An industrial dataspace for automotive supply chain: Secure data sharing based on data association relationship

Abstract

The automotive supply chain (ASC) is a complex system involving every aspect of automobile manufacturing, from which the data obtained features such as massive volume, diverse types, and complex relationships. Traditional data management methods no longer meet the demands of handling heterogeneous data from multiple sources or ensuring secure cross-domain data sharing in the ASC, which leads to the isolation of information. Therefore, this paper proposes a data management method based on Industrial Dataspace (IDS), constructs a dataspace architecture for the automotive supply chain (DS-ASC). On this basis, proposes a method for data relationship mining and trusted data sharing that considers implicit associations among ASC members. The improved BiLSTM model promotes the understanding of data, and the improved DPoS algorithm reduces the risk of data leakage. Our method is validated in the practical application of a supply chain master enterprise, and the experiments show that the method proposed in this paper is able to effectively improve the accuracy of mining data association relationship. Meanwhile, it is able to prevent single-point attacks, and ensure the security of data sharing.