AgriTwin : A Sensor Free Digital Twin System for Smart Farming Using APIs and Geo-Maps

Abstract

Farmers currently faces multiple challenges like fluctuating climatic conditions, reduced soil quality, excessive use of resources, and inconsistent market demands. In various regions, farmers still depends largely on direct observation and practical experience while making decisions about crop planning, irrigation, and fertilizer usage, which may lead to excessive resource utilization and reduced productivity. To overcome these disadvantages, this paper develops AgriTwin, a digital twin-based agricultural system that creates a digital representation of farmland for predictive analysis and decision support. The proposed system includes agricultural datasets, real time weather data collected through external APIs, and inputs provided by farmers to represent to soil condition, crop growth patterns, and environmental factors. By applying machine learning methods, the platform analyzes these data sources to generate insights such as crop suitability recommendations, yield estimation, irrigation scheduling, and market trend analysis. The digital twin environment enables farmer to stimulate different farming strategies before applying them in real fields, helping reduce risk and improve planning. In addition, that platform includes an interactive visualization dashboard that presents predictions and analytical insights in a simple and user-friendly format. By combining data- driven analysis with virtual farm modeling, AgriTwin aims to support informed decision-making, optimize resource utilization, and promote conventional agricultural practices.