Scholarlink

<p>With the rapid digitalization of agricultural supply chains, optimizing agricultural warehousing through edge computing has gained increasing attention. This study proposes an edge-enabled predictive optimization mechanism for agricultural product warehousing, combining edge computing and a novel CBA-NGA algorithm. Edge devices are deployed for real-time acquisition and local processing of environmental data—such as temperature, humidity, wind speed, and wind direction—providing immediate feedback for warehouse management. A forecasting model based on the CBA-NGA algorithm is constructed to address inventory loss prediction, improving convergence speed and prediction precision over traditional models. Empirical results show that the model reduces mean squared error (MSE) from 33.5614 to 17.1911, demonstrating substantial performance enhancement. Compared to standard CBA methods, the CBA-NGA model offers better adaptation to complex, nonlinear environmental interactions under edge computing conditions. This research contributes to the literature by integrating CNN-BiLSTM-Attention-based predictive modeling with edge intelligence and NGA-based optimization, providing a scalable and adaptive solution for loss prevention in dynamic agricultural storage environments. The findings have practical implications for improving the efficiency, responsiveness, and sustainability of smart warehousing systems in modern agriculture.</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

Agricultural supply chain warehousing optimization mechanism based on CBA-NGA in the context of edge computing

  • 期刊论文

  • 发表日期：

    2026年6月1日

查看全文
  • Kaiyang Zhong,
  • Yu Feng,
  • Zehao Wang,
  • Zhiguo Zhong,
  • Yifei Lv

摘要

With the rapid digitalization of agricultural supply chains, optimizing agricultural warehousing through edge computing has gained increasing attention. This study proposes an edge-enabled predictive optimization mechanism for agricultural product warehousing, combining edge computing and a novel CBA-NGA algorithm. Edge devices are deployed for real-time acquisition and local processing of environmental data—such as temperature, humidity, wind speed, and wind direction—providing immediate feedback for warehouse management. A forecasting model based on the CBA-NGA algorithm is constructed to address inventory loss prediction, improving convergence speed and prediction precision over traditional models. Empirical results show that the model reduces mean squared error (MSE) from 33.5614 to 17.1911, demonstrating substantial performance enhancement. Compared to standard CBA methods, the CBA-NGA model offers better adaptation to complex, nonlinear environmental interactions under edge computing conditions. This research contributes to the literature by integrating CNN-BiLSTM-Attention-based predictive modeling with edge intelligence and NGA-based optimization, providing a scalable and adaptive solution for loss prevention in dynamic agricultural storage environments. The findings have practical implications for improving the efficiency, responsiveness, and sustainability of smart warehousing systems in modern agriculture.