Petroleum Software License Usage Forecast Based On ARIMA Algorithm

Abstract

In petroleum industry, the expensive software license procurement costs have brought a heavy burden to enterprises. When purchasing the software license, the license management department usually forecasts the reservation, warning, purchase quantity and threshold setting of licenses according to experience and factor analysis method. Due to lack of data support and inaccurate forecast, the enterprises always buy insufficient or surplus quantities of license. In order to solve the above problems, and realize intelligent forecast for license needs, we proposed a method for forecasting the license using status based on ARIMA algorithm, and designed a license management system. Through numerical experiments, it is found that compared with the traditional forecast methods and models, due to the enhanced ability to extract random information, it is more suitable for non-stationary license occupancy sequence forecast in most cases, and the forecast is much more accurate as well.