Demand forecasting is a critical component of supply chain management, enabling businesses to align product supply with customer demand. Various methods exist for predicting demand, each with its strengths and limitations. In this paper, we propose a hybrid grey-neural network model to enhance prediction accuracy. The grey prediction model is effective for forecasting with small data sets but can be sensitive to anomalies, leading to significant errors. On the other hand, artificial neural networks are robust in handling complex data patterns, yet they require large-scale, well-representative training data, which can be challenging to obtain. By integrating these two approaches, our hybrid model aims to reduce prediction errors and improve forecast stability. The results demonstrate the model's effectiveness in achieving more accurate demand predictions, addressing the limitations of each method individually.