This paper addresses a general class of pavement management problems raised in road maintenance and rehabilitation planning. Highway agencies usually face budget uncertainty from year to year. This uncertainty will eventually make the road network condition unstable if the maintenance budget falls short for some of the years. As a result, it is important for highway agencies to adopt maintenance and rehabilitation policies that are resistant to budget uncertainty. However, to the authors' knowledge, the questionof how to minimize the effect of this uncertainty on network-level pavement conditions has not been discussed in the current literature. In this paper, a deterministic linear programming formulation is proposed with its stochastic extension to handle the budget uncertainty effect on road network condition fluctuation. In order to solve the problem effectively, the use of the Nested Decomposition is recommended. A case study was carried outfor a network of roads in the Dallas district. The maintenance and rehabilitation plan was obtained. The stochastic programming solution was compared with the expected value solutions. Significant gain in terms of pavement condition fluctuation was observed by using the proposed approach.
Abstract