HNTB

While granular roads play a key role in rural transportation systems, they are highly vulnerable to damage caused by superload vehicles such as Implements of Husbandry (IoH) and Superheavy Loads (SHL). These vehicles often exceed standard weight and size limits, causing non-standard stress conditions that accelerate granular road deterioration. To address this issue, this study introduces RISAT (Road Infrastructure-Superload Analysis Tool), a data-driven, user-friendly tool developed to help engineers and local agencies evaluate the structural and economic effects of superloads on granular roads. RISAT integrates mechanistic analysis with Artificial Neural Network (ANN) models to predict critical road responses and assess performance indicators such as rutting damage, road damage costs, number of load repetitions, and service life reduction. The ANN models were developed using more than 3,200 cases representing simulated superload traffic and granular roadway structural conditions and subsequently optimized through various back-propagation algorithms. Field validation was conducted using a test site in Iowa, where in-situ data was collected and compared with Layered Elastic Theory (LET)-based predictions. Comprehensive validation showed that RISAT predictions closely align with conventional LET outcomes, confirming RISAT’s reliability for evaluating granular road performance under diverse superload conditions. Although RISAT was developed under Iowa conditions, the tool can be adapted to other regions by modifying input values to match local road structures, traffic types, and environmental conditions. RISAT offers a practical solution for researchers and transportation agencies aiming to manage unpaved road networks more effectively and economically.