A fuzzy queuing model for real-time, adaptive prediction of incident delay for ATMS/ATIS

This paper presents a fuzzy queuing model that can be used to predict the possible delay that a vehicle will experience at an incident location based on real-time information on current queuing conditions, traffic arrivals, lane closings. Compared to most existing methods, the proposed model is unique in three aspects. First, it explicitly accounts for uncertainties involved in all influencing factors and thus allows easy incorporation of imprecise and vague information typically available in this type of prediction environment. Second, the model is adaptive in the way that it allows continuous update of estimates as new information is made available. Third, delays obtained from the model are fuzzy numbers that can be conveniently mapped to linguistic terms for use in systems such as changeable message signs (CMS). A case study is presented to demonstrate the application of the proposed model in facilitating the composition of location-dependent delay messages for CMS.