Research

The MoCA Lab studies how people, platforms, and policies interact in modern mobility systems.
We build models, algorithms, and decision tools that help society design transportation systems that are efficient, fair, and truly human-centered - especially as AI and digital platforms increasingly shape how we move.

This page introduces our vision, the scientific foundations behind it, and the kinds of research problems students can work on.


Overview

Decision-makers

  • Travelers choosing routes, modes, and departure times
  • Operators & platforms managing fleets, routing, pricing, and service quality
  • Regulators designing policies and ensuring public goals are met

Time scales

  • Immediate actions: what travelers and platforms do right now
  • Interactive dynamics: how their actions influence each other over time
  • Long-term design: how policies and technologies shape the future system

Dimension X

Every mobility problem has a context or mechanism that defines the situation. We call this Dimension X, and it can be things like multimodal mobility, pricing structures, or new technologies such as AI-driven agents.

Together, these ideas form our research tensor — a conceptual map of the many challenges in modern mobility.

MoCA Research Tensor

The tensor helps us think systematically about the problems we study and how they connect to one another.
It also helps students find a topic that excites them while staying aligned with the broader research program.


1. Who we work with and why

Our research focuses on questions that matter to:

Policy makers

who want sustainable and equitable cities

Transport agencies

who manage public infrastructure and operations

Mobility platform operators

(e.g., on-demand services, navigation tools)

Travelers

who experience the system every day

These groups all influence each other, and their decisions collectively shape congestion, access to opportunities, emissions, and the overall travel experience.

By understanding these interactions, we aim to help build mobility systems that work for people and society, not only for algorithms or market forces.


2. How we think about mobility systems

We believe that mobility systems are best understood by combining:

Behavioral modeling

How people and platforms make decisions

Network modeling

How actions spread through the transportation network

Policy analysis and design

How decisions influence societal outcomes

The MoCA Lab integrates these perspectives into one research program.


3. How we conduct research

We approach each problem through a flexible three-step process:

1

Theory & Methodology

We develop new models and algorithms that can represent realistic behavior, handle complex networks, and scale to large inputs.

This includes advances in choice modeling, network modeling, and equilibrium analysis.

2

Modeling the real system

We take real-world mobility challenges and identify their essential trade-offs.

This helps us focus on what truly matters and avoid unnecessary complexity.

3

Data & Application

We use data to estimate models, validate predictions, and evaluate different policies or platform designs.

We then use the insights to improve our theory in a continuous feedback loop.

Continuous feedback loop


4. Open Projects


5. Research Directions

Although our lab is young, several research directions already guide our work.

5.1 Behavioral foundations

We study how travelers and platforms make decisions, and how to model these decisions in a way that is flexible, scalable, and consistent with real-world behavior.

This includes modeling:

  • preferences and trade-offs between choice alternatives under different contexts
  • behavioral and physical constraints and habits
  • rich patterns of substitution and heterogeneity

These behavioral foundations support everything we build on top.


5.2 Dynamic decisions and adaptation

Mobility is not static. Travelers learn, adapt, and respond to new information and changing conditions. Platforms adjust prices, allocate vehicles, and update routes in real time.

We study:

  • sequential decision-making
  • repeated interactions
  • adaptive behaviors
  • dynamic system feedback

This helps us understand not only what people do today, but how their behavior evolves over time.


5.3 Information and guidance

Digital platforms now influence travel through suggestions, notifications, and personalized guidance. These tools can improve travel, but they can also reshape network conditions or create new inequalities.

At a high level, we study:

  • how travelers respond to guidance
  • how guidance affects network conditions
  • how to evaluate interventions from a public-interest perspective

5.4 Equilibrium, stability, and system design

When many agents interact, unexpected system-level behaviors can emerge — stability, oscillations, congestion surges, and more.

We study:

  • demand–supply equilibrium
  • pricing and incentive mechanisms
  • multimodal system integration
  • conditions for stability and robustness

These insights support the design of better and more resilient mobility systems.


5.5 Human–AI mobility ecosystems

Digital tools, navigation apps, and emerging AI assistants increasingly participate in mobility decisions. As these technologies become part of everyday travel, mobility systems may evolve in new and unexpected ways.

At a high level, we are interested in:

  • how digital tools and automated systems influence mobility choices
  • how human and non-human decision-makers coexist in mobility networks
  • what opportunities and challenges arise as these technologies become more widely used

This direction helps us prepare for mobility systems where humans and algorithms share decision-making roles.