A generalized computational model for modeling and simulation of complex systems
The use of computer models and simulation is a widely adopted approach to study complex systems. To this end a diverse set of computational models like Cellular Automata, Artificial Neural Networks, or Agent-based simulation is being used. As a common denominator virtually all of these approaches favor different variations of complex systems and are tailored to support the description of systems that fit the corresponding variation well. Although this form of specialization has its benefits like ease of modeling with respect to the particular subset of complex systems, the drawbacks of this specialization are a lack of comparability between structurally different systems and a diminished expressiveness with respect to systems that do not fit any particular subset of complex systems favored by existing, specialized models. In this paper a generalized computational model for complex systems is proposed which allows for the description of most types of systems with a single model. Furthermore, the proposed model provides a high degree of encapsulation and reduces the amount of shared knowledge needed among the constituents of the system. The paper closes with a set of example applications of the proposed model to further illustrate the involved concepts and to provide an intuition on how this model may be used.