Edge AI for Industrial IoT Decision Making: Current Advances and Future Directions

Abstract

The convergence of Artificial Intelligence (AI) and the Industrial Internet of Things (IIoT) has transformed industrial systems by enabling real-time decision making at the edge. Traditional cloud-centric architectures face challenges of latency, bandwidth consumption, and data privacy, which are critical in time-sensitive industrial environments. Edge AI integrates AI capabilities with edge computing, bringing intelligence closer to IIoT devices, thereby enabling faster and context-aware decision-making without over-reliance on centralized cloud infrastructures. This paper surveys current advances in edge AI for IIoT decision making and highlights future directions. It explores architectures, algorithms, and applications that support intelligent decision-making at the network edge, with emphasis on manufacturing, predictive maintenance, supply chain optimization, and energy management. The review discusses enabling technologies such as lightweight deep learning models, federated learning, and hardware accelerators, alongside challenges including scalability, interoperability, and cybersecurity. Findings from literature suggest that edge AI enhances operational efficiency by reducing latency, improving reliability, and enabling autonomous decision-making in distributed industrial systems. However, barriers remain in terms of resource constraints, integration with legacy systems, and lack of standardized frameworks. Future research is expected to focus on adaptive AI models, edge–cloud collaboration, and the use of 6G-enabled IIoT ecosystems. By synthesizing state-of-the-art approaches, this paper provides insights into how edge AI can drive the next wave of industrial automation and digital transformation. It emphasizes the need for robust, secure, and scalable frameworks to fully realize the potential of edge AI for IIoT-based decision making.