By leveraging real-time data from sensors, historical records, and current environmental conditions both inside and outside the greenhouse, the tool delivers precise and dynamic recommendations for the optimal adjustments of windows, curtains, and fans. These recommendations ensure that conditions stay within the user-defined parameters while adapting to environmental fluctuations in real time.

The tool leverages Random Forest Regression, a robust machine learning model, to analyze and predict the most effective settings for greenhouse systems. The process involves:

1. Data Collection:

   Greenhouse interior: Captures temperature, humidity, and radiation levels.

   Rhizosphere (substrate): Monitors substrate humidity and temperature.

2. Data Processing and Prediction:

   The collected data is fed into the Random Forest model, which processes historical and real-time inputs to calculate the optimal settings for windows, curtains, and fans.

   The algorithm accounts for non-linear interactions between variables, making it highly effective in dynamic and complex environments.

3. Recommendation and Adjustment:

   The tool suggests adjustments based on predictions to ensure that the microclimate remains within the user-defined thresholds.

   These adjustments are continuously refined as new data is collected.

• Innovation in Optimization: Utilizes Random Forest Regression to accurately calculate the optimal settings for crop growth

• Increased Efficiency: Maintains conditions that maximize crop yield and quality.

• Cost Savings: Reduces energy and water consumption through precise recommendations.

• Customization: Adapts to the specific needs of different crops as defined by the user.

• Reduced Errors: Minimizes reliance on manual adjustments through automated decision-making.