What is it about?

The way that companies produce goods is changing in the digital age. Production lines need to blend machine learning, data analytics, as well as human elements. This is known as Industry 4.0 and will improve the efficiency of production. But such complex systems consist of many connected parts and subsystems. These make them vulnerable to cyberattacks. This paper describes a “visual analytics” framework for system monitoring. It uses visual interfaces that keep humans aware of the system status. The interfaces cover multiple variables and real time predictions. This allows for quick decision making. The framework can project potential threats. It can predict how long it will take for the system to feel adverse effects. The framework can also detect malicious attacks. Finally, it can determine how long it will take for system failure. The authors of the study successfully tested their approach on data from the aviation industry.

Featured Image

Why is it important?

The production process is one with many moving parts. Some of these parts are "weak links," that can become the focus of a cyberattack. Increasing the number of digital components in the production process also increases the number of weak links. This makes it essential to have a cyber security framework that is robust against such attacks. The framework proposed in this study can quickly assess threats and their severity. It also keeps humans apprised of the state of the system. The machine learning elements in this framework are explainable. This means that users can easily understand the system and its processes. KEY TAKEAWAY: Robust cyber security measures are essential for Industry 4.0 systems. The framework proposed in this paper can be used to monitor and secure such systems. It is robust and human users can easily understand it.

Perspectives

Writing this article was a great pleasure as I got to discuss and expand my knowledge in not just one field, but in broad areas, from visualization to manufacturing systems and beyond. I hope you find this article a broad view of the ways we can combine human and machine intelligence to apply to various domains, in this case, the manufacturing automation system.

Huyen N. Nguyen
Texas Tech University System

Read the Original

This page is a summary of: Visualization and Explainable Machine Learning for Efficient Manufacturing and System Operations, Smart and Sustainable Manufacturing Systems, February 2019, ASTM International,
DOI: 10.1520/ssms20190029.
You can read the full text:

Read

Contributors

The following have contributed to this page