Towards trustworthy Cyber-physical Production Systems: A dynamic agent accountability approach

Abstract

Smart manufacturing is a challenging trend being fostered by the Industry 4.0 paradigm. In this scenario Multi-Agent Systems (MAS) are particularly elected for modeling such types of intelligent, decentralised processes, thanks to their autonomy in pursuing collective and cooperative goals. From a human perspective, however, increasing the confidence in trustworthiness of MAS based Cyber-physical Production Systems (CPPS) remains a significant challenge. Manufacturing services must comply with strong requirements in terms of reliability, robustness and latency, and solution providers are expected to ensure that agents will operate within certain boundaries of the production, and mitigate unattended behaviours during the execution of manufacturing activities. To address this concern, a Manufacturing Agent Accountability Framework is proposed, a dynamic authorization framework that defines and enforces boundaries in which agents are freely permitted to exploit their intelligence to reach individual and collective objectives. The expected behaviour of agents is to adhere to CPPS workflows which implicitly define acceptable regions of behaviours and production feasibility. Core contributions of the proposed framework are: a manufacturing accountability model, the representation of the Leaf Diagrams for the governance of agent behavioural autonomy, and an ontology of declarative policies for the identification and avoidance of ill-intentioned behaviours in the execution of CPPS services. We outline the application of this enhanced trustworthiness framework to an agent-based manufacturing use-case for the production of a variety of hand tools.

Manufacturing Execution System Integration through the Standardization of a Common Service Model for Cyber-Physical Production Systems

“…As an alternative response to the concerns of distributed control in manufacturing, recently a Manufacturing Agent Accountability Framework has been proposed, which is a dynamic authorization framework that defines and enforces boundaries in which agents are freely permitted to exploit their intelligence to reach individual and collective objectives [45].…”