A techno-economic framework for resilient pavement infrastructure: a mechanistic-empirical approach for overload conditions

Premature failure of flexible pavements in developing nations, such as Pakistan, is a critical issue exacerbated by uncontrolled vehicle overloading and the limitations of traditional empirical design methods. This study presents a Mechanistic-Empirical (M-E) framework to address this challenge. Using the Michigan Flexible Pavement Design System (MFPDS), a comprehensive parametric analysis of 144 unique flexible pavement cross-sections was conducted by systematically varying the thickness of the Asphalt wearing course (2–10 cm) and binder course (10–25 cm). Pavement performance was evaluated based on critical mechanistic responses, including tensile strain at the bottom of the AC layer and compressive strain at the top of the subgrade, to predict fatigue and rutting life. A techno-economic optimisation was performed using a Cost-Benefit Ratio that links initial construction cost to predicted design life. Results indicate that the existing local design is critically inadequate, with a predicted service life of only 0.02 years. In contrast, the proposed M-E designs demonstrate that increasing AC layer thickness significantly enhances pavement life, with an optimal design achieving over 15 years of service. A key finding is that the Cost-Benefit Ratio decreases non-linearly, identifying an economic "sweet spot" beyond which further investment in thickness yields diminishing returns. Furthermore, the M-E framework highlights the high sensitivity of pavement life to the AC elastic modulus, indicating that improvements in material quality can yield significantly greater increases in durability than simply increasing material quantity. This study provides a validated M-E design framework tailored to local conditions, offering economically optimised solutions that are resilient to severe loading and making a data-driven case for a paradigm shift toward performance-based design and advanced material selection.