Scholarlink

<p>In the digital era, Supply Chain Management (SCM) is challenged by issues of security, transparency, and operational efficiency. This research introduces a novel model that integrates deep learning with optimization techniques to improve sales forecasting within SCM. Specifically, the model combines the Bidirectional Gated Recurrent Unit (Bi-GRU) with the Mayfly Algorithm (MFA) to enhance forecasting accuracy and optimize decision-making processes. The Bi-GRU model captures the temporal dependencies in sales and demand data, while the MFA refines the model’s parameters for better performance. The proposed approach is tested using the SOK market dataset from Turkey, achieving outstanding results with a Root Mean Square Error (RMSE) of 0.06 and a Mean Absolute Percentage Error (MAPE) of 3.1%. These results demonstrate that the Bi-GRU-MFA model outperforms traditional deep learning models, providing a more reliable and efficient solution for sales analysis in SCM. This study concludes that the integration of Bi-GRU and MFA offers SCM professionals an advanced tool for improving sales management and operational efficiency.</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

Optimizing Retail Supply Chain Sales Forecasting with a Mayfly Algorithm-Enhanced Bidirectional Gated Recurrent Unit

  • 期刊论文

  • 发表日期：

    2025年8月11日

查看全文
  • Guna Sekhar Sajja,
  • Mohan Kumar Meesala,
  • Santosh Reddy Addula,
  • Pavankumar Ravipati

摘要

In the digital era, Supply Chain Management (SCM) is challenged by issues of security, transparency, and operational efficiency. This research introduces a novel model that integrates deep learning with optimization techniques to improve sales forecasting within SCM. Specifically, the model combines the Bidirectional Gated Recurrent Unit (Bi-GRU) with the Mayfly Algorithm (MFA) to enhance forecasting accuracy and optimize decision-making processes. The Bi-GRU model captures the temporal dependencies in sales and demand data, while the MFA refines the model’s parameters for better performance. The proposed approach is tested using the SOK market dataset from Turkey, achieving outstanding results with a Root Mean Square Error (RMSE) of 0.06 and a Mean Absolute Percentage Error (MAPE) of 3.1%. These results demonstrate that the Bi-GRU-MFA model outperforms traditional deep learning models, providing a more reliable and efficient solution for sales analysis in SCM. This study concludes that the integration of Bi-GRU and MFA offers SCM professionals an advanced tool for improving sales management and operational efficiency.