Entropy-probabilistic Methods and Models of Risk Analysis to Increase the Reliability of Complex Adaptive Systems

• Applicant: Professor Dr.-Ing. Michael Beer
• Subject Area: Structural Engineering, Building Informatics and Construction Operation
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 547125816

Every year we observe an increase regarding the efficiency of newly developed systems. Smart systems began to appear, which include hundreds of elements that interact not only with a person, but also with each other. They allow creating smart control systems, intelligent production, offices and homes. But at the same time, the management of such systems is becoming more and more complicated. The main reasons for that are the growth of information uncertainty, the complication of risks, the emergence of systemic risks which are unknown how to take into account. The scientific problem that the project is aimed to solve is ensuring the reliability and stability of complex adaptive systems operating in conditions of information uncertainty and different nature of risks. The main goal of the project is to propose new approaches for timely anticipating of crisis situations in complex adaptive systems and developing management recommendations to ensure the sustainability of their functioning. The scientific novelty of the research lies in the fact that there are no specific approaches that allow us to effectively solve the problem posed in the project. The system regularities will be described using the differential entropy of a continuous random vector, presented in vector form as two components – the randomness entropy and the self-organization entropy. The project proposes to combine the vector entropy model and a risk model. This allows us to consider the reliability of a complex adaptive system within the framework of the category of its “sustainable development”. Under the sustainable development of a complex adaptive system, we mean its dynamics, which consists in a balanced change in the vector entropy while maintaining an acceptable level of risk. The following main tasks will be solved in the project: 1. Development and research of methods for analyzing systemic risk and the risk of complex interconnected networks. 2. Development of a method for modeling complex adaptive systems in the form of multidimensional stochastic interconnected network structures. 3. Development of methods and models for the formation of management recommendations to improve the reliability of adaptive network systems. 4. Development of entropy-probabilistic methods for monitoring the reliability of complex network systems, their verification and validation. Testing on various systems will confirm the results of the study increasing the scientific significance of the study and directly demonstrate its practical usefulness in solving real problems.

• DFG Programme: Research Grants
• International Connection: USA
• Cooperation Partner: Professor Konstantin Zuev, Ph.D.