The emergence of organizational routines from habitual behaviours of multiple actors: an agent-based simulation study

Organizational routines are generative, dynamic systems involving multiple actors, and are crucial to understanding how organizations typically accomplish their tasks. In this paper, by regarding the individual habits of multiple actors as fundamental building blocks, we investigate organizational routines through a ‘bottom-up’ approach. On the basis of an individual actor-based perspective, we provide a general framework for analysing the internal structures of organizational routines and carry out agent-based simulation analysis based on a special empirical background. The results of our research show that performance feedback loops underlying interactions between actants, including both human actors and non-human artefacts, are key factors underpinning the formation and dynamics of organizational routines, and that imitation by individual actors may accelerate the emerging processes of these routines. For routines deliberately designed by management, an extended model from the perspective of evolution games indicates that a better balance of payoffs between different individual actors is important, and that the initial strategic distributions of individual actors are sensitive to the formation of routines. Agent-based simulation and evolutionary game theory thus provide quantitative but effective approaches that allow us to open up the ‘black box’ of organizational routines in two completely distinct forms—those that emerge from repeated interaction between individual actors and those deliberately designed by management—and to explore their formation and micro-dynamics.