Pavement maintenance and rehabilitation programming and budget allocation from strategic perspectives: Multiobjective optimization models

The current roadway system was built several decades ago with billions of dollars, numerous tangible and intangible resources, and enormous effort. However, the massive infrastructure assets started to exhibit the symptoms of declining quality and become more vulnerable to unforeseen climatic conditions. As such, highway agencies adopted two-step approaches for highway assets, focusing on the preservation of existing assets and developing plans to secure the future. However, the task of preserving in-service assets became quite demanding for the managing authorities. These difficulties in managing transportation are due to the limited budget dollars for transportation asset management, particularly pavements, fluctuating construction markets, ever-changing policy, rising concern over the safety of highways, the need for integrating sustainability and inequity, and a lack of engagement from stakeholders. Given the current circumstances, highway authorities must define their strategic themes to integrate these concerns into the pavement management system. Therefore, the transformation of traditional pavement management practices into an interactive and inclusive system based on strategic themes can both optimize pavement performance at the system level and coherently address the growing issues. This dissertation proposes four optimization models to establish management plans and allocate limited budgets for pavement maintenance and rehabilitation programs. This study considers multiple objectives such as preserving pavement in good conditions, management of safety, and energy consumption. Furthermore, this work considers social vulnerability, equity measures, and pavement surface friction in the pavement management decision-making process. In the first model, a multi-period, multi-objective, linear integer programming model is proposed that optimizes the mean network conditions and minimizes the fluctuation of network-level pavement conditions. The model also provides a maintenance and rehabilitation program for the pavement network under multiple tradeoff configurations to help decision-makers manage pavement maintenance programs within financial restrictions. In the second study, an extended goal programming framework combined with a pavement condition prediction model is proposed for the allocation of the pavement M&R budget across multiple management districts of a highway agency and over the planning horizon. The framework considers five objectives in the optimization formulation, including maximizing the proportion of pavement sections with the best conditions and minimizing the severity of crashes through pavement M&R. The model engages stakeholders in the budget allocation process so that they can communicate their preferences with their peers and the central authority in support of their districts' strategies. The third study presents a framework to integrate pavement surface friction into traditional pavement management to promote both ride comfort and safety. The model integrates skid resistance into pavement M&R decision-making using multi-choice goal programming and considers the interaction between pavement conditions, skid resistance, and M&R costs. Also, the proposed model looks at different ways to set goals to see what happens to pavement M&R planning when the goals change. The fourth and final study explores pavement maintenance and rehabilitation scheduling and a budget allocation model that prioritizes pavement M&R based on social vulnerability while achieving a certain level of equity both vertically and horizontally. The uncertainty of the available budget is modeled as a chance-constrained optimization problem. The proposed model provides a maintenance and rehabilitation program for the pavement network that achieves a certain level of equity both vertically and horizontally. The proposed model will assist decision-makers in developing social justice-based pavement management plans. Implementing such a plan would allow disadvantaged communities to improve their quality of life through efficien and safe mobility and access to opportunities. Case studies using Texas data are presented to demonstrate the application and effectiveness of the model in developing pavement maintenance and rehabilitation programs. Overall, the proposed models and frameworks offer decision-makers useful tools to optimize pavement maintenance and rehabilitation programs under various constraints and considerations.