分流区作为城市道路常见的交通场景，其复杂的交通特性和频繁的车道变换极易引发交通瓶颈和拥堵问题，对其通行能力的提升提出了挑战。为量化分析智能网联环境下干道分流区通行能力的影响因素，提出了一种基于P值检验和熵权法的综合筛选与量化方法。该方法通过P值检验从潜在影响因素中筛选出显著变量，并结合熵权法对各因素进行权重赋值，从而精准识别关键影响因素。以干道分流区为实例，从道路和交通两个维度，通过P值检验和熵权法筛选出网联渗透率和分流比为关键影响因素。同时，基于SUMO（simulation of urban mobility）仿真平台分析其对分流区通行能力的影响，结果表明：网联渗透率与通行能力呈显著正相关；分流比对通行能力的影响呈现非线性特征，即当分流比低于0.5时，其与通行能力呈正相关，而当分流比超过0.5时，则转为负相关；网联渗透率对通行能力的影响显著高于分流比。此外，通过探讨不同渗透率下网联自动驾驶车辆（connected and automated vehicle, CAV）专用道的实施效果，为智能网联技术的推广和分流区通行能力的优化提供了理论依据和实践指导。

The diverge area is recognized as a common traffic scenario in urban roads. It is characterized by complex traffic characteristics and frequent lane changes, which are known to easily lead to traffic bottlenecks and congestion. Challenges were posed to the improvement of its traffic capacity. To quantitatively analyze the influencing factors of traffic capacity in diverge areas under the intelligent connected context, a comprehensive screening and quantification method was proposed. This method was based on P-value testing and the entropy weight method. Significant variables were firstly screened from potential influencing factors through P-value testing. Weights were then assigned to each factor using the entropy weight method. Key influencing factors were accurately identified. The diverge area of an arterial road was taken as an example. The connectivity penetration rate and diverge ratio were identified as the key influencing factors from both road and traffic dimensions through P-value testing and the entropy weight method. The impacts of these factors on traffic capacity were analyzed based on the simulation of urban mobility(SUMO) simulation platform. The results indicate that the connectivity penetration rate is found to be significantly positively correlated with traffic capacity. The impact of the diverge ratio on traffic capacity is observed to exhibit nonlinear characteristics, with a positive correlation observed when the diverge ratio is below 0.5 and a negative correlation observed when it exceeds 0.5. The influence of connectivity penetration rate on traffic capacity is determined to be significantly higher than that of the diverge ratio. Furthermore, by exploring the implementation effects of dedicated lanes for connected and automated vehicles(CAV) under different penetration rates, it provides theoretical basis and practical guidance for the promotion of intelligent connected technologies and the optimization of traffic capacity in diversion areas.