Origin-destination (OD) data is fundamental to mobility and transportation analysis, providing critical insights into travel demand patterns, infrastructure planning, and mobility algorithm performance evaluation. However, real-world OD datasets are often sparse, inconsistent, or inaccessible due to privacy constraints. This project addresses these challenges by synthesizing OD data, enabling enhanced benchmarking and reproducibility in transportation research. Our proposed approach comprises two complementary synthesis methodologies: forward and backward. The forward synthesis method focuses on developing generative algorithms that produce realistic OD demand patterns from static datasets or high-level properties.

Given inputs such as population density, census data, and a city’s transportation network, we aim to generate mobility demand data that is spatially and temporally consistent with these inputs. This enables controlled scenario generation, allowing researchers and practitioners to evaluate mobility solutions under diverse conditions. On the other hand, the backward synthesis method operates on existing OD demand datasets, either decomposing them into multiple datasets with varying characteristics or extending them to enhance spatial and temporal granularity. Decomposition acts as a form of down-sampling, extracting distinct demand patterns from dense datasets, while extension enables up-sampling, refining aggregated or incomplete datasets to integrate multiple sources and improve analytical fidelity.

By advancing OD data synthesis, this project enhances the reproducibility and accessibility of mobility research. A key outcome is the development of an open-source tool that democratizes access to high-quality, synthetic mobility datasets, facilitating robust algorithm benchmarking and broader adoption in transportation modeling.

I’m participating in SuperUROP for the opportunity to advance my knowledge in the transportation field specifically as it relates to transportation modeling & data generation. Through this program, I want to apply my experience from previous UROPs and my urban planning background in order to develop a platform for mobility data synthesis. I’m excited to participate and contribute to this project.