Simulation-based engineering of complex systems using EXTEND+MFG+OpEMCSS

A Complex Adaptive System (CAS) is a network of self-organizing, intelligent agents that share knowledge and adapt their operations in order to achieve overall system goals. Three things are needed to understand, design, and evaluate CAS. First, a mathematical model or way-of-thinking about CAS, called Context-Sensitive Systems (CSS) theory, is required to provide a solid foundation upon which to represent and describe the kinds of interactions that occur among the CAS agents during system operation. Second, a graphical modeling language is required that implements CSS theory in a way that enhances visualization and understanding of CAS. Third, a systems design and evaluation tool is required that makes it easy to apply CSS theory, expressed using a graphical modeling language, to understand, design, and evaluate CAS. As an example, an OpEMCSS model of two intelligent agents is discussed that learn rules and maximize their average reward in the prisoner´s dilemma game.