AUFGEHTS Project
Published by open.bydata
About this dataset
This repository contains the outputs of the AUFGEHTS project, which explores a scalable, automated methodology for generating and calibrating mesoscopic traffic simulation models using Floating Car Data (FCD) as the sole input. You can use the data available in this repository under MIT License. roject Overview The core objective of AUFGEHTS was to develop a cost-effective and transferable modeling pipeline that enables: Calibration using commercially available FCD Seamless cross-border traffic modeling Conventional modeling approaches often break down at administrative or national boundaries due to fragmented data sources. AUFGEHTS aimed to address this gap by relying entirely on FCD, enabling simulation of traffic dynamics across jurisdictions. The A93/A12 corridor between Bavaria and Tyrol served as the primary test case—representing a heavily trafficked, cross-border segment of European transport infrastructure. Inputs to the model Data Source: Aggregated edge-based FCD from HERE Technologies, providing average speeds of probe vehicles per 5-minute interval. No trajectory-level data is used. Network Generation: Based on OpenStreetMap data (Feb 2024) combined with HERE’s map. Included road types: motorway, motorway_link, primary, primary_link. Demand Estimation: Generated using the AUFGEHTS pipeline, with speed data as the only external input. No sensor counts, OD matrices, or travel surveys were used. Model Performance Model performance was evaluated using GEH85 statistics, which compare observed and simulated flows: GEH85 range: 10–24 over a full-day simulation Benchmark: FGSV recommends GEH85 < 5 for high-quality calibration It is thus important to note, given these results, the model should be interpreted as a coarse approximation of traffic conditions rather than an operationally valid forecast. The results can be used in understanding the current capabilities and limitations of speeds-based calibration, exploring the traffic condition of the area. But it is not suitable for high-fidelity traffic planning or traffic impact assessments. How to Run the Simulation Ensure SUMO is installed and accessible via your system path. Then, run: sumo-gui -c sumo_withTrips.sumocfg This will generate: edge_data_AFG.xml – summary data for all vehicles edge_data_AFG_LKW.xml – summary data for heavy goods vehicles routes_AFG.xml – generated route definitions Further outputs can be configured in the .sumocfg and additional.add.xml files. Contact For any questions, please contact: Gabriel Tilg gabriel.tilg@transcality.com