EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
1100. Natural gas demand forecasting in Mexico with machine learning algorithms to support making decision process during energy transition
Invited abstract in session WB-28: Advancements of OR-analytics in statistics, machine learning and data science 9, stream Advancements of OR-analytics in statistics, machine learning and data science.
Wednesday, 10:30-12:00Room: 065 (building: 208)
Authors (first author is the speaker)
| 1. | Sergio Ibarra
|
| Operational Research, UNAM | |
| 2. | Aida Huerta-Barrientos
|
| Operations Research, National Autonomous University of Mexico |
Abstract
Mexican energy pool has a big dependence on natural gas, most of which is imported from the United States of America (USA). In this context a reliable, updated and adaptable tool to forecast natural gas demand in the medium term is required to support the Mexican government making decision process. There are now few studies related to natural gas demand forecast in Mexico, but they are either not updated in terms of historical data and variables used as inputs or they do not use the latest known methods such as Neural Networks (NN) losing the opportunity to leverage most recent forecasting techniques.
This study explores natural gas demand forecasting in Mexico using machine learning algorithms to support the energy transition. We compare time series models ARMA & ARIMA and Long Short Term Memory NN based on historical demand data from January 2005 to Nov 2023. Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and Akaike Criteria (AIK) reveal each method's strengths and weaknesses depending on data characteristics, like the presence of outliers. Our contribution lies in applying advanced machine learning to Mexican gas demand forecasting, expanding on existing research predominantly focused on countries like China, India, Turkey, and the USA. This work provides valuable insights for stakeholders navigating the energy transition in Mexico, helping optimize infrastructure planning, resource allocation, and energy security.
Keywords
- Energy Policy and Planning
- Algorithms
- Graphs and Networks
Status: accepted
Back to the list of papers