AI-PROFICIENT

AI-PROFICIENT develops a technical and business ecosystem to demonstrate the potential for improved
performance in production plants by bringing advanced AI technologies to production lines and facilitating the cooperation between humans and machines. In this regard, AI-PROFICIENT proposes an evolution from hierarchical and reactive decision-making for plant automation towards self-learning and proactive control strategies that take full advantage of the integration of advanced AI technologies with production plants. 

• Objective 1: Integrating existing and emerging AI technologies to create the AI-PROFICIENT platform for digitalised production plants, leveraging the platform AI services and local intelligence of smart components at the system edge to enable agile production processes and improved operational planning and execution while increasing the Overall Production Efficiency (OPE).
• Objective 2: Piloting the AI-PROFICIENT solution in 3 production plants which operate in different manufacturing domains under different use case scenarios, involving AI-enabled predictive fault detection and diagnostics, and proactive maintenance, demonstrating its potential to improve the quality of products and processes.
• Objective 3: Identifying the effective means for human-machine collaboration while respecting the privacy, safety and security requirements and following the ethical principles in order to enable the AI decision-making explainability and transparency, as well as the reinforcement mechanisms based on human knowledge and feedback to improve the trustworthiness of AI in the manufacturing domain.

Reason for applying to HSbooster.eu services

• Develop a standardisation gap analysis based on existing industrial maintenance standards but addressing the scope of AI-PROFICIENT, i.e AI application to fault detection, prognostics and condition-based maintenance.
• Select one or two standards for which to consider submission of CWAs or equivalent.
• Strengthen contacts with appropriate standardisation groups, which we have yet to select.

Main Standardisation Interests

Objective:

• Support the standardisation initiatives in relation to AI applications for manufacturing, as well
• as system integration and interoperability
• Dedicated T7.6 on Standardization. It will closely collaborate with WP4 and WP5 in the design of the M2H/H2M interfaces for the connected worker and M2M interoperability mechanisms and communication protocols for the sake of alignment with contemporary standardisation efforts. Task activities will consider different aspects of data interfacing and integration, namely: alignment with the RAMI 4.0 architecture; interfacing with respect to CEN TC 319 maintenance standards; interoperability with production quality control data (QIF); integration with plant management systems (MOM/CMMS/QMS); and M2M connectivity (in particular OPC UA). Specific attention will be paid to the alignment with activities of the CEN/CENELEC Focus Group on AI and building upon the results of the Joint MSP/DEI Working Group.

Current Status:

• A preliminary set of pertinent standards has been identified.
• A map of already achieved deliverables to related standards has been created;
• Possible AI-PROFICIENT contributions to the targeted standardisation items in the CEN-CENELEC Roadmap have been identified;
• We performed an internal evaluation of the AI-PROFICIENT standardisation efforts against the project-specific level recommendations in the EU Draft Code of practice on standardisation for researchers.