Application of Artificial Neural Network Sensitivity Analysis to Identify Key Determinants of Harvesting Date and Yield of Soybean (Glycine max [L.] Merrill) Cultivar Augusta

Genotype and weather conditions play crucial roles in determining the volume stability of a soybean yield. The aim this study was to identify key meteorological factors affecting harvest date (model M_HARV) yield variety Augusta M_YIELD) using neural network sensitivity analysis. dates start flowering maturity, data, average daily temperatures precipitation were collected, Selyaninov hydrothermal coefficients calculated during fifteen-year (2005–2020 growing seasons). During experiment, highly variable occurred, strongly modifying course phenological phases achieved seed cultivar. harvesting mature seeds took place between 131 156 days after sowing, while harvested ranged from 0.6 t·ha−1 2.6 t·ha−1. analysis MLP made it possible which had greatest impact on tested dependent variables among all analyzed factors. It revealed that assigned ranks 1 2 forming M_HARV model total rainfall first decade June August. with highest mean air temperature second May Seljaninov coefficient values for sowing–flowering period.