ABSTRACT
Many Indian manufacturing organizations use natural gas as energy source in their operations. This is attributed to the fact that natural gas emissions contribute very less pollution as compared to traditional fuels such as coal and oil. Additionally, the prices of natural gas are lesser as compared to electricity and oil, but prices of this commodity are volatile and governed by international market. Observing these fluctuations in the price of the natural gas, it is essential to predict the quantity of natural gas to be procured based on the historical data. If any mathematical model predicts future data by considering historical data as input, it is termed as time series forecasting. Analyzing this type of data has become a recent area of focus in artificial intelligence, as accurate forecasting is becoming increasingly vital across all kinds of industries in order to make more informed and accurate decisions. Essentially, applying AI to time series analysis allows us to better uncover the meaning of hidden patterns in the data. There are many statistical methods and machine learning and deep learning algorithms to perform time series predictions, but classical methods such as ETS (error, trend, seasonal) and ARIMA (auto-regressive integrated moving average) outperform machine learning and deep learning methods for one-step and multi-step forecasting on univariate datasets. Thus, this study used traditional time series forecasting to achieve effective energy resource planning that ensures the availability of the desired amount of natural gas for various manufacturing operations at appropriate point of time.
