Artificial intelligence (AI) claims to be a game-changer for manufacturers at every value chain stage. Manufacturers must increase the potentiality of their equipment, minimize waste, shorten cycle times to compete, and increase yields. Direct automation, increased efficiency, reduced downtime, predictive maintenance, overall production, improved safety, quality control, lower operational costs, and faster decision-making are a few of the advantages accessible for businesses that master the deployment of AI throughout their operations. Novel sensor data processing competence and big data, cloud, ML, AI, and edge technologies enable a shift toward more proactive management of processes, equipment, products, and factories. Applying AI and ML and evaluating real-time data from machine sensors makes it desirable to forecast analytical events and take preventative operations to evade complications. Smart factories can examine sensor data streaming and render data on the strength of the equipment and systems using business rules and ML models. Analytics are used for a comprehensive audit trail, version control, authenticated regulatory reporting, and electronic signatures to document changes to analytic processes and reports to automate approvals and workflow.
Sensor data from allied industrial equipment is a beneficial resource in intelligent manufacturing. As a result, the data-acquisition process relies heavily on the production line and factories. These massive data-management systems depend on good data lineage and data cataloging to manage the trail of all available data and flows while making it accessible to multiple users. Manufacturers should plot their primary data objects, machinery, and items such as production facilities and the associated data sources velocities to understand the data volumes and variety they would be dealing with predominantly. They must also develop data quality metrics and monitor them regularly to promote awareness of their importance, which might be difficult when using AI. Manufacturers should begin using a “functional reference architecture,” which outlines the tools they’ll need to collect, store, manage, and analyze data and the analytics and visualization software they’ll need. They can utilize this functional architecture to outline their requirements to evaluate the most appropriate technologies on the market and the technical and infrastructural configuration they want to use.
Some accessible advancements, such as HDFS and Spark, have become de facto big data and AI standards and have been integrated into most commercial AI and big data platform services. Scalable distributed data processing and the construction of complex ML models are both conceivable with these technologies. Manufacturers should pay special attention to analytics skills and “time series” data processing to handle sensor data streams in a production setting. Topics for the special section include, but are not limited to, the following:
- Convergence of ML and DL for real-time monitoring for smart sensing and production systems
- AI-based data analytics for predictive maintenance through smart cybersecurity with intelligent interaction
- Enhanced memory-efficient and computationally AI algorithms for smart sensing and production systems
- AI-enabled data analytics for data acquisition and storage in industrial interaction systems
- Digital twins for smart sensing and intelligent interaction in production systems
- AI-enabled data analytics for multi-modal sensing in smart manufacturing
- Convergence of ML with AI-enabled automation for verification and validation in intelligent production systems
- Decentralized edge computing for AI-enabled big data analytics for smart manufacturing
- AI-enabled big data analytics for planning and decision-making in industrial automation
- Industrial IoT-enabled data-driven predictive analytics for smart sensing
Manuscript submissions due: 30 September 2022
Reviews completed: 17 January 2023
Final manuscripts due: 12 March 2023
For author information and guidelines on submission criteria, please visit the OJ-CS Author Information page. Please submit papers through the ScholarOne system, and be sure to select the special-section name. Manuscripts should not be published or currently submitted for publication elsewhere. Please submit only full papers intended for review, not abstracts, to the ScholarOne portal.
Kyungpook National University, South Korea
La Trobe University, Melbourne, Australia
Università Degli Studi di Milano, Italy