The Engineering of Emergence in Complex Adaptive Systems

Agent-oriented software engineering is a new software engineering paradigm that is ideally suited to the analysis and design of complex systems. The main focus of these methodologies is to engineer a complex system in such a way that the correct emergent behavior results. In a complex adaptive system, however, emergent behavior cannot be predicted during analysis and design as it evolves only after implementation. By restricting emergent behavior as is done in most agent-oriented software engineering approaches, a complex system cannot be fully adaptive as well. We propose the BaBe methodology that will enable a complex system to be adaptive as the system can learn from its environment during run-time and modify its behavior in order to adapt to changes in its environment. This methodology adds a run-time emergence model consisting of distributed Bayesian Behavior Networks to the agent-oriented software engineering lifecycle. These networks are initialized by the human software engineer and deployed by Bayesian Agencies (also complex adaptive systems). The distributed Bayesian Behavior Networks, being specialized Bayesian Networks, will enable the Bayesian Agencies to collectively mine relationships between emergent behaviors and the interactions that caused them to emerge. This technology will further enable the Bayesian Agencies to collectively learn from the emerged behaviors and to modify the behavior of the system in order to adapt to changes in the environment. We describe a prototype implementation of the Bayesian Agencies using Sun’s Enterprise JavaBeansTM component architecture.