Scholarlink

<p>Obtaining optimal environmental values is crucial in agriculture for maximising crop yield and ensuring sustainable production. In this direction, the current study presents a novel approach to optimise statewide wheat yield in India, by developing a Decision Support System (DSS) based on Deep Learning (DL)-integrated with Response Surface Methodology (RSM). The dataset used in the present study consists of over eight years (2012–2020) of environmental data along with the wheat yield data collected across the districts of Madhya Pradesh. Leveraging this extensive dataset, the study employed a DL-based framework to analyse the complex relationships between environmental variables and wheat yield. The study results show that the developed crop yield estimation model’s metrics on testing data include RMSE: 0.83 t/ha, MAE: 0.66 t/ha, MAPE: 24.0%, and RRMSE: 0.27. In addition, by integrating DL with RSM, optimal levels of these environmental variables were determined to maximise wheat yield. The resulting recommendations serve as a foundation for the DSS, offering actionable insights for irrigation management and temperature control in wheat fields.</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

Decision Support System for Maximizing Wheat Productivity in India

  • 期刊论文

  • 发表日期：

    2026年3月10日

查看全文
  • Samarth Godara,
  • Ram Swaroop Bana,
  • K. K. Chaturvedi,
  • S. B. Lal

摘要

Obtaining optimal environmental values is crucial in agriculture for maximising crop yield and ensuring sustainable production. In this direction, the current study presents a novel approach to optimise statewide wheat yield in India, by developing a Decision Support System (DSS) based on Deep Learning (DL)-integrated with Response Surface Methodology (RSM). The dataset used in the present study consists of over eight years (2012–2020) of environmental data along with the wheat yield data collected across the districts of Madhya Pradesh. Leveraging this extensive dataset, the study employed a DL-based framework to analyse the complex relationships between environmental variables and wheat yield. The study results show that the developed crop yield estimation model’s metrics on testing data include RMSE: 0.83 t/ha, MAE: 0.66 t/ha, MAPE: 24.0%, and RRMSE: 0.27. In addition, by integrating DL with RSM, optimal levels of these environmental variables were determined to maximise wheat yield. The resulting recommendations serve as a foundation for the DSS, offering actionable insights for irrigation management and temperature control in wheat fields.