Prediction of Harvest Time of Apple Trees: An RNN-Based Approach

In the field of agricultural research, Machine Learning (ML) has been used to increase productivity and minimize its environmental impact, proving be an essential technique support decision making. Accurate harvest time prediction is a challenge for fruit production in sustainable manner, which could eventually reduce food waste. Linear models have estimate period duration; however, they present variability when chronological apple tree stages. This study proposes PredHarv model, machine learning model that uses Recurrent Neural Networks (RNN) predict start date harvest, given weather conditions related temperature expected period. Predictions are made from phenological phase beginning flowering, using multivariate approach, based on series phenology meteorological data. The computational contributes anticipating information about date, enabling grower better plan activities, avoiding costs, consequently improving productivity. We developed prototype performed experiments with real datasets institutions. evaluated metrics, results obtained evaluation scenarios demonstrate efficient, good generalizability, capable accuracy results.