Periodic Reporting for period 2 - i4Q (Industrial Data Services for Quality Control in Smart Manufacturing)
Okres sprawozdawczy: 2022-07-01 do 2024-05-31
The i4Q Project aimed to deliver IoT-based Reliable Industrial Data Services (RIDS), a comprehensive suite comprising 22 i4Q Solutions, which were designed to manage vast amounts of industrial data from low-cost, efficient, and compact interconnected factory devices, supporting real-time monitoring and control of manufacturing processes. The i4Q Framework ensured data reliability through functions grouped into five fundamental capabilities within the data cycle: sensing, communication, computing infrastructure, storage, and analysis and optimization. The i4Q RIDS featured simulation and optimization tools for continuous process qualification, quality diagnosis, reconfiguration, and certification of manufacturing lines, ultimately ensuring high manufacturing efficiency and fostering a zero-defect manufacturing approach.
With i4Q RIDS, factories can effectively handle large volumes of data, attaining high levels of accuracy, precision, and traceability. This data is used for analysis, prediction, and optimization of both process quality and product quality in manufacturing, promoting a zero-defect manufacturing approach. i4Q Solutions efficiently gather raw industrial data using cost-effective instruments and advanced communication protocols, ensuring data accuracy, precision, reliable traceability, and time-stamped data integrity through distributed ledger technology. i4Q also provided tools for continuous process qualification, quality diagnosis, reconfiguration, and certification of manufacturing lines, guaranteeing high manufacturing efficiency and optimal quality.
The i4Q RIDS were demonstrated in 6 Uses Cases from relevant industrial sectors and representing two different levels of the manufacturing process: machine tool providers and production companies. i4Q pan-European consortium entails Industrial partners: WHIRLPOOL, BIESSE, FACTOR, RIASTONE, FARPLAS and FIDIA. Implementers: TIAG, CESI and AIMPLAS. Technology Providers: IBM, ENGINEERING, ITI, KNOWLEDGEBIZ, EXOS. R&D partners: CERTH, IKERLAN,BIBA, UPV, TUBERLIN, UNINOVA. Specialist partners: FUNDINGBOX, INTEROP-VLAB, DIN, LIF.
With i4Q RIDS, factories can effectively handle large volumes of data, attaining high levels of accuracy, precision, and traceability. This data is used for analysis, prediction, and optimization of both process quality and product quality in manufacturing, promoting a zero-defect manufacturing approach. i4Q Solutions efficiently gather raw industrial data using cost-effective instruments and advanced communication protocols, ensuring data accuracy, precision, reliable traceability, and time-stamped data integrity through distributed ledger technology. i4Q also provided tools for continuous process qualification, quality diagnosis, reconfiguration, and certification of manufacturing lines, guaranteeing high manufacturing efficiency and optimal quality.
The i4Q RIDS were demonstrated in 6 Uses Cases from relevant industrial sectors and representing two different levels of the manufacturing process: machine tool providers and production companies. i4Q pan-European consortium entails Industrial partners: WHIRLPOOL, BIESSE, FACTOR, RIASTONE, FARPLAS and FIDIA. Implementers: TIAG, CESI and AIMPLAS. Technology Providers: IBM, ENGINEERING, ITI, KNOWLEDGEBIZ, EXOS. R&D partners: CERTH, IKERLAN,BIBA, UPV, TUBERLIN, UNINOVA. Specialist partners: FUNDINGBOX, INTEROP-VLAB, DIN, LIF.
From M1 to M18, the project developed a conceptual vision, establishing the state of the art in manufacturing quality technologies. A regulation and trustworthy data management system was defined, along with the specifications guiding the creation of i4Q solutions and key performance indicators. By M9, the i4Q framework was designed with clear and detailed reference architecture, defining viewpoints, KPIs, requirements, and functional specifications. A thorough analysis of various ontologies and data models to support interoperability and data exchange among tools, platforms, and organizations was also conducted. Multiple perspectives, such as business, usage, functional, and implementation viewpoints, were used to describe the reference framework. By M18, the first version of the i4Q solutions was released, outlining the principles, development, design, and communication among solutions. Deliverables and demos containing all necessary information for the development and functioning of these solutions were also published. Subsequently, the experimental base for the i4Q Solutions was initiated. Methodologies were applied in the six pilots participating in the project, along with software tools and technologies developed to meet specified and real-world scenarios and requirements. The exploitation potential of these solutions was also investigated. The first pipelines for the six pilots were prepared, with a set of solutions collaborating in these pipelines to collect data, process it, and provide the required results for each industry. Lastly, a message broker was developed to facilitate communication between the solutions.
From M21 to M36, the i4Q solutions were fully developed and their integration and deployment in the industry environment tool place. Specific i4Q solutions were deployed in each one of the pilots, based on the scenarios and pilots’ needs. During the extension of the project, from M36 to M41, a final tuning of the tools’ integration in the pilots’ sites took place, where some extra features of the tools were added for some pilots, after pilots’ suggestions for the improvement of the tools and the easiest user interaction with machine operators.
During this reporting period, many exploitation activities tool place. A series of tasks have been carried out (e.g. exploitation workshops, Value proposition, buyer persona, business model canvas, an extensive market analysis, customer discovery, go-to market strategy), aimed at gathering and analyzing information to prepare the go-to-market plan for the i4Q solutions. Solutions were also prepared for their exploitation in marketplaces, as well as a detailed plan for the creation of a new start-up was developed.
In conclusion, from M1 to M41, the i4Q project actively participated in various events, conferences, and publications, maintaining a robust dissemination activity with numerous engaging papers and presentations. Additional dissemination materials, providing updates on project progress, i4Q events, workshops, publications, and other exploitable results, include the project website and links to Twitter, LinkedIn, and YouTube, all of which are publicly accessible.
From M21 to M36, the i4Q solutions were fully developed and their integration and deployment in the industry environment tool place. Specific i4Q solutions were deployed in each one of the pilots, based on the scenarios and pilots’ needs. During the extension of the project, from M36 to M41, a final tuning of the tools’ integration in the pilots’ sites took place, where some extra features of the tools were added for some pilots, after pilots’ suggestions for the improvement of the tools and the easiest user interaction with machine operators.
During this reporting period, many exploitation activities tool place. A series of tasks have been carried out (e.g. exploitation workshops, Value proposition, buyer persona, business model canvas, an extensive market analysis, customer discovery, go-to market strategy), aimed at gathering and analyzing information to prepare the go-to-market plan for the i4Q solutions. Solutions were also prepared for their exploitation in marketplaces, as well as a detailed plan for the creation of a new start-up was developed.
In conclusion, from M1 to M41, the i4Q project actively participated in various events, conferences, and publications, maintaining a robust dissemination activity with numerous engaging papers and presentations. Additional dissemination materials, providing updates on project progress, i4Q events, workshops, publications, and other exploitable results, include the project website and links to Twitter, LinkedIn, and YouTube, all of which are publicly accessible.
i4Q solutions:
• tackled the analysis of manufacturing data by combining simulation and real data in the form of digital twins, while employing data fusion techniques and supporting the distribution, deployment, and monitoring of AI analysis models in real manufacturing environments with the final aim of increasing the industrial equipment productivity through real-time error localisation. Error localization is accomplished through the application of explainable AI techniques, which identify the source of the problem once a failure has been detected in the machine.
• included manufacturing process qualification and reconfiguration as an essential step during ramp-up and after reconfiguration of production processes and ensuring good manufacturing practice and final product quality through adequate control over processes, data collection and statistical procedures for evaluation of process stability and process performance.
• decreased the time-to-market of new products or variants, through the inclusion of Digital Twins using simulation and optimisation strategies for rapid line reconfiguration considering the intelligibility of the needed upgrades, ensuring that non-simulation experts may also exploit the prescriptive analyses.
• potentiated the role of AI in manufacturing for taking actions and decisions, identifying defects in products in a manufacturing lines, identifying hazards, or tuning machines based on dynamically changing conditions, connecting devices, gathering, collecting, and storing data for being analysed with the aim of help on taking the pertinent decisions and actions to increase the quality of manufactured products leading to zero-defects and the elimination of scrap.
• tackled the analysis of manufacturing data by combining simulation and real data in the form of digital twins, while employing data fusion techniques and supporting the distribution, deployment, and monitoring of AI analysis models in real manufacturing environments with the final aim of increasing the industrial equipment productivity through real-time error localisation. Error localization is accomplished through the application of explainable AI techniques, which identify the source of the problem once a failure has been detected in the machine.
• included manufacturing process qualification and reconfiguration as an essential step during ramp-up and after reconfiguration of production processes and ensuring good manufacturing practice and final product quality through adequate control over processes, data collection and statistical procedures for evaluation of process stability and process performance.
• decreased the time-to-market of new products or variants, through the inclusion of Digital Twins using simulation and optimisation strategies for rapid line reconfiguration considering the intelligibility of the needed upgrades, ensuring that non-simulation experts may also exploit the prescriptive analyses.
• potentiated the role of AI in manufacturing for taking actions and decisions, identifying defects in products in a manufacturing lines, identifying hazards, or tuning machines based on dynamically changing conditions, connecting devices, gathering, collecting, and storing data for being analysed with the aim of help on taking the pertinent decisions and actions to increase the quality of manufactured products leading to zero-defects and the elimination of scrap.