Estimation of network based incident delay in a transportation network using dynamic traffic assignment

The reduction of incident-induced delay is one of the main objectives of transportation management in many states. Many studies were performed in the past that developed models to estimate the delay resulting from an incident on the roadway. These methods are based on the impact of incidents on a single link that captures only the delay for travelers traversing a freeway segment upstream of the location of the incident. They do not take into account its spatial and temporal characteristics of the transportation system and the effect of information provided to travelers during an incident. This dissertation introduces a more comprehensive methodology to estimate the incident impacts on travel times at the network level. The method can be applied to more types of operational traffic conditions. The incident impact is estimated by calculating the incident delay as the difference of travel times on average between the travel times under normal and incident conditions. Using the VISTA computer transportation simulation DTA model, two transportation networks are simulated under a range of traffic demand levels, incident durations, lane blockages, and deployment of ATIS. Dynamic Traffic Assignment (DTA) can provide a more realistic representation of the traffic patterns on a network given the time-dependent demand. DTA produces the time-space trajectory of each individual vehicle from its origin to its destination. The estimates of travel times, vehicle flows, and corresponding statistics are obtained under varying incident schemes. The incident-induced delay is calculated using the proposed method to estimate the incident impacts on the roadway at the link, OD pair, and network levels. The results suggest that incidents have a different impact on different OD pairs. Incident delay depends mostly on the duration of the incident, the number of lanes blocked, and its location. When the network is operating near capacity conditions, the incident delay may be less than the corresponding delay if the network is operating on non-congested conditions. An effective traveler information system can alleviate the impacts of an incident to various travelers at varying degrees.