Parallel Intelligence for CPHS: An ACP Approach

Abstract

Cyber–Physical–Human Systems (CPHS) provide a new perspective and framework to support effective and efficient operations of complex systems, especially those that involve human and social factors in their management and control. However, the complexity of human and social behaviors in CPHS intensifies the differences between the models of systems and the systems to be modeled, thus leads the so-called "Cognitive Gap" or "Modeling Gap" for complex systems. Consequently, Merton's Laws rather than Newton's Laws must be applied in describing, predicting, and prescribing entities for CPHS. The ACP approach, which consists of "Artificial Systems" for description, "Computational Experiments" for prediction, and "Parallel Execution" for prescription or control and management, has transformed system models from system analysis into software-defined processes for data generation to make complex systems computable, testable, and verifiable. Based on ACP, Parallel Intelligence (PI) is constructed from the interactions and entanglement between actual systems and artificial systems. This chapter summarizes the research on PI and CPHS over the past 20 years. After a brief description of the history and framework, various applications are presented along eight aspects: parallel control and intelligent control, parallel robotics and parallel manufacturing, parallel management and intelligent organizations, parallel medicine and smart healthcare, parallel ecology and parallel societies, parallel economic systems and social computing, parallel military systems, and parallel cognition and parallel philosophy. Finally, the technical support and future direction for CPHS development are addressed.