Forecasting ethanol demand in India to meet future blending targets: A comparison of ARIMA and various regression models

oleh: Bishal Dey, Bidesh Roy, Subir Datta, Taha Selim Ustun

Format: Article
Diterbitkan: Elsevier 2023-03-01

Deskripsi

Accurate fuel demand forecast is crucial to meet the demand. India is aiming to roll out 20% ethanol blended gasoline by 2025 and planning to make this transition complete by 2030. But due to lack of ethanol availability this target cannot be achieved yet. So, accurate forecast of ethanol demand based on future gasoline demand is important. In this study, various linear and non-linear regression models and Autoregressive integrated moving average (ARIMA) models are developed and compared for forecasting gasoline demand in India. Historical gasoline consumption data from 1997 to 2021 is used to develop and evaluate these models. ARIMA (1,1,0) model is found to be the most accurate among all the developed models for this specific time series. Forecast made with ARIMA (1,1,0) model shows that ethanol demand for achieving 20% blending will be 9777 million Litres by 2025 and 11247 million Litres by 2030. Hence, considering the result of this study and the present domestic ethanol production it can be said that India will continue to face ethanol shortage to meet future blending targets.