多源数据驱动的高速公路服务区运营状态评价

针对各地高速公路服务区运营状态缺少标准评价指标的问题，利用高速公路智慧服务区所采集的大量多源异构数据，构建了一个高速公路服务区运营状态分级评价体系。首先，制定相应数据清洗规则来处理服务区所采集的原始数据。其次，基于k-means聚类算法搭建评价模型，对服务区运营状态进行分级评价研究。最后，以山西省盂县高速公路服务区为案例进行实证分析，对该服务区一周内每天同一时段以及一天内各个时段两方面的运营状态，分别进行横向及纵向比较分析，结果发现一天中下午及午高峰时段运营状态最差，平峰及凌晨时段运营状态较好。

Classification Evaluation of Expressway Service Area Operation State Driven by Multi-source Data

Aiming at the lack of standard evaluation indexes in expressway service areas, a comprehensive evaluation system for expressway service areas is built by using a large number of multi-source heterogeneous data collected by expressway intelligent service areas. Taking Yuxian Expressway Service Area in Shanxi Province as an example for empirical analysis. Firstly, the corresponding data cleaning rules are formulated to process the original data collected by the service area. Secondly, a hierarchical evaluation model based on k-means clustering algorithm is built to evaluate the service area. Taking the week data from March 15 to March 26 as an example, the horizontal and vertical comparative analysis is conducted on the operation state of the same period of each day within a week and each period of a day in the service area. It is found that the operation state on March 17 is the worst and that on March 20 is the best. In a day, the operation state is the worst in the afternoon and midday peak hours, and better in the flat peak and morning hours. Finally, the advantages, disadvantages and application scope of this evaluation method are analyzed.