DS-DUC: A Fine-Grained Data Usage Control Method for the Automotive Supply Chain Based on Industrial Data Space and Extended Usage Control Model
Main article
Abstract
The automotive supply chain (ASC) constitutes one of the world's most complex multi-tier industrial ecosystems, encompassing thousands of suppliers, manufacturers, and service entities that collectively produce and maintain approximately 90 million vehicles annually. The competitive advantage, regulatory compliance, and operational efficiency of ASC participants are increasingly dependent on seamless, secure, and trustworthy data sharing across organizational and national boundaries. However, the ASC is uniquely challenging for data governance: its high data complexity and diversity span geometric CAD models, vehicle ECU firmware, real-time telematics, warranty records, and supplier quality certifications; frequent data updates follow product development cycles with update frequencies ranging from hourly (telematics) to annually (vehicle specifications); and the multi-party collaboration structure creates heterogeneous trust relationships with asymmetric power dynamics between OEMs and their supplier tiers. Existing data governance solutions fail to adequately address the concerns of data providers regarding continuous oversight of data usage and access conditions after data release—a capability known as data sovereignty. This paper proposes DS-DUC (Data Space-based Data Usage Control), a novel data usage control method for the ASC tailored to the Industrial Data Space (IDS) architecture. DS-DUC introduces an improved Extended Usage Control (EUCON) model that augments the classical UCON framework with mutable obligation attributes and persistent post-access monitoring capabilities, seamlessly integrated with IDS connector technology. The proposed DS-ASC-UC framework separates usage policies from enforcement mechanisms to achieve modularity and adaptability, enabling fine-grained, context-aware data access management across the entire ASC. Experimental evaluation on a prototype implementation involving three simulated ASC tiers demonstrates that DS-DUC achieves policy evaluation latency of 18.9 ms (58% reduction vs. XACML-based baseline), throughput of 680 req/s at 1,000 concurrent nodes (224% improvement over XACML), and violation detection rates exceeding 88% across all five policy violation categories. Security analysis confirms comprehensive coverage of data sovereignty requirements including confidentiality, access control, audit traceability, and dynamic policy adaptation, positioning DS-DUC as a practical solution for next-generation ASC data governance
