This paper proposes a bi-level mathematical programming framework for pavement maintenance to minimize fuel consumption in equilibrated networks. The proposed model extends research in the area by formulating the interaction between network equilibrium and various maintenance activities. The volume of traffic, age of the pavement, initial conditions, and interference due to maintenance are considered in developing this long-term deterioration model. The fuel consumption induced by roughness and traffic disruption is further investigated in the optimization process. A modified active set algorithm with nested sub-programming is developed to generate the detailed solutions. The performance of the proposed model was tested by comparing it with two other mainstream strategies: worst first and threshold control. The results show that eco-based method outperformed prevalent models by reducing extra fuel consumption by 20%. They thus show that eco-based optimal scheduling has the potential to aid in long-term maintenance decisions and reduce the energy cost.

本文提出了一种用于路面维护的双层数学编程框架，以最大程度地减少平衡网络中的油耗。拟议的模型通过制定网络平衡与各种维护活动之间的相互作用来扩展该领域的研究。在开发此长期恶化模型时，应考虑交通量，人行道的年龄，初始条件以及维护引起的干扰。在优化过程中，将进一步研究由粗糙度和交通中断引起的燃料消耗。开发了带有嵌套子程序的改进的活动集算法，以生成详细的解决方案。通过将其与其他两种主流策略进行比较来测试所提出模型的性能：最差优先控制和阈值控制。结果表明，基于生态的方法通过将额外的燃料消耗降低了20％，胜过了现有模型。因此，他们表明，基于生态的最佳调度有可能有助于长期维护决策并降低能源成本。