AI-PROFICIENT: Objectives & expected results
As mentioned in previous posts, AI-PROFICIENT will develop a technical and business ecosystem to demonstrate the potential for improved performance in production plants, by bringing the advanced AI technologies to production lines and facilitating the cooperation between humans and machines.
To this aim, our project proposes an evolution from hierarchical and reactive decision making for plant automation towards self-learning and proactive control strategies that take full advantage of integration of advanced AI technologies with production plants.
Which are AI-PROFICIENT’s objectives?
Goal #1: Integrating existing and emerging AI technologies to create the AI-PROFICIENT platform for digitalized production plants, in order to enable agile production processes and improved operation planning and execution, while increasing the Overall Production Efficiency (OPE).
Goal #2: Piloting the AI-PROFICIENT solution in 3 production plants which operate in different manufacturing domains, under different use case scenarios, involving the AI-enabled predictive fault detection and diagnostics, and proactive maintenance.
Goal #3: Identifying the effective means for human-machine collaboration, while respecting the privacy, safety and security requirements and following the ethical principles, to enable both the AI decision-making explainability and transparency, as well as the reinforcement mechanisms based on the human knowledge and feedback to improve the trustworthiness of AI in manufacturing domain.
Which are AI-PROFICIENT’s expected results?
The AI-PROFICIENT enabling technologies and concepts of AI-PROFICIENT solution are the following:
- Smart components for embedded AI at system edge
- IIoT for smart component integration and interoperability
- AI prognostics for system degradation and health state assessment
- AI enabled decision-making for quality assurance
- Semantic lifting and model agnostic techniques for XAI
- Hybrid digital twins and process modelling
- Generative optimization of production processes (human in the loop)
- Role-specific visualization for transparent AI decision support
- ‘Ethics by Design’ Approach
We must point out that in order to demonstrate the effectiveness and flexibility of the overall concept, the AI-PROFICIENT solution will be fully integrated, deployed and validated in three different operation environments provided by two manufacturing enterprises, CONTINENTAL and INEOS.
Last but not least, in collaboration with their plant operators, AI-PROFICIENT will ensure the end-user engagement throughout the project lifetime, considering requirements specification, deployment and validation, as well as development of recommendations for ethical principles for trustworthy AI in manufacturing domain.