FABILUT: Flexible Agent-Based Integrated Land-Use Transport model

• Applicants: Professor Dr.-Ing. Rolf Moeckel; Professor Dr. Kai Nagel
• Subject Area: Traffic and Transport Systems, Intelligent and Automated Traffic
• Term: from 2016 to 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 322166923

Integrated land use/transport models are widely used to analyse the impact of policy scenarios, infrastructure investments and global trends in a simulation environment before they happen in reality. Integrating transport with land use models allows representing the land use/transport feedback cycle in both directions: On the one hand, the location of households and jobs defines the origins and destinations of most trips. On the other hand, the ease of access, which is driven by travel times, influences the attractiveness of locations for households to live and firms to locate. Only the integration of transport with land use models allows the realistic representation of this feedback cycle in a simulation environment. Such integrated land use/transport models have been implemented and applied since over 50 years. However, existing approaches attempt to represent individual behaviour with highly aggregated solutions. In a few exceptional cases, microscopic models were used. Yet the coupling between the models has always been implemented in aggregate form, a shortcoming this project aims at overcoming. In this research, a microscopic transport model will be integrated with a microscopic land use model. For the first time, this integration will be accomplished entirely based on the exchange of microscopic data at the individual level. The microscopic transport model MATSim will be integrated with the also microscopic land use model SILO. In this integration, individual travellers defined in SILO will make trips that are assigned to the network in MATSim. Vice versa, MATSim will provide individual travel times for all household members when searching for a new housing location. This ensures that a commuter who has to be at work by 6:00 am sees different travel times and modal options than someone going to work at 9:00 am. This microscopic coupling is expected to represent more efficiently and more realistically the impact of travel times and congestion on household location choice and travel behaviour. The integrated MATSim/SILO modelling suite will be implemented in four study areas: Munich (Germany), Dublin (Ireland), the State of Maryland (USA) and Cape Town (South Africa). The diversity of study areas will help develop universally valid and transferable solutions for integrated land use/transport models. Furthermore, results are expected to be transferable to other modelling subjects, such as health, tax revenue or climate models.

• DFG Programme: Research Grants
• International Connection: Japan
• Cooperation Partner: Professor Dr. Masanobu Kii