Answers are grounded in this paper and cite their sources.
The increasing complexity of collaborative manufacturing systems is driving the transition from dyadic human-robot interaction toward multi-human multi-robot (MH-MR) configurations. In these systems, process performance depends both on individual task execution and on the quality of interactions among multiple agents, with varying degrees of intelligence and autonomy. In such settings, human operators and robotic agents interact through task allocation, control signals, and shared process-state information. However, conventional failure analysis methods, such as Failure Mode, Effects and Criticality Analysis (FMECA), are primarily task-centred and do not fully capture failures arising from interaction-level misalignments. This paper proposes a coherence-driven framework for the design and analysis of MH-MR systems, where failures are interpreted as the result of coherence violations across interacting agents. The methodology is structured as a stepwise process that starts from process decomposition, introduces agent configuration, derives interactions, and identifies critical coherence requirements. Potential coherence violations are then translated into generalized failure mode categories which are subsequently structured using conventional representation tools. The approach is analysed through a case study involving the collaborative assembly of a skateboard performed by two human operators and two robotic agents. The results show that, while execution-level failures remain relevant, additional failure mechanisms emerge from coordination, communication, and collaboration among agents. The proposed framework enables a shift from task-level failure analysis to an interaction-driven design approach, providing a novel tool to support practitioners during the design of MH-MR manufacturing processes.
No takes yet. Share an insight, caveat, or question.
Capponi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: