Meu SciELO
Serviços Personalizados
Artigo
Indicadores
- Citado por SciELO
Links relacionados
- Similares em SciELO
Compartilhar
Revista Cubana de Ciencias Informáticas
versão On-line ISSN 2227-1899
Resumo
PERDIGON LLANES, Rudibel e GONZALEZ BENITEZ, Neilys. Comparison and selection of artificial intelligence techniques for forecasting bovine milk productions. Rev cuba cienc informat [online]. 2021, vol.15, n.2, pp. 24-43. Epub 01-Jun-2021. ISSN 2227-1899.
Forecasting is an effective decision-making tool, especially in the dairy industry, because it helps to improve dairy herd management, save farm energy and optimize long-term capital investments. The application of artificial intelligence techniques to forecasting milk productions is a topic of interest for the scientific community. However, defining a technique or model to forecast these productions with an absolute performance at a global level is a challenging and complex activity, because none is accurate in all scenarios. In this research, artificial intelligence techniques used in the literature to forecast bovine milk productions were compared and the technique with the best adjustment to these forecasts was selected through the application of the Analytic Hierarchy Process. The synthetic analysis, the survey and experimental method were used as scientific methods. The results obtained allowed identifying artificial intelligence techniques based on Artificial Neural Networks as the best fit for forecasting bovine milk production, superior to Decision Trees and Support Vector Machines. It was determined that the most relevant selection criteria in the dairy production sector are the capacity of these techniques to handle data that present uncertainty and their ability to obtain precise results in an optimal way. The analysis carried out supports decision making in milk producing organizations.
Palavras-chave : multi-criteria analysis; analytic hierarchy process; forecasting; decision making.