Improved Method for Validating the Simulated Travel Behavior by Using Both the Smart Card and Navigation Data

Abstract

Incorrect estimations of travel time, link traffic volume and trip distribution due to the inaccurate report from the survey data relying on respondents' memory are particularly problematic when an activity-based travel simulator conducts any sensitivity analysis. The dairy data of Seoul for example reports travel times longer than that of the four-step network assignment. Owing to the rapid development of big data engineering, the data of public transportation use is often used for correcting the estimation by the diary data only. The smart card data however includes only the public mode use but does not cover the car use information. The paper aims to validate an activity-based travel simulator, FEATHERS Seoul (FS), by analyzing the simulated travel behavior against the smart card and car navigation data, recently collected in Seoul.