Can We Engineer the Emergent Behavior of A System-of-Systems?

摘要

Emergent behaviors exist in biological systems, physical systems and human performance. It isrnan inherited nature of a System-of-Systems (SoS). SoS displays a global complexity that cannotrnbe adequately managed by hierarchical structures and central control; therefore, traditionalrnsystems engineering and management approaches are necessary but insufficient for a SoS.rnLittle is currently known about constructing an interoperable network of systems and thernincorporation of emergent behaviors. The purpose of this panel is to explore the possibilities ofrndeveloping an architecture model including the emergent behavior.rnThe challenge is how to understand the initiation mechanisms of the emergent behaviors for arnparticular system architecture model so that the resident beneficial or harmful emergentrnbehaviors can be enhanced or mitigated with selected changes in the model. Is model-based thernonly feasible approach to develop the architecture model with emergent behavior? If this is thernanswer, what kind of modeling methodology? Should it be solely based on agent-basedrnmodeling or a combination of SysML and agent-based? Is SysML ready to deal with emergentrnbehavior?rnFor a non-modeling consideration, can we plan for the beneficial or harmful emergent properties?rnHow do we overcome development friction that is bound to arise when there are complex,rnindependent, overlapping governances, for example, the customer requirements for a SoS evolvernover time, etc.? There may only be SoS modeling at the level of the government agency, whichrnactually procures but does not build, and not at the level of contractors who build the nextrngeneration of technologies.