Abstract: With the cellular signaling data, spatial patterns of intercity tourists of 5A and 4A attractions in 26 cities in the Yangtze River Delta were analyzed under the current circumstance that there is insufficient consideration about tourism connection when forecasting intercity trip distribution. The data showed that spatial patterns of intercity tourists were similar among attractions whose tourists rank high and the overall 5A and 4A attractions in the same city. Meanwhile, the spatial pattern stayed stable on different days other than festival holidays. Based on this characteristic, the tourism preference parameter was constructed with the help of the preference scale and added to the basic gravity model to build an intercity trip distribution forecasting model. The calibration results, which are based on the current intercity trip distribution data of 26 cities in the Yangtze River Delta, show that the fitting degree of the model is improved, the root mean square error decreases significantly, and results of 75% of the OD pairs are improved. Therefore, the introduction of tourism preference can improve the accuracy of the intercity trip distribution forecasting model.

Key words: integrated transportation, intercity trip distribution, tourism preference, preference scale, gravity model

摘要： 针对目前城际客运出行分布预测过程中对城际旅游关联性考虑不足的问题，借助手机信 令数据，对长三角26个城市5A和4A级景点的城际游客客源地分布进行分析发现，在同一个城市 中，无论是从单个头部景点层面还是全部5A和4A级景点总体层面上来看，城际游客客源地分布 规律均较为相近，且在非节日的不同日期，这种分布也非常稳定。基于这种特性，借助亲景度指 标提出一个可以表征两个城市间旅游关联性的旅游偏好变量，并将其引入基本重力模型中构建 城际客运出行分布预测模型。基于长三角26个城市出行分布现状数据对模型系数进行标定后的 结果表明，模型拟合程度显著提高，标准误差大幅下降，75%的OD对预测结果得到改善。因此， 可以认为引入旅游偏好变量有利于提高城际客运出行分布预测模型的精度。

关键词: 综合运输, 城际客运出行分布, 旅游偏好, 亲景度, 重力模型