Modeling Multi-Layer Access Control Policies of a Hyperledger-Fabric-Based Agriculture Supply Chain

Transparency and immutability properties of blockchains enhance traceability and trust in supply chains. However, supply chain participants resist storing most data on blockchains due to business confidentiality concerns. Permissioned blockchains can partly address such concerns. However, even on a permissioned blockchain, it is nontrivial to determine who can see what data given multiple access control models spanning multiple layers and various complex business relation-ships among the partners within a blockchain-based consortium. Consequently, most data are kept off-chain, vastly limiting the on-chain business logic execution that could otherwise enhance au-tomation, efficiency, and real-time compliance enforcement. This paper presents a multi-layered and multi-model access control analysis of an agriculture supply chain based on Hyperledger Fabric. First, we model five layers of access control in Fabric-based applications where different layers adopt different access control models. The smart contract layer may also adopt multiple and hierarchical access control models. Second, using NIST's access control rule logic circuit simulation technique, we modeled and validated the safeness of policies governing both Fabric configuration management and data access within the supply chain. Finally, we conducted a reflective privacy assessment to enhance supply chain participants' confidence in storing data on-chain by answering queries like “which other participants can see my data?”.