Spatial Clustering of Activity-Travel Fragmentation in California

Keywords: spatial clustering, activity-travel fragmentation, sequence analysis, travel survey
Session Type: Virtual Paper Abstract
Day: Monday
Session Start / End Time: 2/28/2022 05:20 PM (Eastern Time (US & Canada)) - 2/28/2022 06:40 PM (Eastern Time (US & Canada))
Room: Virtual 72

Abstract

This study employs sequence analysis to estimate a fragmentation indicator for people's daily schedules, spatial clustering to identify groups of observations with similar levels of fragmentation in space, and hierarchical clustering to identify distinct patterns of time allocation within each spatial cluster. Using the Local Indicator of Spatial Association (LISA), we discover that a substantial chunk of the National Household Travel Survey component in California has a spatial distribution fragmentation that is not different from a random scatter. We observe systematic and substantial differences in membership to these clusters based on the living environment, household structure, household income, and employment status for the component of the sample with significant spatial clustering of fragmentation. Within LISA spatial clusters, sequence analysis pattern recognition reveals a recurring pattern of time allocation that includes typical work and school schedules, as well as the pattern of staying at home. Each spatial LISA cluster is comprised of various time allocation clusters. Taken together, this analysis demonstrates significant and quantifiable variety in spatial clustering of fragmentation and the necessity for tailored policy responses across space. Policy initiatives should provide more activity opportunities and services surrounding residences while also relieving constraints for those with time shortages and allowing for greater time allocation flexibility in work and education, thereby alleviating stress-inducing time demands.