Fostering urban resilience and accessibility in cities: A dynamic knowledge graph approach
- Development of versatile tool for disaster and city planning.
- Disaster response system that focuses on static and dynamic planning through agents interplay.
- Holistic 15-minute city planning by status quo monitoring and scenario planning.
- Cross-domain analyses in planning via the integration of complementary data sources.
This paper explores the utilisation of knowledge graphs and an agent-based implementation to enhance urban resilience and accessibility in city planning. We expand The World Avatar (TWA) dynamic knowledge graph to support decision-making in disaster response and urban planning. By employing a set of connected agents and integrating diverse data sources — including flood data, geospatial building information, land plots, and open-source data — through sets of ontologies, we demonstrate disaster response in a coastal town in the UK and various aspects relevant to city planning for a mid-sized town in Germany using TWA. In King’s Lynn, our agent-based approach facilitates holistic disaster response by calculating optimal routes, avoiding flooded segments dynamically, assessing infrastructure accessibility before and during a flood using isochrones, identifying inaccessible population areas, guiding infrastructure restoration, and conducting critical path analysis. In Pirmasens, for city planning purposes, the knowledge graph-driven isochrone generation provides evidence-based insights into current amenity coverage and enables scenario planning for future amenities while adhering to land regulations. The implementation of agents and knowledge graphs achieves interoperability and enhances urban resilience and accessibility by enabling cross-domain correlation analysis that extends various areas including geospatial buildings, population demographics, accessibility coverage, and land use regulations.
- This paper draws from preprint 320: Fostering Urban Resilience and Accessibility in Cities: A Dynamic Knowledge Graph Approach
- Access the article at the publisher: DOI: 10.1016/j.scs.2024.105708
Associated Themes:
