A Cyber Manufacturing IoT System for Adaptive Machine Learning Model Deployment by Interactive Causality Enabled Self-Labeling

TL;DR

The paper addresses the challenge of deploying adaptive ML in cyber manufacturing by introducing AdaptIoT, a containerized, microservice platform that integrates an end-to-end data streaming pipeline, ML services, and an interactive causality–enabled self-labeling workflow. The core innovation is the Interactive Causality Engine (ICE), which leverages a causal knowledge graph and coordinated submodels (ESD, ITM, and task models) to autonomously label data and retrain models post-deployment, mitigating data distribution shifts. The authors demonstrate a field-ready system in a makerspace, detailing the software/hardware architecture, data flows, and a real self-labeling experiment that shows superior performance over baseline semi-supervised methods, with scalable throughput and low latency suitable for SMBs. Overall, AdaptIoT provides a practical pathway to deploy personalized, self-labeling, edge-enabled ML in small-to-medium manufacturing environments, with open-source components and a clear route to future capability expansion.

Abstract

Machine Learning (ML) has been demonstrated to improve productivity in many manufacturing applications. To host these ML applications, several software and Industrial Internet of Things (IIoT) systems have been proposed for manufacturing applications to deploy ML applications and provide real-time intelligence. Recently, an interactive causality enabled self-labeling method has been proposed to advance adaptive ML applications in cyber-physical systems, especially manufacturing, by automatically adapting and personalizing ML models after deployment to counter data distribution shifts. The unique features of the self-labeling method require a novel software system to support dynamism at various levels. This paper proposes the AdaptIoT system, comprised of an end-to-end data streaming pipeline, ML service integration, and an automated self-labeling service. The self-labeling service consists of causal knowledge bases and automated full-cycle self-labeling workflows to adapt multiple ML models simultaneously. AdaptIoT employs a containerized microservice architecture to deliver a scalable and portable solution for small and medium-sized manufacturers. A field demonstration of a self-labeling adaptive ML application is conducted with a makerspace and shows reliable performance.

Figures (9)

• Figure 1: An illustration of the overall procedure of self-labeling.
• Figure 2: A high-level block diagram of the proposed IIoT system for self-labeling applications.
• Figure 3: (a) A simplified knowledge graph for a 3D printer use case; (b) Corresponding state transition relationships of the causally related Hand&Arm node and Controller node.
• Figure 4: Hardware and software infrastructure. Each block represents a containerized software service.
• Figure 5: The web GUI to display real-time data and ML results where users can also apply control to the system.