Abstract: The impact mechanism of cycle length difference of adjacent intersections on correlation degree index ( CI) was analyzed. Then took two signalized intersection as example, typical experimental scenarios were selected and the required traffic data were generated based on simulations. The relation models between normalized index of cycle length difference and CI in two coordinated directions were developed respectively. The threshold value of CI that could partition adjacent intersections into one subarea was determined. At last, by taking two adjacent intersections that located in Fuzhou city as example, the model was validated and the results show that the model was effective. The model can quantify the impact of cycle length difference on CI dynamically with the change of intersection traffic states and provide theoretically support to dynamic subarea partition.