TENGOS

Transport and Epidemic Networks: Graphs, Optimization and Simulation

The ongoing COVID-19 pandemic shows that maintaining critical infrastructure during such a scenario is imperative to preserve essential basic services and to limit economic harm to a minimum. Accordingly, authorities steadily balance between preserving the functionality of infrastructure and confining epidemic spreading. For transportation networks, which remain crucial for the movement of people and goods to keep the economy and daily supply operational, finding the right balance between optimal functionality and reducing infections remains an inherently complex task; especially, as transportation networks remain major foci of infection - in particular with recent ambitions towards sustainable and shared transportation, which focus on public transport and ride pooling. In this context, two questions are central. At tactical level, it is crucial to understand which combination of operating modes and restrictions allow us to limit infections in order to allow for safe operations. At strategic level, the design of transportation networks remains a central challenge.
Transport networks shall allow for sustainable operations during normal times but also for robust and resilient operations in extreme situations when epidemic mitigation strategies are in place. To be efficient, the networks should not rely on parallel infrastructure or redundant capacities. To answer these questions, we study multilayer network dynamics by coupling epidemic networks and transportation networks, where each person that travels in a transportation network is also part of an epidemic network. To allow for a holistic assessment and decision support, we combine methodologies from three disciplines: epidemic modeling, transport optimization, and transport simulation. Specifically, we combine two dynamical systems modelling epidemic networks in combination with ow-based transportation network models. We develop optimization-based algorithms that allow to design and operate robust transportation networks during epidemics, we study reduced multilayer transport-epidemic dynamics via differential equations, and we benchmark our results against agent based transport simulations. This multifaceted approach allows us to quantify the impact of changes in the transportation networks and  perational concepts. We also use this algorithmic environment to design and test new transport networks and potential epidemic mitigation strategies to develop efficient and safe transportation systems for a new normal.