Time Matters: Integrating Temporal Constraints into Essential Service Accessibility Equity Analysis

• In Event: Governance, Digital Inequality, Service Accessibility, and Lived Experience

Abstract

Urban land use and transportation planning policies increasingly rely on accessibility metrics to determine where to locate public transit, food stores, and essential services. However, most commonly used models are static, assuming people are always available to travel, begin trips from home, and that transit services and businesses operate continuously. These assumptions overlook how people actually live: many of them work non-standard hours, rely on limited transit, and may only be able to travel when services are closed.
As a result, this disconnect leads to an overestimation of real-world accessibility, and worse, misinformed policy and planning decisions. For example, a grocery store may be geographically close to shift workers' homes but unreachable for them because most stores are closed and transit is unavailable when they get off work. Policymakers would be blind to the barriers that shift workers are facing if they only rely on static accessibility measures.
To address this problem, this study develops a time-sensitive, scalable, people-based accessibility model to account for the time aspect of individuals' schedules, available travel modes, and businesses' hours. Using the Southern California Association of Governments (SCAG)'s Activity-Based Model, Point of Interest (POI) data with opening hours, and street networks from OpenStreetMap (OSM) across Los Angeles County, I compute time-sensitive accessibility metrics to essential services at both traffic analysis zones (TAZ) and individual levels.
Findings show that essential services access varies drastically across time and space in Los Angeles. In particular, it drops sharply during early morning and late night, revealing potential "temporal access deserts" that static models fail to detect. As an example, the Gini index of 15-minute transit access to groceries jumps from around 0.58 at noon to around 0.95 at 1 a.m. In addition, regression results show that low-income and African American neighborhoods—despite having reasonable daytime transit access to groceries—become significantly underserved after 9 p.m., revealing a time-based disparity that static models overlook. By identifying not only where, but also when and who are experiencing a lack of access, this paper addresses a blind spot in public policy making and provides a new tool to guide transportation and land use planning. Such insights can inform targeted policies such as adjusting store hours, improving off-peak transit frequency, or initiating mobile service deliveries to better serve disadvantaged populations. 

Author

• Yuquan Zhou, University of Southern California
  Presenting Author