Mapping the knowledge frontiers and evolution of decision making based on agent-based modeling

摘要

With increasing attention paid to the application of agent-based modeling in decision-making issues, a large number of related studies have been published in management science and operational research areas. This study adopts multiple methods, including bibliometric mapping, text mining, and qualitative analysis, to comprehensively review relevant research to explore knowledge frontiers and evolution in decision-making research based on agent-based modeling (DM-ABM). This review is based on 1190 relevant journal articles from the Web of Science Core Collection dataset and 37,167 collectively cited references of the articles. The top ten most-cited studies that constitute the intellectual milestones of DM-ABM were identified. Keywords and research topics develop rapidly; recent research have paid most attention to the keywords “model,” “system,” and “simulation” and topics “learning,” “contracts,” “protocols,” and “self-learning.” The top 24 references with the strongest citation bursts were displayed to show that the area was increasingly active from 2001 to 2010. Transition points were mapped to reveal the top five studies with the highest betweenness centrality, which considerably influences knowledge evolution. Then, the top three clusters are identified as the frontier areas and analyzed by text mining, including intelligent agents, model validation, and collaborative decision making. Finally, the most recent research in this field is investigated, and four future research directions are proposed: the advanced intelligence of agents, approach to reality, group decision making, innovative modeling methodologies and diversified applications.