城市道路重要度与交通状态存在复杂的相互作用，影响着人们的出行效率与出行满意度。该文从出租车GPS轨迹数据提取出GPS点位数、早高峰平均速度与夜间自由流速度之比两类交通状态指标，根据复杂网络理论获取拓扑结构中的节点度、介数中心性、接近中心性和特征向量中心性，来表征不同层面的道路重要度，利用耦合协调度模型探究了交通状态与4类道路重要度指标之间的耦合关系。以耦合关系为因变量，选用13个自变量，利用CatBoost算法结合SHAP图对耦合关系产生的机理进行回归分析。结果表明：交叉口数量、绿化面积、公交站点数量对交通状态与道路重要度的耦合协调程度起促进作用，道路分隔型式对交通状态与介数中心性、接近中心性、节点度的耦合关系具有较为显著的正向影响；对于正向不协调的道路，公共管理用地、居住用地强度对交通状态与接近中心性的耦合协调程度有负面影响，居住用地强度对交通状态与特征向量中心性的耦合协调程度有负面影响；对于负向不协调的道路，居住用地强度对交通状态与接近中心性的耦合协调程度有负面影响。

Urban road significance and traffic state have a complex interaction relationship, which affects people’s travelling efficiency and satisfaction. In this paper, two types of traffic state indicators are extracted from the GPS trajectory data of taxis, namely the number of GPS points and the ratio of average speed during the morning traffic peak to the free-flow speed at night. Based on the theory of complex networks, the node degree, betweenness centrality, closeness centrality and eigenvector centrality in the topological structure are obtained to characterize the road significance at different levels, and the coupling relationship between traffic state and the four significance indexes is explored by using the coupling coordination degree model. With the coupling relationship as the dependent variable, thirteen independent variables are selected to regress the mechanism of the coupling degree using CatBoost algorithm combined with SHAP. The results show that the number of intersections, the green area, and the number of bus stops promote the coupling coordination between traffic state and road significance, and that the road separation type has a significant positive effect on the degree of coupling between traffic state and betweenness centrality, closeness centrality as well as node degree. In addition, for the positively uncoordinated roads, the intensities of public management land and residential land have a negative effect on the coupling coordination between traffic state and closeness centrality, and the intensity of residential land has a negative effect on the coupling coordination between traffic state and eigenvector centrality; while for the negatively uncoordinated roads, the intensity of residential land has a negative effect on the coupling coordination between tra-ffic state and closeness centrality.