FAIR: Facilitating Artificial Intelligence Resilience in Manufacturing Industrial Internet
Yingyan Zeng, Ismini Lourentzou, Xinwei Deng, Ran Jin
TL;DR
The paper addresses AI resilience in Manufacturing Industrial Internet (MII) by defining quantitative resilience metrics and a diagnosis framework. It introduces a Multimodal Multi-head Self Latent Attention (MMSLA) model to diagnose root causes across data, AI pipeline, and cyber-physical layers, and couples this with targeted mitigation strategies. The framework is validated on an Aerosol Jet Printing (AJP) testbed, where MMSLA outperforms baselines in identifying hazards (Y1–Y5) and enables effective mitigation actions, quantified via temporal and performance resilience metrics. The work provides a practical path toward reliable AI-enabled manufacturing by enabling online hazard detection, root-cause diagnosis, and context-aware mitigation in a cloud-fog ecosystem.
Abstract
Artificial intelligence (AI) systems have been increasingly adopted in the Manufacturing Industrial Internet (MII). Investigating and enabling the AI resilience is very important to alleviate profound impact of AI system failures in manufacturing and Industrial Internet of Things (IIoT) operations, leading to critical decision making. However, there is a wide knowledge gap in defining the resilience of AI systems and analyzing potential root causes and corresponding mitigation strategies. In this work, we propose a novel framework for investigating the resilience of AI performance over time under hazard factors in data quality, AI pipelines, and the cyber-physical layer. The proposed method can facilitate effective diagnosis and mitigation strategies to recover AI performance based on a multimodal multi-head self latent attention model. The merits of the proposed method are elaborated using an MII testbed of connected Aerosol Jet Printing (AJP) machines, fog nodes, and Cloud with inference tasks via AI pipelines.