Machine Learning in the Internet of Things for Industry 4.0

TL;DR

This paper explores how to organize machine learning-enabled IoT systems in Industry 4.0, focusing on dataflow processing, latency considerations, and architectural patterns for edge and cloud deployment.

Contribution

It introduces a flow processing stack and architectural patterns for distributed machine learning in IoT systems, considering latency and response time requirements.

Findings

Communication latency impacts system response times.

Edge and cloud deployment strategies depend on application latency needs.

Recommendations for machine learning pattern selection in IoT systems.

Abstract

Number of IoT devices is constantly increasing which results in greater complexity of computations and high data velocity. One of the approach to process sensor data is dataflow programming. It enables the development of reactive software with short processing and rapid response times, especially when moved to the edge of the network. This is especially important in systems that utilize online machine learning algorithms to analyze ongoing processes such as those observed in Industry 4.0. In this paper, we show that organization of such systems depends on the entire processing stack, from the hardware layer all the way to the software layer, as well as on the required response times of the IoT system. We propose a flow processing stack for such systems along with the organizational machine learning architectural patterns that enable the possibility to spread the learning and inferencing on the edge and the cloud. In the paper, we analyse what latency is introduced by communication technologies used in the IoT for cloud connectivity and how they influence the response times of the system. Finally, we are providing recommendations which machine learning patterns should be used in the IoT systems depending on the application type.