<![CDATA[Smart farming has some limitations regarding the management of streaming data from IoT sensors. This is necessary to support real-time decision-making in areas with less infrastructure. This paper discusses the practical use of the N8N platform as a low-code/no-code workflow automation tool for monitoring IoT sensors in smart farming. A mixed-method approach was used, with a prototype design based on Research and Development. The system was built using IoT-A architecture, which includes the perception layer (soil moisture, temperature, humidity, pH, NPK, and ultrasonic sensors on ESP32), network layer (MQTT and HTTP), processing layer (N8N workflow for ingestion, validation, transformation, and decision logic), and application layer (dashboard and alerts). Testing was done in a controlled environment for 72 hours with scenarios such as normal operation, high load, network disruption sensor failure, and scalability up to 20 nodes. Results showed an average response time of 150–300 ms, throughput of up to 500 data points per minute end-to-end latency below 450 ms availability greater than 99% and processing accuracy between 98.7% and 99.2%. The system detected failures accurately and restored operations within an average of 45 seconds. These results proved that N8N can improve the efficiency and reliability of real-time monitoring as an adaptive solution for tropical agriculture in Indonesia. It also suggested long-term field trials together with AI integration for predictive forecasting to enhance scalability and practical adoption.]]>
