Converging on Emergence: Reconnoitering to Optimally Adapt to Changes in System Environment

Abstract

Emergence in complex systems can result in dynamic and unpredictable changes in the optimal operating state of deployed systems. This paper considers the implications of emergence from the perspective of a continuously changing objective function within a real-time systems optimization problem. To account for potential changes due to emergence, an algorithm and numerical operator are presented which facilitate detection of changes in the environment and adjustment to altered conditions, with the goal of minimizing the time to re-converge to optimal system operation. A numerical case study is performed to demonstrate the approach and validate its statistical effectiveness.