基于出行时间可达性的公共交通竞争力评估模型

为实现区域公共交通与小汽车间出行竞争力的定量化评价和影响关系解析，利用公共交通动静态数据、出租车运行数据及出行调查数据提取公共交通个体出行链和小汽车出行用时等信息，结合路径规划API(Application Programming Interface)计算公共交通及小汽车全过程出行耗时，基于全过程出行时间可达性视角构建公共交通竞争力评价模型；考虑公共交通竞争力的空间效应，分别选取土地利用和交通设施因素作为解释变量，提出基于空间杜宾模型的公共交通竞争力影响模型；以北京市为例，探究早、晚高峰等不同时段下区域公共交通竞争力与各影响因素之间的交互关系。结果表明：早、晚高峰期间公共交通竞争力的平均值均在1.50以内，平峰时段约为1.74，市中心区、地铁沿线和大部分居住社区周边区域的公共交通出行竞争力相对较高；各时段下公共交通竞争力呈现明显的空间依赖性，且存在“低-低聚集”“高-高聚集”的典型聚集区域；土地利用混合度、办公就业密度和地铁站点密度等因素具有显著的负向空间溢出效应，而道路网密度和绕行系数则表现出显著的正向效应。综上，所提出的评价模型能够定量化评估公共交通竞争力，所建影响模型能够在考虑空间依赖性的基础上解释竞争力与因素间的相互关系。

A Competitiveness Model of Public Transport Based on Travel Time Accessibility

In order to achieve quantitative evaluation and analysis of key influencing factors on the competitiveness between regional public transport and private car travel, the information on the whole travel time by public transport and private car were calculated based on dynamic and static public transport data, taxi travel data, travel survey data and route planning data. A competitiveness evaluation model was then built from the perspective of travel time accessibility. Since public transport competitiveness has spatial effects, a spatial Dubin model was developed by using land use and transport facility as explanatory variables to investigate the impact on public transport competitiveness. Taking Beijing City as an example, the relationship between regional public transport competitiveness and its various influencing factors was analyzed with different periods of morning, evening, and peak hours on weekdays. The resultsshow that the average value of public transport competitiveness is less than 1.50 during both the morning and evening peak periods, while around 1.74 during the flat peak. The competitiveness is relatively higher in the city center, along the metro lines, and in the areas around large residential communities. The results also indicate a clear spatial dependence and the existence of typical agglomeration areas of "low-low agglomeration" and "high-high agglomeration". The factors of land use, residential service density and metro station density have significant negative spatial spillover effects, while road network density and bypass coefficient show significant positive spillover effects. The proposed evaluation model could quantitatively assess the competitiveness of public transport, and the model can explain the interrelationship between competitiveness and factors taking into account spatial dependence.