This volume presents selected papers from the 22nd International Workshop on Multi-Agent-Based Simulation (MABS 2021), a workshop hosted by the 22nd International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS 2021), originally planned to be held in London, which took place as a fully virtual event during May 3–7, 2021. The meeting of researchers from multi-agent systems (MAS) engineering and the social/economic/organizational sciences is recognized as a source of cross-fertilization, and it has undoubtedly contributed to the body of knowledge produced in the MAS area. The excellent quality of this workshop has been recognized since its inception and its proceedings have been regularly published in Springer’s Lecture Notes in Artificial Intelligence series. More information about the MABS workshop series may be found at https://www.pcs.usp.br/~mabs/. The goal of the workshop is to bring together researchers interested in MAS engineering with researchers aiming to find efficient solutions to model complex social systems from areas such as economics, management, organization science, and social sciences in general. In all these areas, agent theories, metaphors, models, analyses, experimental designs, empirical studies, and methodological principles all converge in simulation as a way of achieving explanations and predictions, exploration and testing of hypotheses, and better designs and systems.