Fuel Sales Price Forecasting using Time Series, Machine Learning, and Deep Learning Models
Mohammad Alwadi
Abstract
Advances in machine learning have led to an important shift in the petroleum service industry with the development of predictive analytics, using production records, staff data, and fuel sales profiles, to enhance machine learning-based systems that support decisions. These cutting-edge tools can forecast gasoline sales, identify significant trends, and pinpoint key factors that affect fuel sales. Fuel managers have a greater ability to make informed decisions, use resources efficiently, increase operational effectiveness, and increase sales of all fuel types based on the information obtained from these systems. Additionally, managers can efficiently manage risks while boosting efficiency by using predictive analytics and machine learning to quickly adapt to changes in the environment. This study proposes an innovative method for predictive fuel sales and pattern analysis, utilizing a real-life dataset from a service station in Jordan that features a real-time instant sales record, connected to an online sales system.