An Analytics Environment Architecture for Industrial Cyber-Physical Systems Big Data Solutions.

Area of Science:

• Computer Science
• Industrial Engineering
• Data Science

Background:

• Industrial processes face challenges in managing and analyzing large datasets.
• Integrating Big Data technologies into industrial cyber-physical systems (iCPS) requires a robust architectural framework.
• Existing solutions may not adequately address the complexities of data integration from Industrial Internet of Things (IIoT) environments.

Purpose of the Study:

• To propose a reference architecture for industrial Big Data analytics systems within iCPS.
• To support the design of Big Data solutions for data modeling, predictive analysis, KPI inference, and real-time analytics in iCPS.
• To facilitate the integration of IIoT, communication networks, and cloud platforms within the iCPS context.

Main Methods:

• An attribute-driven design (ADD) approach was employed for architectural design.
• Requirements were gathered from smart production planning, manufacturing process monitoring, and maintenance, repair, and overhaul (MRO) scenarios.
• The proposed architecture addresses data management drivers and novel Big Data analytics techniques.

Main Results:

• A reference architecture for Big Data analytics in iCPS is presented.
• The architecture effectively integrates IIoT environments, communication systems, and cloud computing.
• A fault diagnosis case study demonstrates the architecture's applicability in meeting functional and quality requirements.

Conclusions:

• The proposed architecture provides a foundational framework for industrial Big Data analytics in iCPS.
• Data is recognized as a critical asset in iCPS, necessitating advanced management and analytics.
• The architecture enables enhanced data-driven decision-making and operational efficiency in industrial settings.