Paving Equity: Unveiling Socioeconomic Patterns in Pavement Conditions Using Data Mining

Abstract

Accessibility is a key metric in transportation equity, yet a critical aspect of accessibility is the quality of access. This study presents a data-driven social equity assessment of pavement conditions across the United States using the International Roughness Index (IRI) and data mining techniques applied to over 8 million records from the Highway Performance Monitoring System (HPMS). Data mining, as an exploratory tool, revealed hidden patterns that conventional methods might overlook. The analysis uncovered disparities in pavement conditions across socioeconomic and demographic groups. On average, road segments in the National Highway System with lower traffic volumes, higher minority populations, greater racial diversity, thinner pavement designs, and more non-English speakers tend to have poorer pavement conditions, as observed in the HPMS data. While the study identifies correlations rather than causal relationships, the findings underscore the inequities in access quality and highlight the need for transportation agencies to integrate social equity considerations into pavement maintenance and budgetary strategies. The observed gap is larger within clusters representing rural and small urban areas compared to urban sections. By factoring socioeconomic variables into decision-making processes, policymakers can better allocate resources to ensure equitable access to well-maintained infrastructure, regardless of community demographics, traffic levels, or other influencing factors.