The Framework for Designing Autonomous Cyber-Physical Multi-agent Systems for Adaptive Resource Management

Abstract

The paper contributes to design of autonomous cyber-physical multi-agent systems for adaptive resource management providing increase of efficiency of business operating in uncertain and dynamic environment. Evolution of multi-agent systems from purely decision-making support and simulation tool to cyber-physical system including Digital Twins and fully autonomous systems is analyzed. The main paper contribution is the proposed conceptual framework for designing autonomous cyber-physical multi-agent systems for adaptive resource management. It is shown in the paper that, in cyber-physical multi-agent systems for adaptive resource management, the ontology-customized multi-agent engine and ontology-based model of enterprise are forming ontology-driven “Digital Twin” of the enterprise providing opportunity to combine operational scheduling of resources with ongoing real-time simulations and evolutional re-design of configuration of enterprise resources. The functionality and architecture of the autonomous cyber-physical multi-agent systems for adaptive resource management are developed to support for the full cycle of autonomous decision making on resource management. Time metrics for measuring event-based response time and level of adaptability of autonomous cyber-physical multi-agent systems for adaptive resource management are proposed. Results of developments can be applied for smart transport and smart manufacturing, smart agriculture, smart logistics, smart supply chains, etc.