The application of a pavement management system (PMS) to optimise the allocation of scarce budgeting resources for a network of highways is becoming more common both in developed and in developing countries. The use of expert opinion and expert systems may help to improve a PMS as well as to ease the computational burden in some cases. In such systems, degradation models are necessary to predict the impact of scheduled maintenance so that both the long- and short-term results are optimised. Expert opinion is also often used to determine the feasible maintenance and rehabilitation actions for pavement in different condition states. From the set of feasible actions, the network level optimiser will select a multiyear optimal strategy. Two different approaches are illustrated: one for the Ohio Department of Transportation, and the second for Saudi Arabia. An algorithm is presented for updating expert opinion-based degradation models for pavements. Bayesian updating procedures are given that automatically update the degradation models with new network survey data. These procedures continually self-adjust the PMS to fit the specific conditions found in the network. This process results in improved prediction models and a better use of resources.
Samenvatting