MSD CoP Webinar: Graph Neural Networks for Modelling Water Distribution Systems (and beyond)

This webinar was hosted by the MultiSector Dynamics Community of Practice (MSD CoP; https://multisectordynamics.org). Abstract: Graph Neural Networks (GNNs) are particularly well-suited for modelling networked infrastructure systems - like water networks, power grids, and transportation systems - because they can incorporate the underlying topology and interactions between components. In this talk, we’ll begin with a quick primer on how GNNs work and why their structure lends itself to transferability - learning patterns on one network and applying them to others. We’ll then explore how we-ve applied GNNs to water distribution systems, progressively embedding more physics into the modelling process. We’ll close by looking at how we employed GCC-based methods in other domains, such as urban drainage systems, spatiotemporal flood modelling over meshes, and distributed hydrological models, showcasing how this approach can scale across water systems. Presenters: Dr. Riccardo Taormina (Delft University of Technology) Moderator: Stefano Galelli (MSD CoP WG Co-Lead), David Gold (MSD CoP WG Co-Lead), Jillian Sturtevant (MSD CoP WG Communications Officer) The webinar was held on: April 24th, 2025 from 11 AM - 12PM ET.