Creating Synthetic Data to Support Advanced Mobility Modeling for Small and Medium-Sized Urban Areas

The future of mobility is expected to be shaped by the integration of shared mobility services, vehicle electrification, and automated vehicle technology. To prepare for this transformation, it is crucial for policymakers and urban planners to envision the travel demand changes for emerging technologies and services. Traditional travel demand modeling approaches (e.g., the four-step travel demand modeling) can simulate the travel demand in aggregated zone (e.g., Traffic Analysis Zones) level, but fail to capture the detailed interactions between individuals and other elements in transportation systems including infrastructure and services. To address this challenge, planners and researchers are exploring advanced approaches for tracking an individual's daily activities and movements such as Agent/Activity-based Modeling (ABM). To support ABM, the common approach involves generating synthetic populations and travel demand that match aggregated numbers of the study population.

Nonetheless, creating synthetic data remains a significant challenge for researchers and planning practitioners, particularly in small and medium-sized urban areas (SMMAs) where resources are scarce. To tackle this issue, we have developed a framework with models and algorithms to generate inputs for ABM modeling in over 400 SMMAs (population < 500,000), by fusing  a range of data sources including the American Community Survey (ACS) data, Longitudinal Origin-Destination Employment Statistics (LODES) data, Microsoft building footprint (MBF) and SafeGraph point-of-interest (SGPOI) data, and the 2017 National Household Travel Survey (NHTS) data. Results are currently being processed and will be made publicly available in open format files along with comprehensive documentation and metadata. The main objective of this effort is to provide synthetic high-resolution population and travel demand data to facilitate the adoption of ABM by practitioners in the transportation and planning fields on SMMAs. The outcomes of the project can support research on public policymaking at disaggregated levels and for specific target populations and can also serve as a valuable resource for interested stakeholders looking to integrate future mobility technologies and services into their planning initiatives. By sharing this data and continually updating it with new and improved information, this project aims to foster advancements in mobility modeling and planning practices for more sustainable and efficient transportation.