Towards a Guide to Domain-specific Hybrid Simulation Anatoli Djanatliev, Reinhard German

Due to an increasing complexity of current models, many established simulation and modeling techniques nare not sufficiently powerful to solve large-scaled problems solely. Using only discrete modeling methods can lead to an exponentially growing model complexity, and continuous approaches are not able to represent individuality.

These limitations are a major reason why hybrid simulation modeling is receiving a growing attention in recent times. The main idea is to combine two or more different simulation paradigms aiming to take profits from the advantages of all applied methods. In particular, the Discrete-Event Simulation (DES), System Dynamics (SD), and the Agent-Based Simulation (ABS) are highly relevant techniques in this scope

In this paper, we particularly outline three major processes which are important to define a specialized hybrid approach in domain-specific contexts. Most important concepts will be explained using an example hybrid simulation definition for prospective healthcare decision-support. Furthermore, we focus on practical issues aiming to sustainable model developments (e.g., tooling, framework building).

Overview of our hybrid simulation approach for prospective healthcare decision-support
Download full paper (PDF)