MULTI-CRITERIA OPTIMIZATION OF SINGLE INTERSECTION UNDER OVER-SATURATED CONDITIONS

Abstract

Intersections are junctions where at least two roads or streets meet or cross. Consequently, they represent the most critical spots in the city traffic network, in terms of traffic safety. When determining the Level of Service at signalized intersections, the Highway Capacity Manual (HCM) does not take into account the traffic safety factors. In order to overcome this shortcoming, we propose in this paper the multi-criteria approach to the single intersection traffic control problem. The first objective function to be minimized is the average control delay experienced by all vehicles that arrive at the intersection. The second objective function to be minimized is the traffic safety risk index on intersection. These two criteria are in conflict, and the task is to find a good compromise between them. The analysed problem is solved by the combination of dynamic programming, and compromise programming. The proposed approach was tested in the case of the intersection that is controlled with two phases, under over-saturated traffic flows.