Forecasting the Fuel Consumption and Price for a Future Pandemic Outbreak: A Case Study in the USA under COVID-19

被引:1
作者
Sakib, Ahmed Nazmus [1 ]
Razzaghi, Talayeh [2 ]
Bhuiyan, Md Monjur Hossain [1 ]
机构
[1] Univ Oklahoma, Sch Aerosp & Mech Engn, Norman, OK 73019 USA
[2] Univ Oklahoma, Sch Ind & Syst Engn, Norman, OK 73019 USA
关键词
forecast; power; monkeypox; COVID-19; fuel; price; energy; pandemic; stochastic; ARIMAX; ELECTRICITY DEMAND; IMPACT;
D O I
10.3390/su151712692
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
The COVID-19 epidemic and the measures adopted to contain it have had a significant impact on energy patterns throughout the world. The pandemic and movement restrictions led to unpredictable fluctuations in power systems demand and the fuel price for a delayed period. Monkeypox, another viral disease, appeared during the post-COVID period. It is assumed that the outbreak of monkeypox is unlikely due to the implication of preventive measures experienced from COVID-19. At the same time, the probability of an epidemic cannot be blindly overlooked. This paper aims to examine and analyze historical data to look at how much petroleum fuel was used for generating power and how the price of petroleum fuel changed over seven years, from January 2016 to August 2022. This period covers the time before the COVID-19 pandemic, during the pandemic, and after the pandemic. Several time-series forecasting models, including all four benchmark methods (Mean, Naive, Drift, and Snaive), Seasonal and Trend decomposition using Loess (STL), Exponential Smoothing (ETS), and Autoregressive Integrated Moving Average (ARIMA) methods have been applied for both fuel consumption and price prediction. The best forecasting method for fuel price and consumption has been identified among these methods. The best forecasting method for fuel consumption observed is ETS based on the RMSE value, which is 799.59, and the ARIMA method for fuel price, with RMSE 4.67. The paper also utilizes the ARIMAX model by incorporating multiple exogenous variables, such as monthly mean temperature, mean fuel price, and mileage of vehicles traveling during a certain period of pandemic lock-down. It will assist in capturing the non-smooth and stochastic pattern of fuel consumption and price due to the pandemic by separating the seasonal influence and, thus, provide a prediction of the consumption pattern in the event of any future pandemic. The novelty of the article will assist in exploring the potential energy demand in terms of cost and consumption of fuel during any pandemic period, considering the associated abnormalities.
引用
收藏
页数:26
相关论文
共 25 条
[1]   Analysis of the electricity demand trends amidst the COVID-19 coronavirus pandemic [J].
Abu-Rayash, Azzam ;
Dincer, Ibrahim .
ENERGY RESEARCH & SOCIAL SCIENCE, 2020, 68
[2]   Time Series Prediction of Electricity Demand Using Adaptive Neuro-Fuzzy Inference Systems [J].
Acakpovi, Amevi ;
Ternor, Alfred Tettey ;
Asabere, Nana Yaw ;
Adjei, Patrick ;
Iddrisu, Abdul-Shakud .
MATHEMATICAL PROBLEMS IN ENGINEERING, 2020, 2020
[3]   Impact of the COVID-19 Pandemic on the US Electricity Demand and Supply: An Early View From Data [J].
Agdas, Duzgun ;
Barooah, Prabir .
IEEE ACCESS, 2020, 8 :151523-151534
[4]   Day-ahead industrial load forecasting for electric RTG cranes [J].
Alasali, Feras ;
Haben, Stephen ;
Becerra, Victor ;
Holderbaum, William .
JOURNAL OF MODERN POWER SYSTEMS AND CLEAN ENERGY, 2018, 6 (02) :223-234
[5]  
[Anonymous], 2011, ELECTRICITY
[6]  
Baker S.R., 2020, COVID INDUCED EC UNC, DOI [DOI 10.3386/W26983, 10.3386/w26983]
[7]  
Brockwell PJ, 2016, SPRINGER TEXTS STAT, P1, DOI 10.1007/978-3-319-29854-2
[8]  
Camp K, 2020, MON LABOR REV
[9]  
Fairlie R.W., 2020, Working Paper No. 20-022
[10]  
Gungor O., The Impact of COVID-19 Outbreak on Turkish Gasoline Consumption