Smart Agriculture Management System Using Virtual IOT and AI Decision Support

Modern agriculture increasingly depends on data-driven decision support to improve productivity and resource efficiency; however, most existing smart farming solutions rely heavily on physical IoT sensor infrastructures that are costly, difficult to maintain, and limited in scalability for experimentation and deployment in resource-constrained environments. This dependency restricts widespread adoption and reduces accessibility to intelligent farming technologies. To address these limitations, this paper presents SAMS, a Virtual IoT and AI-based Smart Agriculture Management System designed to provide integrated decision support without mandatory reliance on physical sensing hardware. The proposed system introduces a virtual sensor simulation layer that generates realistic environmental data streams, enabling continuous monitoring, analytics, and system evaluation under diverse agricultural conditions. The platform incorporates multiple AI-driven modules, including crop recommendation, yield prediction, disease risk assessment, and weather-based analysis, all integrated within a unified architecture supported by centralized data processing and an interactive analytics dashboard. System implementation demonstrates stable data flow, consistent decision outputs, and effective integration of AI models within a scalable software framework. Experimental evaluation confirms that the virtual IoT approach can support intelligent agricultural decision-making while significantly reducing infrastructure constraints. The proposed system provides a practical, flexible, and cost-efficient alternative for smart agriculture applications, particularly suitable for educational, experimental, and early deployment scenarios.