AgriFlow: A Closed-Loop Cyber-Physical-Human Decision Support System for Precision Agriculture
Year: 2026
Type: article
Abstract: Smart-agriculture tools are usually built as separate applications for leaf-disease diagnosis, irrigation scheduling, and farm record keeping, so the outputs of one rarely correct or inform the others. This paper reviews ten peer-reviewed studies from Springer and Springer Nature venues, published between 2014 and 2025, covering plant-disease detection, machine-learning irrigation decision support, rainfall-forecast uncertainty, drought water allocation, and farm management information systems, and positions AgriFlow against them. AgriFlow is an integrated multi-tier agro-decision support system that combines a multi-crop MobileNetV2 leaf-pathology classifier benchmarked EfficientNet-B0, a closed-loop rainfall confirmation mechanism, variety-conditioned phenology modelling with supervisor ground-truth verification, a multi-factor Field Priority Score with what-if drought simulation, and a dual-photo maintenance ticket loop across four user roles. For each of the ten studies the paper states what AgriFlow adds, and also what the study offers that AgriFlow does not. In the abstracts and summaries examined, none of the ten studies combines these components in a single platform. Quantitative field evaluation of AgriFlow is identified as the main direction for future work.
Author: Vismaya Koorma
Institution: Amal Jyothi College of Engineering
Language: English
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7224161097
FWCI:
Cites:
Cited by:
Related to: