Scalable Fleet Monitoring and Visualization for Smart Machine Maintenance and Industrial IoT Applications.

Area of Science:

• Industrial Internet of Things (IIoT)
• Smart Manufacturing
• Predictive Maintenance

Background:

• Widespread adoption of smart machine maintenance is hindered by Industrial Internet of Things (IIoT) challenges in robustness, scalability, and security.
• Effective predictive maintenance relies on well-trained machine learning algorithms that require substantial volumes of reliable data.

Purpose of the Study:

• To address the challenges in IIoT for smart machine maintenance.
• To present an open test and research platform, the Smart Maintenance Living Lab, for validating IIoT applications.

Main Methods:

• Development of a Smart Maintenance Living Lab featuring a fleet of drivetrain systems for accelerated bearing tests.
• Implementation of a scalable IoT middleware cloud platform for data ingestion and persistence.
• Creation of a dynamic dashboard for fleet monitoring and visualization.

Main Results:

• Demonstration of the feasibility of IIoT applications for smart machine maintenance through component validation.
• Generation of benchmark data for enhancing machine learning algorithms.
• Provision of insights into designing, implementing, and validating robust, scalable, and secure IIoT architectures.

Conclusions:

• The Smart Maintenance Living Lab reduces industry reticence towards adopting IIoT technologies for smart maintenance.
• The platform facilitates the improvement of machine learning algorithms and validates IIoT system architectures.
• The study confirms the viability of IIoT for mission-critical industrial operations.