高速公路ETC客户细分方法研究

应用大数据技术，提出一种基于车辆通行特征的ETC客户细分方法.构建了ETC客户细分指标体系并利用ETC收费数据提取了最近消费间隔、年通行频次和年消费金额等细分指标；为克服大数据聚类的失效问题，结合CLARA算法完成了ETC客户全样本数据聚类分析；建立了ETC客户细分决策树并提取出细分规则，最终实现了ETC客户星级评定.分析结果表明，本文提出的ETC客户细分方法能够解析各类客户通行特征、当前价值和增值潜能，可为高速公路运营管理单位探索ETC客户精准营销和分级费率优惠提供创新思路，同时可为进一步提高ETC客户规模和支付比例、提升ETC管理决策水平提供理论支持.

Customer Segmentation for Expressway ETC System

Applying big data technology, a customer segmentation method of electronic toll collection (ETC) system is presented based on vehicle behavioral characteristics. A segmentation index system of ETC customer consisting of Recency, Frequency, and Monetary is constructed and extracted using ETC data. The whole sample clustering analysis of ETC customer is accomplished by CLARA algorithm while overcoming the invalidation problem on big data clustering. The decision tree of ETC customer segmentation is finally built and transformed into a set of segmentation rules. The empirical results indicate that the proposed method has a better ability to analyze the traveling characteristics, present values and appreciation potentials for different classifications of ETC customers. It provides an innovative idea to implement precision marketing and make hierarchical discount rates for ETC customers. Meanwhile, it provides theoretical support to further increase of ETC customer scale and paying ratio, as improve decision-making level in expressway operation and management.