Study On Urban Arterial Signal Coordination Method

Urban traffic control is one of the effective means to alleviate the city traffic congestionand reduce traffic air pollutant emissions. Because the arterial intersections are stronglycorrelated, it is necessary to study the arterial coordination control method in order to improvethe efficiency of urban traffic control. This study has conducted a thorough study into existingarterial signal coordination methods, and proposes a new graphic method of arterial signalcoordination for undersaturated traffic conditions and a new mixed integer linear planningmethod for adjacent signals under oversaturated conditions. Both methods have been testedby field work, and have achieved expected results.This new graphic method for arterial signal coordination control is applicable forundersaturated conditions. The innovation lays in adopting a series of continuous andindependent procedures to have the offsets, phase patterns optimized in one drawing. Theseprocedures include: first, maximum one-way bandwidth optimization, which is to set theinitial offset of adjacent intersections equal to the forward travel time between them; second,split control box designation, which is built to facilitate the choice of phase patterns; third,phase pattern optimization, which is to obtain the phase pattern that maximize the bandwidthfor the other way; and fourth, the offset optimization, which is to adjust the size of two-waygreen wave bandwidth and to approximate the expected coordination control effect.This new mixed integer linear planning method for adjacent signals is designed for thebalanced state under oversaturated traffic conditions. The innovation here lays in designingtwo continuous and independent steps to complete the optimization of phase patterns, splits,public cycle length and offsets. These steps are: first, construct a mixed integer linearprogramming model and its branch and bound solution to optimize split, public cycle lengthand offset, using signal groups as the variables; and second, choose the phase patterns that cancontribute to a mimimum queue length among the road.