Data-driven based evaluation method for urban signalized intersection

Urban highway traffic consists of intersections and links. It is the intersection where traffic jams always happen, so the intersection traffic evaluation is the key of the urban highway traffic evaluation. Because the intersection is a complicated system and affected by many uncertainties, it is very difficult to model accurately for intersection traffic. This paper proposes two performance measures, system average speed and level of service index, for intersection traffic evaluation, which can be got from the data of traffic detects. In the end, the simulation results prove that they work well.