Road maintenance optimization using a probabilistic approach calibrated with 15-year monitoring data

Efficiently maintained transportation infrastructure system dictates the hallmark of a well- functioning economy. However, road degradation is often unpredictable and demands frequent and costly maintenance activities. Despite substantial annual investments, practitioners struggle to adopt the optimum maintenance strategy due to a lack of knowledge on cost over socio-economic benefit. Therefore, this study proposes an optimum pavement intervention strategy by analyzing the wealth of monitoring data collected over 15 years (2006–2021) using a probabilistic pavement degradation framework based on roughness [International roughness index (IRI)]. The distribution of the IRI values is evaluated based on road clusters categorized by location, road use type, and pavement maintenance categories (PMC). One-way ANOVA is used to identify patterns of IRI influenced by external factors like rainfall, terrain type, and drainage condition and the results emphasize the significance of external factors in road pavement IRI variation. The calibrated discrete state Markov models are used to predict road conditions including associated costs and offer a detailed comparison of available intervention options aiding intervention prioritization. The study suggests that prioritizing the worst sections may not yield the best cost-benefit ratio; instead, maintaining all sections fairly is more cost-effective. These findings can be used in pavement management systems to rank intervention options based on cost and benefit.