Multimethod Simulation Modeling for Business Applications

Simulation modeling stands apart within the range of modeling tools and technologies. Simulation models enable experimentation without the use of real-life objects. The virtual modeling world is completely risk-free. Managers, researchers, and engineers can create and test models of various system designs and answer hundreds of “what-if” questions, all by virtually experimenting in a risk-free environment.

In modern simulation modeling for business systems, there are three main approaches: system dynamics, discrete event and agent-based modeling. All three modeling approaches are quite independent. Each has its own tool set, unique features, strengths, and weaknesses. The choice of method should be based on the system under study and the purpose of the modeling.

However, most real-world cases are too complex to be modeled with one method, and it is often convenient to describe different parts of a system with different modeling approaches. A combination of approaches provides the opportunity to create an accurate and multi-functional model of a system without any workarounds.

In this paper, we cover the different approaches and define multi-method (also called hybrid) modeling. Moving from theory to practice, we build a multi-method simulation model example, analyze it, and demonstrate a solution to the most common issue for novice modelers: how to choose the most appropriate modeling method or combination of methods to meet your challenge and purpose.