Generating insight in discrete event simulation and agent-based modelling: Experimental evidence

Abstract

This study aims to understand the difference between discrete-event simulation (DES) and agent-based modelling (ABM) user and non-user in generating insight. In total, 41 undergraduate students who were DES and ABM users and non-users participated in the experiment and given task of solving problem using simulation models. The insight generation is measured by four variables: (1) task performance, (2) problem understanding, (3) discontinuity in thinking, and (4) change in problem understanding. This study concludes that there are no significant differences between DES and ABM users and non-users in generating insight particularly for simulation with low complexity as utilised in this study. However, both ABM users and non-users had higher solution rates, indicating higher insight occurrence, than DES users and non-users. Since this study limited to simulation with low complexity, there is a room of improvement for further study to investigate different complexity level of the system simulated in generating insight.