Procedural Knowledge (PK) is knowing-how to perform some tasks. For industry workers, PK is the knowledge required to carry out a specific job, like correctly executing the safety procedure during maintenance interventions, or configuring an industrial system by following the necessary steps in the right order, or adopting practices and behaviours to optimise energy consumption in a plant. The challenge in PK management is that this kind of knowledge may be hard to explain and describe, oftentimes it is poorly digitalised, and, even when documented, it may be still difficult to access and retrieve by industry operators. The PERKS project supports the holistic governance of industrial PK in its entire life cycle, from elicitation to management and from access to exploitation. PERKS bases its solutions on AI (both symbolic and subsymbolic) and data technologies, by advancing and integrating existing methodologies and tools in terms of readiness, flexibility and user acceptance. The results are applied and tested in three industrial scenarios (white goods production plant, computer numerical control machines, microgrid testbed) providing different use cases in terms of PK complexity and industrial requirements. Besides AI and data, the third pillar of PERKS is people: the goal is to satisfy the concrete needs of industry workers, providing them with digital support to better and easier perform their tasks following a human-in-the-loop paradigm, and putting them at the centre according to Industry 5.0 approaches. The outcomes of PERKS are: a reference architecture for PK management; a set of modular, interoperable and complementary digital tools to be composed and customised to industrial requirements; specific integrated solutions to solve the challenges of the project use cases; a set of methodologies and best practices for broader application in other industrial settings, paving the way for a wider transferability across contexts and sectors.

The adoption of robots in lower volume, diverse environment is heavily constrained by the high integration and deployment complexity that overshadows the performance benefits of this technology. If robots are to become well accepted across the whole spectra of production industries, real evidence that they can operate in an open, modular and scalable way is needed. ODIN aspires to fill this gap by bringing technology from the latest ground breaking research in the fields of a) collaborating robots and human robot collaborative workplaces b) autonomous robotics and AI based task planning c) mobile robots and reconfigurable tooling, d) Digital Twins and Virtual Commissioning and e) Service Oriented Robotics Integration and Communication Architectures. To strengthen the EU production companies’ trust in utilizing advanced robotics, the vision of ODIN is: “to demonstrate that novel robot-based production systems are not only technically feasible, but also efficient and sustainable for immediate introduction at the shopfloor”. ODIN will achieve this vision through the implementation of Large Scale Pilots consisting of the following components: - Open Component (OC): A small footprint, small scale pilot instance allowing the development, integration and testing of cutting-edge technologies. - Digital Component (DC): A virtual instance of the pilot implementing an accurate Digital Twin representation that allows the commissioning, validation and control of the actual pilot - Industrial Component (IC): A full-scale instance of the pilot, integrating hardware (HW) and software (SW) modules from the Open and Digital components and operating under an actual production environment. - Networked Component (IC): An integration architecture with open interfaces allowing the communication of all robotics HW and control systems through safe and secure means. ODIN will demonstrate its result in 3 Large Scale Pilots in the automotive, white goods and aeronautic sectors.

Humans are at the center of knowledge-intensive manufacturing processes. They must be skilled and flexible to meet the requirements of their work environment. The training of new workers in these processes is time consuming and costly for companies. Industries, such as the Italian textile sector suffer from the shortage of skilled workers caused, e.g. by the demographic change. A second challenge for the manufacturing sector is the continuous competition through high quality products. COALA will address both challenges through the innovative design and development of a voice-first Digital Intelligent Assistant for the manufacturing sector. The COALA solution will base on the privacy-focused open assistant Mycroft. It integrates prescriptive quality analytics, AI system to support on-the-job training of new workers, and a novel explanation engine - the WHY engine. COALA will address AI ethics during design, deployment, and use of the new solution. Critical components for the adoption of the solution are a new didactic concept to reach workers about opportunities, challenges, and risks in human-AI collaboration, and a concurrent change management process. Three use cases (textile, white goods, liquid packaging) will evaluate the results in common manufacturing processes with significant economic relevance. COALA will contribute its results to the European AI community, e.g. via the AI4EU platform, and it will involve Digital Innovation Hubs to replicate its demonstrators for Europes first trustworthy digital assistant for the manufacturing industry. We expect to reduce the failure cost in manufacturing by 30-60% with the prescriptive quality analytics feature and the assisted worker training. For the change over time we expect a reduction of 15% to 30% by shortening the worker training time.

Despite the indisputable benefits of AI, humans typically have little visibility and knowledge on how AI systems make any decisions or predictions due to the so-called “black-box effect” in which many of the machine learning/deep learning algorithms are not able to be examined after their execution to understand specifically how and why a decision has been made. The inner workings of machine learning and deep learning are not exactly transparent, and as algorithms become more complicated, fears of undetected bias, mistakes, and miscomprehensions creeping into decision making, naturally grow among manufacturers and practically any stakeholder In this context, Explainable AI (XAI) is today an emerging field that aims to address how black box decisions of AI systems are made, inspecting and attempting to understand the steps and models involved in decision making to increase human trust. XMANAI aims at placing the indisputable power of Explainable AI at the service of manufacturing and human progress, carving out a “human-centric”, trustful approach that is respectful of European values and principles, and adopting the mentality that “our AI is only as good as we are”. XMANAI, demonstrated in 4 real-life manufacturing cases, will help the manufacturing value chain to shift towards the amplifying AI era by coupling (hybrid and graph) AI "glass box" models that are explainable to a "human-in-the-loop" and produce value-based explanations, with complex AI assets (data and models) management-sharing-security technologies to multiply the latent data value in a trusted manner, and targeted manufacturing apps to solve concrete manufacturing problems with high impact.

Several products embed different types of electronic components, and they are even more fundamental in some of the European strategic markets (e.g. automotive). However, reference producers come from extra-EU countries in the far-east side of the world (e.g. China and Taiwan). Trying to cope with all these challenges and the current semiconductors crisis, the European Commission (EC) published (and in some cases is still working on) specific EU strategies/directives for automotive, e-waste (e.g. Digital Product Passport) and, specifically, semiconductors (e.g. European Chips Act). However, trying to make these sectors more sustainable, circular and resilient, it is mandatory to boost both EoL strategies (e.g. sorting, reuse, remanufacturing and recycling) and intra-EU production through innovations and investments. The current international scenario represents a good chance to decouple the European economy from both natural resource depletion (e.g. Critical Raw Materials - CRMs) and dependency from extra-EU supplies of strategic products. In order to better prove what the benefits are of a joined circular/resilient use of secondary resources, the automotive and mass electronics sectors have been identified as the reference contexts for establishing a set of innovative solutions. To this aim, the CIRC-UITS project will focus on demonstrate the improvement to the circularity of automotive and mass electronics sectors by reuse of semiconductors from different sources, as well as support the reuse & remanufacturing of semiconductors into new (high added-value) components and products in these sectors.