Edge Computing for Real-Time Decision-Making in Industrial Automation Systems - A Comprehensive Review

Abstract

The real-time decision-making potential brought by edge computing is disrupting industrial automation, transcending the speed-and-bandwidth limitations of traditional cloud computing. By enabling on-site data processing, edge computing effectively low-latency and responsive applications, e.g., predictive maintenance, process optimization, and autonomous robotics. This paper discusses edge computing architecture and deployment models, its hardware and software components, and networking infrastructure. The objectives to review are to analyze how edge computing facilitates real-time decision-making in automated industrial processes. The discourse is then fully addressed around the different core technologies, advantages, concerns, real-life instances, and future trends toward increasing efficiencies, lowering latencies, and making smart manufacturing possible. Integration of AI and machine learning at the edge levels make real-time analytics, thus improving efficiency and agility. Key challenges under consideration include security, scalability, and economic impact, along with their applications in the real world. Future direction encompasses AI-based analytics, energy-efficient edge devices, and seamless interoperability poised to change its status from enabler to disrupter concerning industrial automation.