This paper presents a new optimization model to the problem of pavement maintenance, rehabilitation andreconstruction (MRR) planning for a large-scale road network. In the past, this problem has usually beenformulated as linear programming or integer programming model The solutions obtained from thosemodels determine the timing, location and type of treatment to perform the MRR operation for a givenplanning horizon. However, a shortcoming of those models is that the selected sections for MRR areusually distributed spatially across the network, making it dillIcuh to plan and implement MRR activities ina coordinated manner. In light of the scale of economy, adjacent road sections with similar MRR needsshould be maintained within a single project. Unfortunately, the idea of automatically combining adjacentsections into one large project has not been given serious attention in existing optimization models forpavement MRR planning. In this paper, a new solution approach to pavement MRR planning is proposedthat utilizes the spatial structure of the road network. The road network is first partitioned into groups ofadjacent sections, or MRR projects, with similar MRR needs. Then, in order to maximize systemperformance a Knapsack problem is solved to optimally allocated resources to selected MRR projects.
展开▼
