Coensys, Inc.

Multi-Paradigm Simulation Using AnyLogic

 Anylogic is unique in its ability to develop multi-paradigm models. An Anylogic model can contain concepts from: 

• System dynamics (SD), 

• Discrete Event systems (DES), and 

• Agent Based modeling (ABM). 

One of the key mechanisms used to bring these different paradigms together is the Hierarchical State Machines (HSM). Anylogic supports Hierarchical state machines and use HSMs to not only add richness to models, but also to enable entities modeled in different paradigms to interact with each other.

As modelers we often face the dilemma as to how best to model a complex situation or system. Traditionally, modelers decision has been tainted by the tools they use, because most tools only support one way of modeling a system. This may lead a modeler to miss out on interesting aspects of a system's behavior. Anylogic has solved this problem. With Anylogic you can start with any modeling paradigm and at any time add new aspects of a system in another paradigm that best describe the system or situation being modeled.

There are many debates as to which approach is better. In our experience, no approach is better than the other. It is just that any single approach cannot be effectively employed for all situations. Each approach has its strengths and weaknesses. When to apply which approach requires experience with each one. Modeling the same problem from multiple perspectives can sometimes provide great insights into the problem. Having a multi-paradigm tool like Anylogic helps achieve this understanding and learning. With Anylogic you can focus on appropriately modeling your problem as opposed to wasting time in methodology debates.

So one could start with a system dynamics model, discover some aspect of the system that requires modeling at individual level, add that aspect of the model using agent based modeling capabilities, integrate the two aspects using hierarchical state machines, all within the same model.

We have just begun to realize the power of this approach. Still not convinced, download a free evaluation version to see for yourself. You will soon find yourself creating models using multiple paradigms.