Generation of scenarios for ethanol demand time series: An analysis based on classical methods and ensemble techniques

- 325730
Complete Articles (CA)
Favorite this paper
How to cite this paper?
Abstract

Ethanol plays a key role in the decarbonization of the energy sector, gaining prominence in the transition to renewable energy sources. This study explores the use of time series methods to generate demand scenarios for this biofuel in the United States, using data from 1981 to 2024. To mitigate the impacts of the COVID-19 pandemic, the Seasonal Naive method was applied to remove outliers. A total of 100 demand scenarios were generated using STL decomposition, Box-Cox transformation, and Moving-Block Bootstrap. Three predictive approaches-Seasonal Naive, ETS, and ARIMA-were evaluated with and without bagging techniques to capture temporal dynamics of consumption. The evaluation, based on RMSE, MAPE, and DTW metrics, highlights the robustness of the scenarios generated with bagging and the potential of ARIMA combined with this technique to preserve structural patterns of the series in future forecasts. The results suggest the usefulness of these methods for planning under uncertainty in the biofuel sector.

Share your ideas or questions with the authors!

Did you know that the greatest stimulus in scientific and cultural development is curiosity? Leave your questions or suggestions to the author!

Sign in to interact

Have a question or suggestion? Share your feedback with the authors!

Institutions
  • 1 Petrobras (Brazil)
  • 2 PUC-Rio
  • 3 Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
Track
  • 7. EN&OG – OR in Energy, Oil and Gas
Keywords
Ethanol Demand
Time Series
Consumption Scenarios
Bootstrap