Assumptions and simplifications in discrete-event simulation modelling

All discrete-event simulation models are approximations of the real system that they represent. It is through making assumptions and incorporating simplifications into such models that they become an approximation. However, despite their importance, the 4definition, role and management of assumptions and simplifications is not well understood. In this paper we explore assumptions and simplifications in discrete-event simulation modelling in detail. Our approach is firstly to explore the existing literature on this topic, and then to draw this together with our extensive experience of real projects to recommend an approach to managing assumptions and simplifications in a simulation study. In doing so, we locate the role of assumptions and simplifications in a simulation study against knowledge acquisition and model abstraction respectively. We define the terms "assumptions" and "simplifications", also identifying the role of "simplifying assumptions". An approach to documenting, assessing and treating assumptions and simplifications which has been implemented in real projects is then described; and it is illustrated through application to a Ford Motor Company case study. This paper provides much needed clarity on the topic of assumptions and simplifications in simulation modelling, and it sets out a means for managing assumptions and simplifications during a simulation study.