Research On Establishment Of Electronic Refueling Data Warehouse And Its Data Mining

With the increasing competition in the market environment and the enhancement of the consumer's rights and interests,the managers require to consider more and more complex constraints in making decision.So managers urgently need a high efficiency of information decision support system(DSS)to achieve the rapid calculation of the business(such as marketing strategy),and to avoid the restrictions on individual cognitive limits,cut costs,information support,quality support,enterprise business process reengineering,employee empowerment and other needs.In the end,the decision makers can reduce the blindness in making decision,so as to achieve more efficient use of information resources,so as to improve the ability with the change of the market.Now,with the increasing of private cars in the oil sales industry,the demand for oil products sales is gradually growing,and the gas station site and service cannot meet these needs.At the same time,the original business analysis system cannot meet the needs of the present.This paper mainly focuses on the existing problems of the original business analysis system of YanChang Petroleum Sales Company.For example,the system performance is low,and it cannot achieve the multi-dimensional query statistics.In the other hand,customer credit cannot be measured,and there is lack of decision support for data mining.This paper mainly completed the following work.Firstly,based on the key technologies such as Data Warehouse(DW),On Line Analysis Processing(OLAP)and Data Mining(DM),we investigated and analyzed the original oil sales system of Yanchang Petroleum Sales Company and the needs of customers for the expansion of new business cooperation.Based on the single entity information of oil sales,customer dimension,regional dimension,time dimension and oil dimension,we finished the design and implementation of oil sales management data warehouse.Secondly,based on the data warehouse established above,this paper presents the use case diagram of the oil sales decision support system,the overall framework diagram of the system,and the creation and implementation of the multidimensional data set.This paper also presents some operation results,such as refueling dynamic data query,abnormal refueling data query,analysis of oil sales,and etc.Finally,the decision tree algorithm,ID3,is used to establish data mining model and predict the customer cooperation degree of oil sales,and the results are used to assist decision makers to develop more reasonable customer sales strategies.