Scholarlink

<p>This study develops an AI-driven Decision Support System (DSS) to optimize Eucalyptus replanting schedules using machine learning models, including Random Forest, XGBoost, and Long Short-Term Memory (LSTM). Among these models, Random Forest demonstrated the best performance, achieving a Mean Absolute Error (MAE) of 54.91 days, a Root Mean Square Error (RMSE) of 86.54 days, and an R² value of 0.8582. Additional validation confirmed the model’s robustness, showing consistent performance across the validation and test datasets and indicating strong generalization capability. The results suggest that Eucalyptus replanting typically occurs approximately 1000 to 1500 days (3 to 4 years) after tree cutting. The proposed DSS enhances operational efficiency by optimizing replanting schedules using both real-time and historical data, ensuring the timely availability of raw materials. Furthermore, it has the potential to reduce operational costs by minimizing manual monitoring and optimizing resource utilization, such as water and fertilizers, thereby supporting more sustainable plantation management. Key factors influencing Eucalyptus growth include rainfall, temperature, humidity, soil pH, and nutrient levels. Future work may focus on incorporating additional environmental variables, such as long-term climate projections, and continuously updating the model with new data to further improve predictive performance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial intelligence decision support system for optimizing Eucalyptus replanting in sustainable forest management

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月15日

查看全文
  • Nurhayati Sembiring,
  • Tuti Sarma Sinaga,
  • Delima Sitanggang,
  • Juliza Hidayati

摘要

This study develops an AI-driven Decision Support System (DSS) to optimize Eucalyptus replanting schedules using machine learning models, including Random Forest, XGBoost, and Long Short-Term Memory (LSTM). Among these models, Random Forest demonstrated the best performance, achieving a Mean Absolute Error (MAE) of 54.91 days, a Root Mean Square Error (RMSE) of 86.54 days, and an R² value of 0.8582. Additional validation confirmed the model’s robustness, showing consistent performance across the validation and test datasets and indicating strong generalization capability. The results suggest that Eucalyptus replanting typically occurs approximately 1000 to 1500 days (3 to 4 years) after tree cutting. The proposed DSS enhances operational efficiency by optimizing replanting schedules using both real-time and historical data, ensuring the timely availability of raw materials. Furthermore, it has the potential to reduce operational costs by minimizing manual monitoring and optimizing resource utilization, such as water and fertilizers, thereby supporting more sustainable plantation management. Key factors influencing Eucalyptus growth include rainfall, temperature, humidity, soil pH, and nutrient levels. Future work may focus on incorporating additional environmental variables, such as long-term climate projections, and continuously updating the model with new data to further improve predictive performance.