考虑交通流动态分布的路面养护最优决策

为准确评估路面全生命周期用户成本以制定最优的路面养护策略，提出基于网络交通流均衡条件的路面养护多阶段动态规划最优决策模型，通过交通流的仿真，模拟用户成本在路网养护前、后以及生命周期各阶段的动态变化，并设计启发式迭代算法。该模型最小化的目标函数包含了生命周期的养护成本、用户成本、路面资产残值，约束条件主要由交通量守恒约束条件、养护预算约束、路面性能转移约束等。为证明模型的有效性，以一个具有10个路段的路网为例，采用提出的模型制定该路网连续20年的最优养护决策方案。研究结果表明：在每年养护预算确定的情况下，全生命周期路面养护最优方案是一个决策方案组合，该最优方案可以提高养护资金使用效率并有效减少用户成本，可为路面养护管理提供决策支持。通过对不同养护约束情景下的全生命周期用户成本的模拟表明：随着养护预算的增加，养护路段增多，导致用户成本增加，使养护成本的边际收益发生递减，养护预算在理论上具有一个最优值，并且养护成本收益与养护预算之间呈倒U形函数关系。

Optimal Decision of Pavement Maintenance Considering the Dynamic Distribution of Traffic Flow

To accurately evaluate the life cycle of user costs for pavements and to formulate an optimal pavement maintenance strategy, a multi-stage dynamic programming optimal decision model of pavement maintenance based on network traffic flow equilibrium conditions was proposed in this study. This model can mimic dynamic changes of user cost before and after maintenance of the network and the various stages of the life cycle through a simulation of traffic flow. A heuristic iterative algorithm was then developed to solve the model. The objective function of the model minimizes maintenance and user costs as well as the salvage value of pavement assets. The model is subject to the constraints of traffic conservation, maintenance budget, and road performance transfer. To validate the model, an optimal maintenance decision-making plan was formulated for a sample road network having 10 road segments for 20 consecutive years.Results show that the optimal life cycle pavement maintenance strategy combines decision-making schemes, which can improve the efficiency of maintenance funds and effectively reduce user costs.It can also provide policy supports for road maintenance management. The simulation of a full life cycle of user costs under different conservation constraints reveals that a greater maintenance budget can increase the number of maintenance roads, resulting in an increase in user costs.This in turn leads to a diminishing marginal benefit of maintenance costs. In theory, the maintenance budget has an optimal value, and the relationship between the maintenance cost benefit and maintenance budget is an inverted Ushape.