Developing a predictive maintenance model for the rehabilitation of roads’ assets

Abstract

Maintaining road infrastructure is a significant challenge in minimizing public fund waste from reactive maintenance. Preventive maintenance is considered a new era in this field. To achieve cost savings and prolong the life cycle of road networks, it's crucial to determine their lifespan at the time of construction. This article proposes an emerging predictive methodology to select road network rehabilitation policy. This methodology facilitates long-term planning based on technical and budgetary perspectives to identify the most cost-effective rehabilitation policy for the road network's assets over the planning horizon. To accomplish this, the Pavement Rehabilitation Predictive Model (PRPM) is developed for long-term forecasting of future asset conditions as the network deteriorates over time. It takes into account effective planning for different rehabilitation policies predicts asset deterioration, and aids in making informed decisions by identifying the best rehabilitation policy for pavement sections. An empirical example is used to illustrate the proposed approach in detail, based on the evaluation of three scenarios of the rehabilitation policies. The results indicate that the prediction model is highly efficient, resulting in an average budget savings of up to 20% compared to other relevant maintenance methods.