Open-source METANET calibration tool enables reproducible freeway traffic simulation

A team from MIT (Monica Chan, Shreyaa Raghavan, Cathy Wu) has open-sourced the first comprehensive calibration toolkit for the METANET macroscopic traffic flow model. The tool solves a nonlinear programming problem using the IPOPT interior-point method, jointly estimating ramp flows alongside model parameters. This fills a critical gap: despite METANET's widespread use in traffic simulation and control, no open-source calibration pipeline existed, making results hard to reproduce or adapt to new networks.

Validated on two major real-world datasets—I-24 MOTION (one of the largest open-road trajectory instruments) and Caltrans PeMS (nearly 40,000 detectors across California)—the calibrated model captures complex traffic patterns including stop-and-go congestion waves. The code is publicly available on GitHub, supporting reproducible research and practical deployment for ramp metering and variable speed limit control. As sensor networks expand, this tool translates raw data into actionable models for intelligent transportation systems.