EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
878. Forecasting mobile sources risks using artificial neural networks
Invited abstract in session WC-28: Advancements of OR-analytics in statistics, machine learning and data science 10, stream Advancements of OR-analytics in statistics, machine learning and data science.
Wednesday, 12:30-14:00Room: 065 (building: 208)
Authors (first author is the speaker)
| 1. | Thiago Antonini Alves
|
| Department of Mechanical Engineering, Federal University of Technology - Parana | |
| 2. | Hugo Valadares Siqueira
|
| Instituto Brasileiro de Informação em Ciência e Tecnologia | |
| 3. | Tiago Emmanuel Nunes Braga
|
| Instituto Brasileiro de Informação em Ciência e Tecnologia | |
| 4. | Deivdy Silva
|
| Instituto Brasileiro de Informação em Ciência e Tecnologia | |
| 5. | Yago Honda
|
| Instituto Brasileiro de Informação em Ciência e Tecnologia | |
| 6. | Yara Tadano
|
| Federal University of Technology - Parana |
Abstract
To implement public policies to reduce continuous emissions of green house gases (GHG) and other air pollutants are paramount. One of the main sources of GHG emissions are the use of fossil fuel. In this way, research related to emissions of pollutants from vehicles is paramount. Thus, seeking public policies encouraging the use and the development of more sustainable are essential to preserve populations’ health. The World Health Organization estimates that 6.5 million premature deaths are related to air pollution. To better understand the health risks caused by emissions from mobile sources, is important to select the most important input variables. Therefore, this research aims to analyze and select the input variables that most affect the air pollution health risks. To do so, we applied three Artificial Neural Networks (Multilayer Perceptron, Extreme Learning Machines, and Echo State Neural Networks) to estimate the impacts of air pollution on outcomes for respiratory diseases (hospital admissions and mortality) using road vehicles fleet, distributed and sold fuels amount, and vehicle average mileage. The results showed that the best performance was achieved considering all the input variables. The ELM reached the best overall performance for hospital admissions, and ESN for mortality, both using deseasonalization methods.
Keywords
- Artificial Intelligence
- Forecasting
- Machine Learning
Status: accepted
Back to the list of papers