Towards Automated Handling of Object State Exceptions in Smart Contracts

Blockchain-based process execution has gained significant attention from the BPM community in the previous years. The ability to guarantee the correct execution of processes is considered an important factor for applications in low-trust environments and for the implementation of smart contracts in general. However, enforcing the correct execution of the control flow is a two-edged sword as it naturally minimizes flexibility. At the same time, the enforcement of the process is only partial since the blockchain cannot control the state of real-world objects: physical objects and off-chain data objects. Consequently, the actual object state may deviate from the one assumed by the smart contract. Smart contract models should allow the automatic handling of exceptions arising from such deviations while still guaranteeing contractual behaviors. This paper introduces a modeling approach for smart contracts that utilizes imperative modeling of processes in combination with declarative modeling of object life-cycles and consensus mechanisms. This combination allows us to agree on exceptions, to automatically derive plans for resolving object state exceptions at runtime within the boundaries of the agreed smart contract, and to agree on a selected plan.