Applications of human-machine collaborative decision-making: A review of research with recent developments

This study comprehensively reviews human-machine collaborative decision-making (HMCD) methods and applications across management science, the military, healthcare, and manufacturing. We propose a dual-layer analytical framework. The first layer decomposes HMCD into four sequential stages, namely attribute determination, weight assignment, information aggregation, and decision-making. The second layer identifies four cross-cutting collaboration mechanisms, namely role configuration, interaction and deliberation, trust and explanation, and authority and responsibility migration. A feedback loop connects decision outcomes to earlier stages, capturing iterative adaptation between agents. Using this framework as an analytical lens, we synthesize methods and applications, examine stage-specific collaboration patterns across domains, and trace how failures propagate through mechanism dependencies. This review provides a structured map of HMCD research and guidance for collaborative system design.