Abstract. The increasing volume and diversity of spatial data generated from heterogeneous sources such as geosensor networks, aerial and satellite sensors, and volunteered geospatial information necessitate the integration of these disparate datasets for informed decision-making in natural disaster management, particularly earthquakes. This study aims to design and implement Extract, Transform, Load (ETL) processes to integrate heterogeneous spatial data and visualize the results in management dashboards to enhance earthquake disaster management in Tehran Municipality District 2. Diverse datasets including building height, population, urban dilapidation, fault lines, building materials and structures, slope, and historical seismic events were collected and processed using Python libraries such as GeoPandas and Pandas, and stored in a PostgreSQL spatial database. In addition, seismic vulnerability modeling of buildings was conducted using the Random Forest algorithm and the results were presented through a Power Business Intelligence (BI) techniques dashboard. The novelty of this research lies in combining advanced data mining and BI techniques with customized ETL processes for spatial data and developing an intelligent dashboard equipped with machine learning algorithms to assist in analysis and enable interactive user exploration of various scenarios, thereby improving disaster resource management and prediction capabilities. The findings demonstrate that integrated data approaches and specialized BI tools significantly enhance the quality and speed of decision-making in earthquake disaster management. Over 1,200 spatial records were processed and integrated into a centralized database using the Python-based ETL, significantly enhancing analytical accuracy and response times in disaster management.
Abbaspour et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: