, Civil, Environmental and Geomatic Engineering

ETH Zürich

Title: Micromobility in Switzerland: Demand and Competition

Video recording: Youtube; Bilibili

Abstract: Based on detailed data for micromobility systems in Switzerland this talk will present two analysis: 1. Demand estimation for two cities and their comparison. It will highlight that current models explain each city, but are difficult to transfer. 2. Based on 5 different firms’ data the second paper will offer the first analysis of the competition between such services known to us. We highlight the impacts of prices, saturation of supply and of the terrain.

Hai Yang

Chair Professor, Civil and Environmental Engineering

Hong Kong University of Science and Technology (HKUST)

Title: Some Emerging Research Issues in Ride-sourcing Markets

Video recording: Youtube; Bilibili

Abstract: Urban mobility has undergone drastic changes in recent years with the introduction of application-based taxi and car service e-hailing systems. These systems provide timely and convenient on-demand ride services to anyone, anywhere and anytime. E-hailing is now prevalent in the traditional taxi industry by effectively mitigating information asymmetry and uncertainty between customers and taxi drivers; E-hailing in the form of ride-sourcing can efficiently match a requesting customer with an affiliated private car driver nearby for on-demand ride services. This talk highlights some emerging research issues and latest developments in ride-sourcing markets, including demand forecasting; surge-pricing; matching, pricing and ride-pooling, joint optimization of matching time interval and matching radius; customers’ maximum willingness to wait with sunk waiting time; optimal resource allocation with bundled options of choice; impact of ride-pooling on traffic congestion; competition and third-party platform-integration; Pareto-efficient market regulations; and analysis of human mobility and network property with big car trajectory data, etc.

Chandra R. Bhat

University Distinguished Professor

Joe J King Chair Professor in Engineering

University of Texas at Austin

Title: On Sharing the Road with Autonomous Vehicles: Perceived Safety and Regulatory Preferences

Video recording: Youtube, Bilibili

Abstract: Technology providers, car manufacturers, and public agencies all need to work together to undertake extensive testing of fully autonomous vehicles (AVs) on public roads before such AVs are allowed to freely travel in ways similar to human-driven vehicles. This raises the importance of understanding public perceptions regarding safety considerations when traveling alongside AVs. This study makes use of a national survey conducted by the Pew Research Center to identify the affective, socio-demographic and technology-use attributes that affect an individual’s perception of the safety of sharing the road with AVs (PSSRAV) and identifies measures and interventions that can be undertaken to improve PSSRAV. Additionally, we evaluate individual preferences for AV regulations. Our results underscore the importance of the need for service providers and public agencies to be cognizant of the demographic and lifestyle/affective emotion considerations shaping AV safety perceptions and opinions about AV regulations. In particular, there is a need not only to focus on technological and other infrastructure components of AV development, but also to recognize the socio-technical considerations and human-related factors of the end-users. Our findings should be of substantial interest in the planning, design, deployment, and introduction of AVs within a safe and minimally regulated public operating arena.

Jieping Ye

Vice President, Head of DiDi AI Labs

Didi Chuxing

Professor

University of Michigan

Title: AI for Transportation

Video recording: No recording for this special event

Abstract: Didi Chuxing is the world’s leading mobile transportation platform that offers a full range of app-based transportation options for 550 million users. Every day, DiDi’s platform receives over 100TB new data, processes more than 40 billion routing requests, and acquires over 15 billion location points. AI has been used in numerous components of DiDi’s platform to improve travel safety, experience and efficiency. This talk systematically presents the challenges and opportunities in the core area of modern transportation systems, and highlights some of our recent works on order dispatching and fleet management via deep reinforcement learning.

Hani Mahmassani

Professor, Civil and Environmental Engineering

William A. Patterson Distinguished Chair in Transportation

Northwestern University

This talk has been cancelled.

Peter I. Frazier

Associate Professor, Operations Research and Information Engineering

Cornell University

Staff Data Scientist

Uber

Title: Matching Queues, Flexibility and Incentives

Video recording: No recording for this special event

Inspired by ridesharing dispatch at Uber, we consider queuing control systems in which strategic agents have heterogeneous privately known constraints. At Uber, such constraints arise through the driver destination filter, which lets drivers customize the subset of destinations toward which they are willing to take trips. In similar systems with non-strategic agents, it is beneficial to reserve capacity from flexible agents by preferentially dispatching to less flexible ones when possible. However, in ridesharing, because drivers report their availability over destinations strategically based on private information, reserving capacity in a naive way can hurt flexible drivers’ earnings per hour and lead them to under-report the set of destinations they can serve, actually degrading throughput. We describe a new mechanism that effectively reserves capacity from flexible drivers and show that this new mechanism always performs at least as well as not trying to reserve capacity. This new mechanism uses temporal rewards in place of monetary ones, a technique we view as generalizable to other platform-based service systems with heterogeneous strategic servers.

Martin Savelsbergh

James C. Edenfield Chair Professor,  H. Milton Stewart School of Industrial and Systems Engineering

Co-Director Supply Chain & Logistics Institute

Georgia Institute of Technology

Title: Machine Learning for Meal Delivery

Video recording: Youtube, Bilibili

Presentation slides: pdf

Abstract: Meal delivery is a prime example of an instant delivery service, in which an order has to be delivered only a short time after the order has been placed.  Meal delivery has given rise to many interesting optimization challenges.  We discuss a few of these challenges, but we focus on dynamically adjusting delivery capacity.  Actual order arrivals can differ substantially from expected order arrivals, which means, in the case of higher than expected order arrivals, additional delivery capacity may be needed.  When and how to add delivery capacity is challenging as it has to balance the cost of additional delivery capacity and the cost of missed orders or missed delivery promises.  We explore whether machine learning, specifically neural networks, can be used effectively to dynamically adjust delivery capacity in an instant delivery service environment.

Samitha Samaranayake

Assistant Professor, Civil and Environmental Engineering

Cornell University

Title: Emerging Mobility and Public Transit

Video recording: No recording for this special event

Abstract: Affordable, equitable and efficient access to personal mobility is a fundamental societal need—with broad implications for personal well-being, economic mobility, education, and public health. Emerging mobility services have disrupted the urban transportation ecosystem and instilled hope that new data-driven mobility solutions can improve personal mobility for all. While these apps provide a valuable service, as evident by their popularity, there are many questions regarding their scalability, efficiency, impact on equity, and negative externalities (e.g. congestion). On the other hand, traditional public transit systems provide affordable and community-oriented access to personal mobility, but have their own operational limitations. In this context, there have been many recent efforts to integrate emerging mobility services with public transit (e.g. first last mile microtransit). This talk will focus on fully integrating public transit operations with agile, demand-responsive services to enable personal mobility for all. We will discuss some of the underlying optimization problems (e.g. how they differ from ridepooling), present a general framework for formulating these problems, and explore the corresponding algorithmic (and non-algorithmic) challenges.

Meet best researchers in

Emerging Mobility Systems and Services

“Innovations in transportation have brought us ride-hail service, autonomous vehicles, bike share, carpooling, scooters, and more. New technologies are fundamentally changing the way residents and visitors get around.” 

SFCTA

Online seminar etiquette:

  1. To ask the speaker a question, please type in the Q&A function in Zoom. The moderator will collect questions and ask the questions after the talk.
  2. Please subscribe to our mail list to get the updates of seminars.

Seminar organizing committee

qiluo@cornell.edu

Qi Luo

Postdoc Associate, Cornell University
Assistant Professor, Clemson University (2021)

weima@cmu.edu

Wei Ma

Assistant Professor,
Hong Kong Polytechnic University

nmasoud@umich.edu

Neda Masoud

Assistant Professor,
University of Michigan

xtsun@umich.edu

Xiaotong Sun

Postdoc Associate,
University of Michigan

zhengtian@gwu.edu

Zhengtian Xu

Assistant Professor,
George Washington University

Participating institutes:
Media partners: