Por favor, use este identificador para citar o enlazar este ítem:
http://www.alice.cnptia.embrapa.br/alice/handle/doc/1188870Registro completo de metadatos
| Campo DC | Valor | Lengua/Idioma |
|---|---|---|
| dc.contributor.author | GEBLER, L. | |
| dc.contributor.author | DIAS, J. M. | |
| dc.date.accessioned | 2026-08-03T21:34:10Z | - |
| dc.date.available | 2026-08-03T21:34:10Z | - |
| dc.date.created | 2026-08-03 | |
| dc.date.issued | 2026 | |
| dc.identifier.citation | In: INTERNATIONAL CONFERENCE ON PRECISION AGRICULTURE, 17., CONGRESSO BRASILEIRO DE AGRICULTURA DE PRECISÃO DIGITAL, 11., 2026, Porto Alegre, RS. Anais...Porto Alegre: AsBraAP; AP&D, de 13 a 16 de julho de 2026. | |
| dc.identifier.uri | http://www.alice.cnptia.embrapa.br/alice/handle/doc/1188870 | - |
| dc.description | Implementation of precision fruit farming faces challenges in accurately defining these zones, mainly because, as the orchard reaches the productive phase, the relevance of soil fertility decreases compared to other phytotechnical and physiological parameters. Correct generation of management zones is crucial for the success of the operation, but the accurate interpretation of the collected data requires highly qualified professionals with years of experience, a gap that limits the adoption of precision fruit farming. The overall objective of this work is to develop a computerbased decision support system (DSS) using a binary decision tree to assist in the generation of management zones in precision fruit farming. This tool aims to increase the accuracy of the process and reduce the dependence on specialized technical expertise, standardizing procedures and minimizing the need for years of experience in management. The system was designed to guide the generation of these zones, considering factors such as plant physiology, climate, and soil fertility. The methodology was based on a perceived demand in the productive sector and used a new algorithm structured on the principle of decision trees, using binary questions and answers (yes or no) in the form of flowcharts. The algorithm was divided into specialties for the generation of management zones. The computational system was planned to use structured programming in Python 3.11 (as the back-end), and the front-end in Django for the creation of the RAG, but ended up being developed using the n8n software, whose code is in JSON, allowing publication on the Web. The system was modeled as an Artificial Intelligence Agent (AI Agent) with a deliberative (goal-oriented) architecture, orchestrated in n8n. This agent is responsible for the coordination and execution of an LLM model (ChatGPT), which compiles the information from the flowcharts and offers the user detailed answers in a guiding text format, with a meticulous step-by-step process, ideal for lay users. For preliminary evaluation purposes, the completed system was subjected to a semi-empirical benchmark, where the URL was informally distributed so that users could interact and verify the consistency of the responses. The feedback from the evaluations, conducted personally, demonstrated that the decision support system is functional, modular, and adaptable. The research confirmed that the combination of binary decision trees with deliberative agents constitutes an effective solution for guiding the process of generating management zones. It is expected that the tool, which acts as a digital manager, can be made available to the general public, possibly through Embrapa, allowing for its future expansion and improvement. In short, the work offers a technological alternative capable of reducing dependence on specialists and democratizing access to data-driven management practices, contributing to the digitalization and sustainability of agribusiness. | |
| dc.language.iso | eng | |
| dc.rights | openAccess | |
| dc.subject | Large Language Models | |
| dc.subject | Digital Agriculture | |
| dc.title | A model to support decision-making in the generation of management zones for fruit growing. | |
| dc.type | Artigo em anais e proceedings | |
| dc.subject.nalthesaurus | Precision agriculture | |
| riaa.ainfo.id | 1188870 | |
| riaa.ainfo.lastupdate | 2026-08-03 | |
| dc.contributor.institution | LUCIANO GEBLER, CNPUV; JOVANIA MENEZES DIAS, EMBRAPA UVA E VINHO. | |
| Aparece en las colecciones: | Artigo em anais de congresso (CNPUV)  | |
Ficheros en este ítem:
| Fichero | Tamaño | Formato | |
|---|---|---|---|
| Gebler-Dias-ICPA-ConBAP-2026.pdf | 673,48 kB | Adobe PDF | Visualizar/Abrir |
