|
Customer segmentation and value evaluation method based on data mining for electric vehicles |
DOI:10.7667/PSPC171615 |
Key Words:electric vehicle customer segmentation data mining K-MEANS algorithm customer value evaluation |
Author Name | Affiliation | ZHANG Lu | Electric Power Research Institute, State Grid Beijing Electric Power Company, Beijing 100075, China | LI Guochang | Electric Power Research Institute, State Grid Beijing Electric Power Company, Beijing 100075, China | CHEN Yanxia | Electric Power Research Institute, State Grid Beijing Electric Power Company, Beijing 100075, China | SUN Zhou | Electric Power Research Institute, State Grid Beijing Electric Power Company, Beijing 100075, China | WANG Weixian | Electric Power Research Institute, State Grid Beijing Electric Power Company, Beijing 100075, China | TIAN Heping | Electric Power Research Institute, State Grid Beijing Electric Power Company, Beijing 100075, China |
|
Hits: 4177 |
Download times: 1465 |
Abstract:Customer segmentation is of great significance for charge service operators to obtain the features of charging behaviors and individual differences between various Electric Vehicle (EV) users. Based on large numbers of charging service data fast accumulated by operations management system, exploratory data analysis is applied to all the historical data in the database. Firstly key variables are screened out to the segmentation model, and then the EV customer segmentation method by data mining technique and K-MEANS algorithm is presented. Secondly the customer value evaluation method is proposed, and charging behavioral features and customer values are analyzed based on Beijing EV customers. At last, conclusions and suggestions are given, which would provide data supports for the improvement of operation and maintenance management and decision-making of the precision marketing. This work is supported by Science and Technology Project of State Grid Corporation of China (No. 52020116000J). |
View Full Text View/Add Comment Download reader |
|
|
|