Abstract: This paper proposes a concept of “data-driven, lean-oriented and closed-loop” management for distribution network and explain how to implement this kind of management, as shown in fig 1. Firstly, a big data platform is constructed to integrate and combine the multi-source data. Secondly, big data analysis technologies such as data mining, machine learning and data visualization are applied to solve problems in distribution network production. For example, we can achieve more accurate load prediction with machine learning. And we can also be aware of the risk points in distribution network through history data analysis. Finally, this paper explains how to promote lean management of distribution network in the fields of asset, operation, maintenance and investment based on the big data platform and big data analysis methods. In addition, the feedback procedure sets up a bridge between application and data collecting, which further improve the data quality. Those management measure have been piloted in several cities in Jiangsu. The result proves that they can improve power supply reliability and reduce operating costs significantly. A practical case shows how they work.