Servicios Personalizados
Revista
Articulo
Indicadores
Citado por SciELO
Accesos
Links relacionados
Similares en SciELO
Compartir
Biotecnia
versión On-line ISSN 1665-1456
Resumen
SERVIN-PALESTINA, M et al. Prediction of bean production and yields, with artificial neural network models and climate data. Biotecnia [online]. 2022, vol.24, n.2, pp.104-111. Epub 19-Mayo-2023. ISSN 1665-1456. https://doi.org/10.18633/biotecnia.v24i2.1664.
The state of Zacatecas ranks first in the production of rainfed beans in Mexico. Due to the importance of climatological variables, the objectives of this work were 1) to develop artificial neural networks (ANN) models of for the harvested surface (SC), yields (Rto) and production (P) prediction of rainfed beans in the state of Zacatecas, using data from the 1988-2019 period, and 2) perform the sensitivity analysis to determine the input variables that have the greatest influence on bean production and yield. The Climatol library of the R statistical package was used to fill in missing data. The results show that the ANN models capture the influence of climate on bean production, with an overall efficiency of 0.89 for Rto and 0.86 for SC. The production was estimated using the outputs, Rto and SC, from ANN models and an R2 =0.80 was obtained. According to the sensitivity analysis, evaporation of the cycle (Eva) was the most important variable for predicting yield, while precipitation in August (Pp_Ago) and minimum temperature (Tmin) had a greater influence on production.
Palabras llave : Artificial intelligence; Zacatecas; temperature; rainfall; rainfed crops; Phaseolus vulgaris; L.