Simulating Accessibility from CAVs and ICTs (SACI) develops a simulation tool that helps transportation agencies understand and plan for the transformative impacts of connected and automated vehicles (CAVs) and information and communication technologies (ICTs) on travel behavior and network demand. As CAVs and ICTs reshape how people choose destinations and routes, new models are needed to predict future network demand and usage.
The project develops a first-of-its-kind framework that captures cognitive, perceptual, and behavioral effects of CAVs and ICTs on travel, drawing on recent transportation and social science research. This framework is implemented in an agent-based model using an integrated simulation suite—SILO for land use, MITO for travel demand, and MATSim as a transport simulator—calibrated with multimodal transportation network data from the DC, Maryland, and Virginia region.
The final product is a flexible microsimulation tool designed for use by state, regional, and local DOTs that enables users to test a wide range of policy scenarios, CAV adoption rates, and regulatory frameworks. The modeling framework incorporates AI-based methods for population synthesis and data integration, improving the realism and resolution of the simulations. Technology transfer will occur through workshops, webinars, technical briefs, and an online distribution site for the simulation tool.
Universities Involved
University of Virginia
University of Maryland
Principal Investigators
Andrew Mondschein
Cinzia Cirillo
Expected Research Outcomes & Impacts
The project will deliver an advanced microsimulation tool that enables transportation agencies to test policy scenarios for CAV-ICT deployment and assess impacts on travel behavior, land use, and network demand across the mid-Atlantic region. The modeling framework incorporates AI-based methods for population synthesis and data integration, improving behavioral realism and spatial resolution. These models can shape policies and infrastructure investment decisions at multiple scales, identifying locations for investment and potential CAV-efficient land use strategies.
At least two graduate students will be supported at UVA and UMD, gaining experience with advanced data analytics, microsimulation modeling, and AI-based methods applied to real-world transportation problems. The technical and analytical competencies acquired position students to make immediate and long-term contributions to the transportation workforce.
Subject Areas
Connected and Automated Vehicles, Travel Behavior, Land Use Modeling, Agent-Based Simulation, Transportation Planning