APPLICATION OF DATA MINING WITH R PROGRAMMING TO IMPLEMENT ERP FOR DIGITIZATION OF VALUATION REPORTS ON COMMERCIAL VEHICLES AND PASSENGER CARS

This paper proposes an approach towards data mining with R programming implemented in ERP for complete digitization of valuation reports for commercial vehicles and passenger cars. This paper entails entry of basis data, customary for the operating system of ERP. This data is stored on cloud through server. The model proposed in this paper could be used with the proper selection of cloud and custom designing for ISO system and making the system ready. The model proposes a system with java programming which will be used for the front end design. In the back end, Oracle database can be used to store and manage data from both ERP database and the Cloud storage with structured orientation. The data contains details about types of car and the age of car and the month for optimum category. For betterment of the predictive model more number of parameters has been considered. In the Core End, The R programming is used for Multivariate statistical and predictive Analysis which plays the most important role for valuation of a used vehicle. The values of the parameters have been shown using Boxplot and histogram. According to the structure, a model has been proposed named as Java Enabled Data Mining Interface (JEMI) Model. For the purpose, all the data were collected from both the primary and secondary sources. First data had been collected from commercial database and internet. It was on valuation report for all types of cars. These data were segmented based on types of car and insurance companies. These valuation report formats had been differentiated for each company. Now data had been furnished in MS Excel according to their product name, variant's name, classification and company's activity. After that one new excel sheet was used for individual commercial and passenger car. Then all the parameters which were used for valuation of a car, they were collected from previous report and were furnished in one new sheet according to number of individual car. The research was further conducted to collect primary data from experts of the concerned area about their opinion on the major factors that decide the reselling price of a car. A number of parameters were considered for valuation of car, like TMFL, CMFL, and RCL. This total procedure is network based so it will require the cloud hosting. And different software related issues are also dealt herewith.