Under the combined influence of urban heat islands and climate change, extreme heat increasingly impacts travelers, especially vulnerable populations such as older adults, people with disabilities, and those with chronic diseases. While many DOTs, MPOs, and transit agencies have attempted to address this issue, there is a mismatch between their goals of enhancing mobility and accessibility and their use of design-based or asset-based approaches. The goal of this dissertation is to develop a comprehensive system to evaluate the cumulative heat impact on travelers at the trip level, with a special focus on vulnerable populations including elderly and people with disabilities.

There have been two broad genres of studies that attempt to tackle this problem. At the micro-level, urban designers and physiologists use physiological simulation and computational fluid dynamics to model building- and block-level air ventilation and thermodynamics. Their approach provides tangible and quantifiable outcomes but is often limited in scale and constrained by the number of criteria or hazard types included. A more comprehensive analysis at the macro level, led by urban planners, often uses qualitative approaches like surveys to obtain people’s opinions over an entire trip. However, these are often constrained in quantifiability. Recently, some researchers have started to combine these approaches for a “meso-level” analysis that addresses both comprehensiveness and tangibility, using transportation models and large-scale meteorological data. Yet, these approaches are primarily additive, missing the dynamic nature of transportation trips and the sophisticated mechanisms of heat exposure risk accumulation.

Despite increasing temperatures and impacts on transit users and pedestrians, a comprehensive system to quantify and evaluate heat impact on travelers at the network level does not currently exist. The goal of this study in addressing this issue is to develop comprehensive and quantifiable estimates of people’s thermal comfort during transit and walking trips at the network level to connect with theory and practice.

The dissertation approaches the problem with a comprehensive framework composed of three analytical modules. The first module conducts computer vision based analysis to identify sidewalk locations, delineate sidewalks, and model tree and building shades on sidewalk at different times of day throughout a year. The second module builds TransitSim and SidewalkSim to derive second-by-second transit trip trajectories and generate detailed activity profiles. The third module retrieves and extrapolates the meteorological data and applies National Weather Service’s standards to the calculation of thermal comfort indices. The new ThermoRoute Analyzer program estimates cumulative exposure based on NIOSH’s Work/Rest schedule.

The effectiveness of the analytical modules is demonstrated through two case studies. The first case study involves an exposure analysis using data from over 40,000 samples collected in the Atlanta Regional Commission’s Transit Onboard Survey, which is representative of transit travel conditions on an average weekday. By calculating the cumulative extreme temperature exposure experienced during each trip and assessing the associated thermal risk, this case study assesses the overall severity of exposure in the region, highlights disparities across different demographic groups, and examines trends under future climate change scenarios. The second case study focuses on a constrained area in Downtown Lawrenceville, applying shortest path routing for hypothetical walk trips on July 21, 2024, the hottest day recorded in Earth's observatory history at the time of this dissertation's publication. This case study provides insights into potential changes in travel behavior when individuals are informed about their thermal exposure, illustrating how access to such information might influence decision-making.

This research contributes to the fields of transportation resilience and urban thermal comfort in three ways. Theoretically, this research enhances the understanding of travel decisions and network-level resilience concerning cumulative health risks, and bridges the gap between health risk assessments and transportation engineering. Methodologically, the study integrates analytical techniques across disciplines into a framework to address the comprehensive, tangible, and dynamic needs for trip- and individual-level exposure analysis. Practically, the tools developed are all open source, and the secondary data are all freely accessible online, allowing transportation and planning agencies to use the tools with their own primary data. The findings provide planning agencies with concrete and actionable insights.