Forecasting Oil Production and Consumption Using the Time-delayed Fractional Discrete Grey Model with Multiple Fractional Order

Abstract

Accurate production and consumption forecasts play a crucial role in economic development, environmental protection, and market investment. By introducing fractional accumulation and time delay effects, the timedelayed fractional discrete grey model with multiple fractional order can more accurately capture the dynamic changes in data. The versatility and flexibility of this model allow it to adapt to various data characteristics and complexities, thereby providing higher forecasting accuracy compared to traditional grey models. Therefore, this study employs the existing time-delayed fractional discrete grey model with multiple fractional order and combines it with the Particle Swarm Optimization algorithm to optimize the fractional order. Experimental results show that the model demonstrates significant advantages in both fitting and forecasting capabilities. Through an in-depth analysis of oil production and consumption data in the Asia-Pacific region, the Commonwealth of Independent States (CIS), and the Middle East, this study proves the prediction accuracy and reliability of the TDF-DGM model in these regions. The average percentage error of oil production forecast in Asia-Pacific region is 3.0892%, the percentage error of oil consumption forecast in CIS is 2.3307%, and the percentage error of oil consumption forecast in the Middle East is 4.4986%. The prediction results are obvious due to other prediction models. The main objective of this study is to improve the accuracy and reliability of oil production and consumption forecasting using advanced modeling techniques. Based on model reliability, TDF-DGM model is used for strategic planning and investment decisions to improve accuracy and reduce risk.