Integrating CyberGIS-Jupyter and spatial agent-based modelling to evaluate emergency evacuation time

As large-scale disasters are increasing in severity and frequency, agent-based modeling enables the simulation of disaster and evacuation processes, while exploring the complex interactions of disasters and human behaviors. In this paper, we employ CyberGIS-Jupyter for spatially explicit agent-based modeling to examine dynamic associations between disaster severity and evacuation processes. We find that as the disaster severity increases, the total time for all vehicles to evacuate increases as more vehicles are become stuck. We find that CyberGIS-Jupyter can simplify cyberinfrastructure access to conduct agent-based modeling of emergency evacuation while enabling intuitive sharing and presentation of model components and results and fosters the reproducibility and replicability of agent-based modeling with data- and computation-intensive geospatial problem solving.