Recently, most agencies responsible for the management of large highway and road networks have been putting increasing emphasis on pavement management issues. With highway networks that are aging and deteriorating rapidly, these agencies are forced to undertake a rigorous planning exercise if the scarce resources available for pavement maintenance and rehabilitation are to be allocated in an efficientfashion. To improve their planning mechanisms, many agencies have resorted to pavement management systems built around optimization models of one type or another. The general tendency in these systems has been to provide planners with a single well-defined optimization model, that is, with a model in which the objective function and the constraints have been stated definitively. When the complexity of the issues and of the trade-offs involved in pavement management is considered, one is forced to conclude that this single-model approach falls short of fulfilling the needs of decision makers: there are always important questions that remain unanswered. To remedy this situation, a different approach is proposed: to provide planners with a flexible decision-support tool that lets them specify the optimization problems that seem meaningful within the scope of the decision-making process at any given time. This paper appears in transportationresearch record no. 1272, Pavement management and rehabilitation 1990.
Abstract