UNDERSTANDING URBAN MOBILITY FLOW: EXTRACTING MOBILITY PATTERN FROM AGGREGATED MOBILE PHONE LOCATION DATA

Abstract

Urban studies often place excessive emphasis on the practical implications of the analysis, leading to a decreased focus on exploring methodologies for examining urban phenomena such as mobility. This often happens that studies relied too heavily on the foolproof traditional data such as origin-destination data which demand substantial resources for collection, excusing other means of understanding. This article specifically investigates this dynamic of exploring the use of aggregated mobile phone location data such as Mobile Spatial Statistics^TM (MSS), an underutilized resource in mobility analysis, to explain mobility flow characteristics of urban populations. By employing the theoretically-sound Wasserstein distance approach, the paper pursues to extract the mobility characteristics inherent in the dataset to explain a dominant flow—a primary patterns and directions that suggest movement within areas. Through various tests and applications, this research enhances the urban studies with new ways of understanding urban mobility data.