A Statecharts-Based Model-Driven Approach for Blockchain Execution of Inter-Organizational Business Processes with Multiple Instances

With the growing need for inter-enterprise collaboration, Inter-Organizational Business Processes (IOBPs) have become essential in modern business networks. BPMN 2.0 choreography diagrams effectively describe cross-organizational interactions, but lack clear semantic support and implementation mechanisms for multi-instance scenarios in permissioned blockchain environments. Although Model-Driven Architecture (MDA) approaches help automate the transformation of BPMN models into smart contracts (SCs), existing methods based on Finite State Machines (FSMs) face challenges such as state explosion and limited concurrency support, as the number of states grows combinatorially with the number of participants (e.g., O ( N 2 )) and parallel behaviors must be flattened, resulting in poor scalability and maintainability, making them inadequate for multi-instance execution. To address these issues, we first provide a formal semantic description for BPMN multi-instance participants and tasks in permissioned blockchain environments, clarifying their expected behaviors in decentralized execution. Based on this, an enhanced statechart-based approach is proposed and a supporting framework named MultiChainCollab is developed. The framework integrates statecharts into the MDA approach to enable independent multi-instance state control while enhancing collaborative process data recording in BPMN SCs. Furthermore, we implement an open-source multi-party collaboration platform based on Hyperledger Fabric that supports multi-instance execution, dynamic participant management, and process visualization. A real-world case study and experimental evaluation demonstrate the correctness, scalability, and practical value of our approach. Code is available at: https://github.com/HIT-ICES/ChainCollab/tree/release-Multi-BlockCollab .