Sustainable supply chain management using Bi-GRU network optimized by hybrid maritime search and rescue algorithm

Area of Science:

• Agricultural Science
• Supply Chain Management
• Artificial Intelligence

Background:

• Agricultural and food supply chains face challenges like waste, traceability issues, and profit optimization.
• Dynamic disturbances (weather, transport delays) and limitations of traditional methods necessitate advanced solutions.

Purpose of the Study:

• To propose and evaluate a hybrid AI framework for sustainable agricultural supply chain management.
• To address challenges including waste reduction, traceability, and profit optimization through enhanced decision-making.

Main Methods:

• A hybrid framework combining Bidirectional Gated Recurrent Unit (Bi-GRU) networks and a Hybrid Maritime Search and Rescue (HMSR) algorithm.
• Utilizing IoT sensors and market trends for time series data with Bi-GRU for prediction.
• HMSR algorithm optimizes decisions based on production capacity, storage constraints, and market demand.

Main Results:

• Achieved 92.4% accuracy in storage decisions.
• Reduced waste by 34.2%.
• Maximized profit margins by 28.7%, outperforming baseline models (GRU, LSTM, GA, PSO, HGSODL-ASCM).

Conclusions:

• Hybrid AI approaches, like the proposed Bi-GRU and HMSR framework, can significantly improve agricultural supply chain sustainability.
• The framework offers adaptable and scalable solutions for profitability and waste reduction in the agricultural sector.