Deep Learning based Forecasting of Consumption of Petroleum Products in India

The forecasting of the petroleum products consumption is very important as the petroleum industry directly affects the transport sector of any country and so decision making regarding import/export of these products based on consumption prediction is extremely crucial. With the increasing ease of implementation of deep learning techniques, various time series forecasting methods have been developed. However, the performance of most methods is highly dependent on the availability of large amount of data. In this paper, we propose a forecasting methodology that use small datasets containing only past values of the consumption of petroleum products including Liquified Petroleum Gas (LPG), High Speed Diesel (HSD), Bitumen and Aviation Turbine Fuel (ATF) in India. The prediction has been done using various algorithms on the test set and the algorithm yielding lowest errors in prediction can be used to further predict the consumption for the future.