PHD Position in Planning Transport Systems under Uncertain Futures trough Agent-Based Simulations and Reinforcement learning

The goal of this doctorate is to advance the state-of-the-art of model-based planning support for transport and urban systems under uncertainty. The doctorate will develop new tools and methods that integrate large-scale agent-based modelling, surrogate modelling, scenario planning and sequential decision analytics. The research conducted during the doctorate will improve the long-term planning of urban and transport systems in Switzerland in close collaboration with the Center for Sustainable Future Mobility (CSFM) and the Chair of Risk, Safety, and Uncertainty Quantification at ETH Zürich. The main contributions of this research will be: 1) integrate both qualitative foresight scenario planning and quantitative scenario simulation workflows; 2) develop new spatially explicit surrogate models that enable the use of large-scale agent-based models for the simulation of a wide range of future scenarios, and 3) advance the use of sequential decision analytics (e.g. reinforcement learning or stochastic optimisation) for the identification of optimal adaptive plans.

The case study to be used in the research is the transport system of the Switzerland, where detailed MATSim agent-based models exist at the cantonal (Zurich) and national levels. Potential interventions to be evaluated include but are not limited to the expansion or repurposing of road, rail and cycling infrastructure, mobility on-demand services, mobility hubs and pricing schemes. The project will help tackling pressing questions such as how emerging transport modes, spatial development, and evolving travel behavior might shape Switzerland’s mobility system by 2060. These methods will allow key actors in Switzerland to identify critical risks, uncover novel opportunities, and co-create strategies that remain effective under uncertain future conditions.

The candidate is expected to work closely with other researchers in Zurich and is expected to participate in regular exchanges and coordination with the Swiss Federal Railways (SBB) and the Federal Office for Spatial Development (ARE).

The successful candidate for this PhD position will have a Master’s degree in spatial planning, urban analytics, transport planning, artificial intelligence or a related field, and will have experience with computer modelling, (e.g. agent-based modelling), as well as good grasp of machine learning, forecasting techniques, scenario planning, optimisation, R, python and GIS. Good knowledge of English is essential. Good knowledge of German is beneficial.