With the continuous development of global economic integration, light industry, as an important part of manufacturing industry, faces the challenge of cost control and resource optimization. In light industry, transparent component cost allocation is a key issue, which involves the collaboration between multiple links and multiple participants. To solve this problem, a data sharing method based on neural network is proposed in this paper. First, we introduce the background and related research of transparent component cost allocation in light industry. We then use convolutional neural networks (CNNS) and data distribution Estimation (DDE) to train and model the data. By learning patterns and trends in the data, we are able to predict future supply needs and optimize transparent cost and cost allocation for data sharing. In addition, we also improved a data volume model (DVM) for data sharing to achieve synergistic light industry effects and improvement space through CNN and DDE. Next, through the learning and training of the historical data, the Volume Warp Model (VWM) model can accurately predict the cost of transparent components. In order to verify the effectiveness of this method, we conducted a set of experiments and carried out experimental analysis. Finally, the experimental results show that the data sharing method based on neural network can accurately predict the cost of transparent components and achieve fair distribution among participants, and the model has higher accuracy and efficiency.