Travellers are intolerant of delays, especially if they are unexpected. The impact of such delays has significant economic impact on the individualand the society, especially if the trip is important. Traffic managers also need a performance measure to measure the stability of the quality of service they provide. A practical need therefore arises for the developmentof modelling techniques that are well able to quantify such impact and include them in both the design and economic appraisal of transport systems.These led to the sharp interest in incorporating reliability measures in traveller's information systems and in route choice models. Reliability measures show the extent of variability and robustness of the network or routes in coping with variable demand under both normal (traffic conditions) and abnormal (natural disaster) situations. To obtain maximum information about reliability measures, one needs to specify the full distribution of travel times across different times of the day. A technique for estimatingfull probability distributions of route travel times, in the light of day-to-day variations in the travel demand has been proposed. The solution method proposed operates in two stages. In the first stage, moments of the total travel time for each route of interest are computed by an analytic method, based on the multivariate moments of the link flow vector. In the final stage, a flexible family of density functions is fitted into these moments. The resulting distribution is used to characterize unreliability. The method also provides means of identifying vulnerable links, and examining the impact on route reliability to changes in link capacities. For the covering abstract see ITRD E145999
Samenvatting