Discrete-Event Simulation and System Dynamics for Management Decision Making

12Multi-Method Modelling: AnyLogic

Andrei Borshchev

Managing Director and CEO, The AnyLogic Company, St Petersburg, Russia

The three modelling methods, or paradigms, that exist today are essentially the three different viewpoints the modeller can take when mapping the real-world system to its image in the world of models. The system dynamics (SD) paradigm suggests abstracting away from individual objects, thinking in terms of aggregates (stocks, flows) and feedback loops. The discrete-event (DE) modelling paradigm adopts a process-oriented approach: the dynamics of the system are represented as a sequence of operations performed over entities. In an agent-based (AB) model the modeller describes the system from the point of view of individual objects that may interact with each other and with the environment.

Depending on the simulation project goals, the available data, and the nature of the system being modelled, different problems may call for different methods. Also, sometimes it is not clear at the beginning of the project which abstraction level and which method should be used. The modeller may start with, say, a highly abstract system dynamics model and switch later to a more detailed discrete-event model. Or, if the system is heterogeneous, the different components may be best described by using different methods. For example, in the model of a supply chain that delivers goods to a consumer market, the market may be described in (SD) terms, the retailers, distributors, ...