Periodic Reporting for period 1 - InterQ (Interlinked Process, Product and Data Quality framework for Zero-Defects Manufacturing) Période du rapport: 2020-11-01 au 2022-04-30 Résumé du contexte et des objectifs généraux du projet In current factories, large amount of data is generated from the existing machine sensors. However, that information is not directly useful for process planners to improve the manufactured products due to:• Large amounts of data with little information. Companies are storing data without relevant information.• Lack of information about the product quality. The quality of the product after an intermediate process is not measured and included in the monitoring platform. • Unreliable data. The actual monitoring systems don’t check the quality of the data.• Lack of data exchange between the supply chain actors. Data silos are largely reducing the potential use of the data. The main objective of InterQ project is to measure, predict and control the quality of the manufactured products, manufacturing process and gathered data to assure zero defect manufacturing by means of AI driven tools powered with meaningful and reliable data. InterQ project wants to boost the quality of manufacturing operations and manufactured products linking the process, product and data quality across the supply chain using newly developed distributed ledger technologies to break the current data silos.InterQ will use direct and indirect quality measurements to optimize both the quality of the manufactured products. Indirect measurements made on the manufacturing system will allow predicting the product quality by means of virtual sensors and digital twins. In parallel, new sensors will be developed to measure process parameters close to the tool centre point (TCP) and the product quality. Specific attention will be paid to the data reliability with continuous verification of the measurement quality in real time as well as long term level. Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent Three industrial cases related to aerospace, energy and automotive sectors have been defined providing information regarding the part, machine tool, process, product quality control and quality issues. In the first year the activity of the project has been concentrated in the development of the new systems for process monitoring (InterQ-Process), final product quality measurement (InterQ-Product) and data quality measurement (Inter Q- Data). This developments will provide the required information to create the PPD Quality Hallmark after each production step.In the case of process monitoring (InterQ-Process), the main idea is to measure or estimate process parameters close to the TCP to obtain more meaningful data. This way, new force sensors for grinding and broaching operations have been developed. Complementarily, six virtual sensors to estimate different parameters of the process have been developed. The goal of InterQ-Product is to measure/estimate intermediate and final product quality using new sensors and digital twins. New sensors to measure automatically and digitally the surface topography and integrity have been developed. In addition, three AI driven digital twins have been built for the use cases.The activities of InterQ-Data are focused on data quality verification considering edge level and historical data verification. At the edge level, algorithms to detect real-time data quality problems have been developed including algorithms to detect broken sensors, excessive noise and signal processing problems. Complementarily, the historical data validation consists on verifying normal production patterns. Finally, an additional machine learning application called Erdre has been developed to perform erroneous data repair. In parallel of this activities, the InterQ - Trusted Framework (TF) has been designed based on permissioned blockchain. Hyperledger Fabric was selected as the blockchain platform that constitutes the backbone of the project. This is a decentralized technical infrastructure that quality-related information (PPD Hallmark) to be shared and leveraged within supply chain. Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent) The main innovations are the next ones:• A new generation of sensors and actuator located close to the tool centre point and virtual sensors has been developed. • Development of six different inspection systems giving information about key variables for product quality that cannot be measured directly. • Predicting the quality of the manufactured products with digital twins that will allow implementing closed loop feedback control to improve the overall manufacturing process. • InterQ has developed a novel application to check on-line the quality of the data provided by sensors • A machine learning pipeline to perform erroneous data repair called Erdre has been developed and supposes a step beyond the state of the art.• The project will suppose an important milestone in the application of the blockchain technology in the quality management in industrial environment. The main project output can be condensed in five modules with new sensors and applications for quality control. Inside each there are a set of results that can be exploited individually.• InterQ-Process: Process monitoring based on the measurements close to the TCT. It will be composed by collocated mechatronic devices, new sensors and AI driven virtual sensors. The module will collect meaningful process data able to estimate the status of TCT, possible machine stops and tool wear.• InterQ-Product: The main idea is to digitalize and automatize the data collection of the final quality of a part after one manufacturing step by means by new sensors and a new generation of AI driven digital twins.• InterQ-Data: Thanks to this module the data quality is measured in motion at edge level and long-term by means of historical pattern analysis.• InterQ-ZeroDefect: This module will offer a Quality hallmark dashboard for human centred optimization, process quality optimization tools at machine level and AI driven applications for production optimization at production line level.• InterQ-TrustedFramework: will provide a data sharing platform based on data integrity, traceability and trustworthiness. This framework is based on Distributed ledger technologies and offers a tool for the management of PPD Quality hallmark in the supply chain.The next points summarize the expected impact in the production lines:• Increased equipment productivity through rapid error localisation: InterQ will achieve this objective with the new sensors that measure the quality in process. • Reduction of ramp-up time using smart sensors/actuators and existing production data sets. Existing quality information stored in the distributed ledger will allow fixing potential quality issues before the actual beginning of the production.• AI-driven instrumentation stimulating the transformation towards smart and fast processes leading to decreased time-to-market. The project has developed several AI driven applications to predict process parameters, final part quality and data repair.• Significant increase in quality of manufactured products leading to a reduction of scrap: The in-process measurement of the quality will bring the major impact to achieve a zero-defect production.The impact of the project will be basically manifested mainly in the end of the activities. In the first half of the project the consortium has focused in the development of different systems and technologies, and therefore the first beneficiaries are the different SMEs of the project and the machine builders. WP6 - InterQ-TrustedFramework. Information flow WP2 – InterQ-Process. Fingerprint result visualization at SmartFactoryCell. Independent Fingerprint WP5 - InterQ-ZeroDefect. Tasks for the identification of the origin of surface topography inaccuraci WP2 – InterQ-Process. Grinding machine's process monitoring capabilities WP3 – InterQ-Product. Pixelized surface of the rotating tool to represent cutter workpiece engagemen WP4 - InterQ-Data. Extraction of the data quality for the PPD Quality Hallmark in casing roughing op WP2 – InterQ-Process. Schematic diagram of the monitoring system: points mark accelerometers in the WP8 - InterQ Dissemination activities InterQ Project PPD Quality hallmark
