Automating agent-based modeling: Data-driven generation and application of innovation diffusion models

Simulation modeling is useful to understand the mechanisms of the diffusion of innovations, which can be used for forecasting the future of innovations. This study aims to make the identification of such mechanisms less costly in time and labor. We present an approach that automates the generation of diffusion models by: (1) preprocessing of empirical data on the diffusion of a specific innovation, taken out by the user; (2) testing variations of agent-based models for their capability of explaining the data; (3) assessing interventions for their potential to influence the spreading of the innovation. We present a working software implementation of this procedure and apply it to the diffusion of water-saving showerheads. The presented procedure successfully generated simulation models that explained diffusion data. This progresses agent-based modeling methodologically by enabling detailed modeling at relative simplicity for users. This widens the circle of persons that can use simulation to shape innovation.