A new model and approach for assessing equity in public transport

Abstract

Public Transportation is a cornerstone of urban mobility, ensuring access to essential services, employment, education, healthcare, and social participation for millions worldwide. However, the equitable distribution of public transport services presents a significant challenge, particularly in rapidly urbanizing regions with pronounced socioeconomic disparities. This thesis investigates the concept of public transport equity in Kocaeli, Turkey, using Smart Card Data (SCD), General Transit Feed Specification (GTFS), and Household (HH) data to critically assess the effectiveness of current travel-based transport models. A fundamental gap in traditional models is identified, as they often overlook the travel patterns of occasional users, thereby failing to account for the needs of the entire spectrum of public transport users. This omission challenges efforts to create inclusive and sustainable urban transport systems aligned with Sustainable Urban Mobility Plans (SUMP). The research methodology is divided into multiple phases, encompassing data extraction, cleaning, clustering, and merging of datasets. By integrating SCD, GTFS, and HH data, the study creates a unified dataset that combines transit behavior with socioeconomic indicators. Key analytical tools, such as the Public Transport Supply Index (PT index), Lorenz curves, Gini coefficients, and the Suits index, are employed to evaluate the equity of public transport services across different zones. The introduction of the frequency-weighted origin-destination (OD) matrix and Inverse Probability Weighting (IPW) method allows for a more nuanced analysis of travel demand, particularly accounting for the variability in travel behavior across frequent and infrequent users. By segmenting user groups based on onboarding patterns, card types, usage frequency, and weekday versus weekend travel, distinct profiles for regular and occasional users are developed. These user groups are analyzed both for daily data distribution and route-based representation. The trip chain method, used to construct an OD matrix, uncovers service gaps, highlighting how current models disproportionately favor frequent users in policy-making and resource allocation. This results in disparities in service, particularly for vulnerable populations such as low-income residents, the elderly, and those living in peripheral areas. The study's findings demonstrate that infrequent users, often reliant on public transport for essential trips, are underserved, while policy-making skews toward the needs of regular commuters. By incorporating frequency-based equity measures, such as the IPW-adjusted OD matrix, the thesis proposes a more equitable framework for transport planning. This new model advocates for a user-centered approach, where service frequency and socioeconomic factors are integrated into decision-making processes. The research concludes with recommendations for improving service frequency in underserved areas, enhancing multimodal integration, and adopting dynamic planning approaches that respond to real-time user data. These strategies aim to create a more equitable public transport system, ensuring that services benefit all city residents, including marginalized groups and occasional users. Ultimately, the thesis contributes to both academic discourse and practical applications in urban transport planning, advocating for sustainable and equitable urban mobility solutions