The National Traffic Control Center (NTCC) in England is developing, prototyping and planning the introduction of an advanced Event Detection and Delay Estimation Engine (EDDEE) at the National Traffic Control Centre. At present abnormal congestion alerts are raised when the journey time on a section of the network increases significantly above expected for that timeof day and day type; this has many disadvantages. Using changes in flow to detect incidents can overcome these problems with the additional benefitthat there are four times more flow measurement sites than journey time sections, so using flow sites would result in better geographic granularity. EDDEE uses real-time raw traffic flow counts to identify the start and end of congestion and capacity reducing events. This involves comparing theflow rates at upstream and downstream loop-based monitoring sites. EDDEE calculates when an event starts, based on the stress (the ratio of the flow rate at the upstream site to the capacity of the road) and the difference between the downstream and upstream flow rates.
Abstract