The project "Understanding and Prediction of Critical Transitions in Complex Systems" is supported by The Collaborative Activity Award for Studying Complex Systems from 21st Century Science Initiative of James S. McDonnell Foundation.

     Nature and Society separately and jointly form many complex systems of interacting elements. This project is aimed at their prominent feature: persistent reoccurrence of abrupt overall changes called critical transitions. Among examples are the Earth's crust generating catastrophic earthquakes; Economy generating recessions and surges of unemployment; Megacities prone to mass violence, etc. In a way, the whole global village consists of overlapping systems of this kind. This project is focused on scenarios of development of critical phenomena, with applications to their algorithmic prediction and control.

    At the heart of this research is the integration of three methodologies: pattern recognition analysis of relevant observations; "universal" models of critical phenomena such as considered in non-linear dynamics; and models of specific systems. In addition, the theory of optimal control used to link disaster prediction with disaster preparedness.

    A decisive part of the project is the rigorous test of results by advance prediction, registering both successes and errors.