Study On Value-added Service For Power Customer Segmentation

In order to meet the needs of social and economic development, China Southern Power Grid Company Put forward the core value concept, Its slogan is "Light of million homes, Love from CSG". Through the establishment of customer segmentation model and the development of customer marketing service strategy, Southern Power Grid provides differentiated services and products with customers, which contributes to the increase in customer service levels further, under the guidance of this goal, Maoming power supply bureau also timely carry out the relevant work of customer classification, so that provide customers with personalized marketing and services via identifying different customer groups with the use of electrical characteristics.however, customer segmentation is finished by artificial means in the actual operation, which have many problems,for example, too many subjective factors and experience factors、low efficiency.at the same time, there is a lack of scientific and detailed index system in the model, which brings a lot of difficulties to customer segmentation.so it’s difficult to provide customers with personalized marketing and service.The customer service marketing strategy is formulated to enhance customer satisfaction, so that the customer characteristics are more fully understood and the customer demands are accurately grasped. In this paper, regarding the Maoming electric power customer as the research object, according to the characteristics of the existing customer base, the practical operation of the power customer segmentation method is discussed. Maoming Power Supply Bureau continue to carry out the construction of information acquisition system, which has a relatively complete database system that stores the vast amounts of customer data, which provides hardware condition for using data mining to achieve customer segmentation.as one of the core technologies of data mining, K-means clustering algorithm is becoming more and more important in the research of customer segmentation.but in the K-means algorithm, the clustering results are greatly influenced by the initial center value and the algorithm is easy to fall into the local optimal problem. In view of this, a K-means clustering algorithm based on the improved K-means algorithm is proposed, which make up for the shortcomings of the K-means algorithm, and a more in-depth study is made. On this basis, a set of system that oriented power supply enterprises of power customer segmentation and value-added service is developed, the information about customers in Maoming area is analyzed by using the new algorithm, so that forms a distinctive customer base, and according to the different demand characteristics, customized value-added services menu is personalized for customers.