Data-Driven Approach to Increase the Spatial Resolution of Household Travel Surveys

Abstract: Household travel surveys have long been a primary data source for urban mobility modeling and transportation planning due to their wide availability and accessibility. As cities worldwide strive to transition towards sustainable urban mobility, a detailed understanding of travel behavior is essential. However, privacy regulations, such as the General Data Protection Regulation (GDPR) in Europe, restrict access to detailed coordinate-level travel data, limiting spatial resolution to Traffic Analysis Zones (TAZs). This paper introduces a heuristic algorithm to enhance the spatial resolution of household travel surveys by estimating probabilistic coordinates for each trip's origin and destination. The algorithm sequentially estimates the coordinates for an individual’s entire trip chain using two primary data sources: travel survey data and land use data, processed through ArcGIS and Python. The methodology links TAZ-level travel data to point-level building data, categorized by trip purposes, and enriches the survey data using an objective function that minimizes discrepancies between reported and calculated distances. Applied to the 2022 Trondheim household travel survey, the algorithm demonstrates a more detailed and accurate representation of travel behavior compared to TAZ-level data. Results are validated against data collected via mobile applications, showing strong potential for improving the utility of traditional travel surveys. A data-driven approach highlights the advantages of high-resolution travel data, especially in modeling active mobility and last-mile services, which are critical for sustainable urban transport systems.

Keywords: Spatial Resolution, Household Travel Survey, Heuristic Algorithm, Sustainable Transport