Deep Learning-Based Intelligent Supply Chain Management for Optimized Member Selection and Operational Efficiency

Introduction: Efficient supply chain management (SCM) is crucial for increasing competitiveness, notably through improved member (supplier/partner) selection and operational decision-making. Traditional techniques frequently rely on manual evaluations or static rule-based systems, which have limited scalability, adaptability, and real-time data processing capabilities. Objective: The goal of this research is to create an intelligent supply chain management (ISCM) framework that uses deep learning (DL) and metaheuristic optimization to improve supplier selection and overall operational efficiency. Method: A real-world supply chain dataset from open source Kaggle, which includes supplier performance measurements, delivery schedules, demand forecasting, and transaction history. The dataset is preprocessed using min-max normalization. Feature extraction is utilizing Principal Component Analysis (PCA). This research proposes a Flying Fox Optimized Artificial Neural Network (FlyFO-ANN) method based on an Artificial Neural Network (ANN) network, which is suggested for predicting supplier reliability and demand fluctuations. In addition, a Flying Fox Optimization (FFO) is used to modify model hyperparameters and optimize member selection criteria. The proposed FlyFO-ANN model is evaluated against baseline methods. Result: The experimental results reveal a significant increase in accuracy (0.9233) compared to other methods. The proposed framework is more adaptable and efficient than existing methods. Conclusion: Therefore, combining DL with intelligent optimization improves SCM decision-making by overcoming constraints in static approaches and enabling scalable, data-driven supply chain operations.