Periodic Reporting for period 1 - ALICIA (Assembly Lines In CIrculAtion – smart digital tools for the sustainable, human-centric and resilient use of production resources)
Período documentado: 2023-01-01 hasta 2024-06-30
The ALICIA project addresses the issue of premature obsolescence in manufacturing assets, such as robotic arms and conveyor belts, with 60%-70% being discarded prematurely. This waste undermines the EU's circular economy goals. ALICIA aims to establish a Circular Manufacturing Ecosystem (CME) to maximize resource utility by reusing and recirculating production assets across factories, enhancing sustainability and resilience against supply chain disruptions.
ALICIA will implement and demonstrate digital tools at Continental and Comau, including a machine-readable ontology, AI matchmaking engine, Plug & Produce middleware, and Digital Shadow/Digital Twin technologies. These innovations aim to design, and deploy second-hand production lines 40% faster, reduce material consumption by 70%, and achieve up to 100% asset reuse. The project also introduces business models for purchasing and operating second-hand equipment to enhance sustainability and productivity.
ALICIA will implement and demonstrate digital tools at Continental and Comau, including a machine-readable ontology, AI matchmaking engine, Plug & Produce middleware, and Digital Shadow/Digital Twin technologies. These innovations aim to design, and deploy second-hand production lines 40% faster, reduce material consumption by 70%, and achieve up to 100% asset reuse. The project also introduces business models for purchasing and operating second-hand equipment to enhance sustainability and productivity.
In Work Package 1 (WP1), led by TUG, the primary objectives were to conduct an ecosystem analysis, develop technical specifications, and integrate the framework for capturing factory owner requirements with ALICIA tools. These objectives were successfully achieved through the completion of a use case exploration, the identification and documentation of requirements, the development of architecture for the whole ALICIA platform, and the implementation of the framework.
For Work Package 2 (WP2), under the leadership of IMT, the goals included developing AI engine specifications, preparing relevant datasets, and building and fine-tuning the AI engine. Significant progress was made in this area with the definition of sub-requirements, the generation of necessary datasets, and the successful construction and training of the AI engine.
In Work Package 3 (WP3), led by LMS, the focus was on specifying APIs, implementing the Digital Shadow (DS) API, and conducting tests and validation. The project team accomplished these tasks by defining the necessary requirements, developing and implementing the DS API, testing it with simulated data, and integrating it with middleware to transform it to a Digital Twin (DT) for real-time validation.
Work Package 4 (WP4), led by ECI, aimed to translate system requirements into action, develop Asset Administration Shells (AAS), and create adaptors. The team effectively met these objectives by defining the system requirements, developing AAS models, and creating and testing the necessary adaptors and middleware.
Under the guidance of INTRA, Work Package 5 (WP5) focused on the design integration, extension of services, and implementation of connectors for the marketplace. The team succeeded in defining the requirements, implementing the connectors, and integrating the AI engine with the marketplace, thus enhancing its functionality. The goal is to create an online marketplace for second-hand production resources, which uniquely integrates DS/DT data sharing and the AI matchmaking engine. This marketplace not only facilitates the efficient exchange of second-hand resources but also supports sustainable practices in the industry.
Work Package 6 (WP6), led by COMAU, is dedicated to testing tools, scaling up for demonstrations, and evaluating and defining post-project activities. This work package included the validation of the ontology and the AI engine, the execution of small-scale and industrial demonstrations evaluation of the project results..
Work Package 7 (WP7), led by YAGHMA, aimed to ensure ethical and sustainable innovation and prepare for the commercial uptake of project outcomes. The team conducted ethics audits and life cycle assessments (LCA), developed business models, designed a human-centric training kit, and researched relevant standards to support the project's objectives.
Finally, in Work Package 8 (WP8), led by TUM, the focus was on project management and communication. The team successfully ensured the overall implementation of the project, managed data effectively, and carried out comprehensive dissemination activities to communicate the project's results.
For Work Package 2 (WP2), under the leadership of IMT, the goals included developing AI engine specifications, preparing relevant datasets, and building and fine-tuning the AI engine. Significant progress was made in this area with the definition of sub-requirements, the generation of necessary datasets, and the successful construction and training of the AI engine.
In Work Package 3 (WP3), led by LMS, the focus was on specifying APIs, implementing the Digital Shadow (DS) API, and conducting tests and validation. The project team accomplished these tasks by defining the necessary requirements, developing and implementing the DS API, testing it with simulated data, and integrating it with middleware to transform it to a Digital Twin (DT) for real-time validation.
Work Package 4 (WP4), led by ECI, aimed to translate system requirements into action, develop Asset Administration Shells (AAS), and create adaptors. The team effectively met these objectives by defining the system requirements, developing AAS models, and creating and testing the necessary adaptors and middleware.
Under the guidance of INTRA, Work Package 5 (WP5) focused on the design integration, extension of services, and implementation of connectors for the marketplace. The team succeeded in defining the requirements, implementing the connectors, and integrating the AI engine with the marketplace, thus enhancing its functionality. The goal is to create an online marketplace for second-hand production resources, which uniquely integrates DS/DT data sharing and the AI matchmaking engine. This marketplace not only facilitates the efficient exchange of second-hand resources but also supports sustainable practices in the industry.
Work Package 6 (WP6), led by COMAU, is dedicated to testing tools, scaling up for demonstrations, and evaluating and defining post-project activities. This work package included the validation of the ontology and the AI engine, the execution of small-scale and industrial demonstrations evaluation of the project results..
Work Package 7 (WP7), led by YAGHMA, aimed to ensure ethical and sustainable innovation and prepare for the commercial uptake of project outcomes. The team conducted ethics audits and life cycle assessments (LCA), developed business models, designed a human-centric training kit, and researched relevant standards to support the project's objectives.
Finally, in Work Package 8 (WP8), led by TUM, the focus was on project management and communication. The team successfully ensured the overall implementation of the project, managed data effectively, and carried out comprehensive dissemination activities to communicate the project's results.
The project has achieved significant advancements beyond the current state of the art in several key areas. Firstly, a machine-readable ontology and ecosystem analysis tool was developed. This tool extends the MaRCO ontology by incorporating, e.g. hazardous and rare materials and human skills, which facilitates the integration of human-centric and resilience factors into the system.
The AI Matchmaking Engine represents another major advancement. This engine utilizes reinforcement learning and mathematical programming techniques to optimize the reuse of resources while considering human-centric variables, thereby enhancing the decision-making process in manufacturing environments.
In the realm of Digital Shadow (DS) and Digital Twin (DT) technologies, the project has integrated real-time data to enhance these technologies' capabilities, offering more accurate and dynamic simulations of physical systems.
The development of Plug & Produce middleware is another key result, featuring adaptors that ensure the interoperability of new systems with legacy equipment. This innovation facilitates the seamless integration of various manufacturing systems and technologies.
Overall, these innovations have a significant impact by increasing manufacturing efficiency, reducing costs, improving working conditions, and minimizing environmental impact.
The AI Matchmaking Engine represents another major advancement. This engine utilizes reinforcement learning and mathematical programming techniques to optimize the reuse of resources while considering human-centric variables, thereby enhancing the decision-making process in manufacturing environments.
In the realm of Digital Shadow (DS) and Digital Twin (DT) technologies, the project has integrated real-time data to enhance these technologies' capabilities, offering more accurate and dynamic simulations of physical systems.
The development of Plug & Produce middleware is another key result, featuring adaptors that ensure the interoperability of new systems with legacy equipment. This innovation facilitates the seamless integration of various manufacturing systems and technologies.
Overall, these innovations have a significant impact by increasing manufacturing efficiency, reducing costs, improving working conditions, and minimizing environmental impact.