UNAIR NEWS – The Instrumentation and Energy Research Community (IMERCY) of Universitas Airlangga (UNAIR) installed the second version of its Soil Quality Monitoring system (SQyM V2) in Plaosan Village, Wonoayu, Sidoarjo, on Sunday (June 29, 2025). The installation was part of UNAIR’s Call for Research (CFR) program. SQyM V2 is a smart tool developed to monitor soil conditions and provide actionable insights to farmers. IMERCY designed the system to classify agricultural soil quality using three Machine Learning algorithms: Random Forest (RF), Decision Tree (DT), and Artificial Neural Network (ANN).
Additionally, SQyM V2 contributes to achieving several United Nations Sustainable Development Goals (SDGs), including Goal 2 (Zero Hunger), Goal 9 (Industry, Innovation, and Infrastructure), Goal 15 (Life on Land), and Goal 17 (Partnerships for the Goals). The system supports efforts to build a more sustainable agricultural sector.
The installation was well received by local farmers and village officials, who expressed hopes that SQyM V2 would improve productivity by streamlining data analysis and helping farmers better classify soil quality. The tool measures seven key parameters: nitrogen, phosphorus, potassium, moisture level, pH, total dissolved solids (TDS), and salinity.
Laboratory testing
To support the classification process, the IMERCY team also conducted lab testing at the Soil Resources Laboratory at UPN Veteran East Java. They analyzed farmland soil samples to categorize them into three levels—low, medium, and high—across the seven parameters.
SQyM V2 is powered by solar panels and utilizes the ESP8266 microcontroller. “The tool uses NPK sensors to collect data, which is then pre-processed through a cloud server to ensure accuracy and reliability. Cleaned data is then run through three machine learning models—DT, RF, and ANN,” explained Ibnu Andhika, an IMERCY representative who presented the research results.
The system identifies the most effective model based on Confusion Matrix and Accuracy Score evaluations, then deploys it through the ESP32 module. Final results are shown as descriptive text, accessible via both the device’s screen and smartphones. “This system is designed to help farmers more easily manage their land,” Andhika said.
Ultimately, the Artificial Neural Network algorithm proved to be the most accurate among the three in generating recommendations. Its superior performance enables SQyM V2 to deliver more precise guidance to farmers. With this tool, IMERCY aims to simplify land management practices and help farmers maximize crop productivity.
Author: Fikarul Mujtahida, Andri Hariyanto
Editor: Yulia Rohmawati
