A MODEL FOR THE ESTIMATION OF USER DELAYS CAUSED BY PAVEMENT MAINTENANCE WORKS

Some tools, like HDM for instance, have been developed in order to optimize the economic benefits given by road maintenance; they include models taking into account the main economic terms which are the user costs (time, safety, comfort, vehicle consumption), the road agency costs (road construction, maintenance and exploitation), and sometimes external costs like environmental damages or non renewable resource consumption. However, recent studies have shown that a full optimization of road maintenance should take into account all the consequences of road works, including user costs at roadwork sites, which are key factors of the maintenance economic profitability, especially in European countries where traffic flows are often close to congestion levels. Reducing the frequency of road works, their duration and their impact on traffic flow (slowdowns and bottlenecks) will decrease the total delays experienced by users in their travels, and thus reduce the total economic cost of the maintenance policy. Such decisions may lead to an increase of the direct cost of works, but this can be accepted if the global balance is positive. The main problem in evaluating user delays caused by road works is that the response of flow condition is very sensitive to small changes in parameters like traffic demand and road capacity, even if these changes are very limited in time. In order to overcome this difficulty, a probabilistic model was developed, which can predict the instantaneous traffic, according to a normal distribution. The model parameters may be estimated through an analysis of traffic hourly counts performed during a series of days that can be distributed into different typical categories. This model will be used to calculate, on a given road network and for a given maintenance policy, the total volume of user delays associated to road maintenance operations, and so their cost and their impact on the total benefit assigned to the proposed solution.