Assessment of pavement distress and performance under mixed traffic in urban highways: an empirical field study

Thesis Abstract

Urban highways in rapidly urbanizing contexts experience diverse, mixed-traffic conditions that challenge pavement performance and management practices. The study addresses the problem of accelerated distress and degraded service levels caused by heterogeneous vehicle types, fluctuating loading patterns, and limited maintenance budgets, which complicate predictive modelling and asset management. The aim is to quantify pavement distress evolution and performance under mixed traffic and to develop a practical, data-driven framework for prioritizing maintenance interventions in urban arterial networks. Specific objectives are (1) to characterize traffic composition, loading spectra, and their temporal variability on selected urban highways; (2) to quantify the spatial and temporal progression of distress (rutting, cracking, potholing, and roughness) using longitudinal surveys and non-destructive testing; (3) to identify the relationships between pavement distress indicators and traffic loading, pavement materials, and structural capacity; (4) to calibrate a predictive performance model integrating traffic, material properties, and structural response; and (5) to formulate a decision-support framework for prioritizing maintenance under budgetary constraints. The methodology adopts an empirical field study design conducted on three urban arterial corridors with contrasting traffic patterns and pavement ages in a mid-sized metropolitan region. The population comprises asphalt and flexible pavement sections on these corridors, with a targeted sample of 60 standard inspection segments and 180 non-destructive test locations. A stratified sampling approach ensures representation across lane types (fast, mixed, and heavy-vehicle lanes), pavement ages (5–15 and 15–30 years), and traffic volumes. Data collection instruments include standardized distress surveys (Pavement Condition Index and International Roughness Index), falling-weight deflectometer (FWD) tests to assess structural capacity, high-resolution distress imaging for automated cracking analysis, dynamic traffic counting and vehicle classification systems to capture loading spectra, and material characterization in the lab (resilient modulus, moisture susceptibility, and asphalt binder grade). Validity and reliability are ensured through inter-rater calibration for distress ratings, repeated FWD testing, and instrument precision checks. Data analysis employs a mixed-methods approach quantitative analysis uses multivariate regression and generalized linear models to link distress indicators with traffic loading, structural capacity, and material properties; time-series analysis evaluates distress progression over two years; and ANOVA tests compare differences across corridors and pavement ages. A discrete event simulation models the impact of interventions under budget scenarios. Where appropriate, aroma of the theoretical framework anchors the empirical work to theories of pavement performance and traffic loading, notably the Vulnerability Theory of Infrastructure under Mixed Traffic and the Structural Number concept adapted to contemporary modelling. A conceptual framework and empirical model specification are explicitly tested, with model performance assessed via R-squared, RMSE, and cross-validation. Expected findings include (i) higher rates of rutting, pothole formation, and roughness in sections with heavy mixed-traffic exposure and aged pavements; (ii) robust associations between FWD-derived structural response, axle-load distributions, and observed distress patterns; (iii) material properties and moisture susceptibility significantly moderating distress progression under repeated loading; and (iv) a parsimonious predictive model capable of forecasting distress evolution and aiding maintenance prioritization. The study is anticipated to reveal threshold loading levels beyond which distress accelerates, enabling proactive interventions. Contributions to knowledge include an empirically validated framework linking mixed-traffic loading to pavement distress in urban arterial contexts, enhanced understanding of how material and structural factors interact with traffic to drive performance, and a decision-support tool that integrates physical measurements with budget-constrained maintenance planning. The main conclusion is that mixed traffic intensities and legacy pavement conditions jointly govern distress trajectories, demanding integrated monitoring and targeted rehabilitation schedules. Recommendations emphasize implementing routine traffic loading characterization, adopting adaptive maintenance strategies based on model predictions, investing in pavement materials with improved moisture resistance, and expanding non-destructive testing coverage to enable timely interventions.