Extracting and Visualizing Individual Space-Time Paths: An integration of GIS and KDD in Transport Demand Modeling

The disaggregate activity-based approach for transport demand modeling requires the acquisition, management, analysis, and visualization of very large, multivariate spatial datasets in order to capture and extract meaningful spatio-temporal patterns and relationships. Knowledge Discovery in Databases (KDD) is a conceptual and methodological framework that was developed in the last decade to address the issue of transforming large amounts of raw geographic data into knowledge. Based on the KDD framework, this paper describes the steps involved to build individual space-time paths from an origin-destination survey and presents the functionalities of an object-oriented GIS prototype to extract and dynamically visualize individual space-time paths. The prototype sustains seamless spatio-temporal queries to the database and provides cast-based animation that mimics continuous individual movement on the street network.