The chapter introduces the idea of cyber-physical multi-agent systems based on models for autonomous resource management (CP MAS). The definition, functionality, and architecture of the CP MAS presented are discussed. It is shown that the virtual world of CP MAS can be designed as a “twin-model” of real world, which plays a core component of “intelligence” of discussed CP MAS. The chapter suggests time metrics for the life-cycle characteristics of events, which can aid in assessing the adaptability level of CP MAS and determining “Time-to-Go” for autonomous decision-making. The proposed metrics will allow smart enterprises to detect the weak points of resource management schedule, which leads to system performance increasing and efficiency improving of resource usage, reduce expenses, risks and penalties. For instance, within the framework of CP MAS, there is a discussion on an Intelligent Cyber-Physical System utilizing digital twins of plants as an example. In the future, this developed solution could form the basis of an autonomous CP MAS designed for resource management in precision farming.