Self-adaptive dynamic decision making processes

Abstract

Decision making is important for many systems and is fundamental for situation awareness and information fusion. When a decision making process is confronted with new situations, goals and kinds of data, it must evolve and adapt. Highly optimized processes and efficient data structures generally have the disadvantage of having little flexibility or adaptability when confronted with new forms of data and new or changing goals. Consequently, optimized processes may only be locally optimal and may deteriorate over time. The normal approach to changing conditions is to manually reconfigure and even redevelop the system, which can be costly and time-consuming. In this article. we propose an architecture for the self-adaptation of decision making processes using flexible data structures and a process that monitors and adapts the decision making process. The objective is to have the ability to adapt both data schemas and decision making processes so that they can be both responsive and efficient.