Periodic Reporting for period 3 - Z-Fact0r (Zero-defect manufacturing strategies towards on-line production management for European factories)
Reporting period: 2019-04-01 to 2020-03-31
The z-fact0r solution comprises five modules that are to be used in multi stage production systems with the following targets:
i. The early detection of the defect (Z-Detect)
ii. The prediction of defect generation (Z-Predict)
iii. The prevention of defect generation by recalibrating the production line (multi stage), as well as defect propagation in later stages of the production (Z-Prevent)
iv. The reworking of the product, using additive and subtractive manufacturing techniques (Z-Repair) and
v. The management of the aforementioned strategies through event modelling, KPI (key performance indicators) monitoring and real time decision support(Z-manage)
The Z-Fact0r is expected to have a strong social character in that it aims to achieve zero defect manufacturing, therefore contributing to:
i. Securing a qualitative and sustainable manufacturing industry by reducing the production cost -including the energy consumption- per good and by increasing customer satisfaction.
ii. Contributing to the reduction of unemployment in Europe, by helping European Industries become more competitive.
iii. Contributing to environment protection by reducing material and energy waste.
WP2 has been completed with all the deliverables submitted. In this WP2, a laser scanning portable solution was developed for rapid defect detection, an effective integration of defect detection results generated by laser scanning solutions within processing algorithm for machine/process defect detection was implement, a robotic deburring cell for repairing defected parts was developed, dispensing techniques for product reworking were also developed and a Knowledge based Decision support system for managing all the defect and action events generated by the system was finally implemented.
WP3 has been almost completed with the submission of most of the related deliverables. Among the most important technical achievements of the work package is the definition of a sensors’ network for each end-user and its integration under of a common middleware environment that communicates with all technical modules.
In WP4, BRUNEL completed Z-PREDICT through installation of Event modeller algorithm.
The technical development of WP5 proceeds on track with the submitted deliverables D5.1 and D5.2 (draft version). The consortium has developed the Integration Incremental Strategy and integrated the various software application into a complex distributed software system. We set a number of integration scenarios, that they were executed in a regression manner to make sure that any change has not broken the so far system. Currently, by following validation scenarios, we are executing the software verification and validation process in order to verify that we build the system right (verification) and that we build the right system (validation).
In WP6, clear progresses have been made in what refers to demonstration activities. Data collection activities have been successfully executed by HOLONIX in all of the three end users; even though, in the NECO case, this was accomplished only in March 2019.
In the Reporting Period of WP7, work has been carried out on tasks T7.1 and T7.3 whereas T7.2 has not yet started.
In the context of WP8, new Customer Segments are currently being investigated in the broader areas of Multi-stage production lines and solution providers. Regarding the commercialization plan, steps towards this direction are already accomplished by asking the End-Users to indicate the resources and activities, as well as the corresponding costs, they foresee for the successful integration of Z-Fact0r.
In the context of WP9 the Innovation Manager is constantly enriching and updating the Plan for Use and Dissemination of the Results (PUDR). The communication and dissemination of the solution has also been done via social media. After an analysis of the most suitable tools to spread the project, the LinkedIn and Twitter pages have been opened.
Extensive dissemination activities took place during the 2nd period with our partners’ participation in events, workshops, presentations, Internet / mass media / social media, leaflets, press releases and scientific publications. Finally, training material and organize of workshops is in progress.
I1: Early Stage-Decision Support System for autonomous online inspection, condition monitoring, performance recommendations improvements at shop-floor, with rule-based engine made to address ZDM scenarios: ATLANTIS
I2: Semantic Framework & Ontology for real time data synchronization and data enrichment of shop floor data: EPFL.
I3: Context aware algorithms for defect prediction in real time: EPFL
I4: Knowledge Management Decision Support System for autonomous system calibration and false alarm filtering: EPFL
I5: Reverse Supply Chain policy for efficient defect management in terms of cost, time and environmental impact applied in multistage manufacturing environments: ATLANTIS.
I6: Laser Scanning Technologies and 3D Point-Cloud techniques for rapid zero defect applications: DATAPIXEL.
I7: Early malfunction system deploying 3D Convolutional Neural Network models for joint defect detection and prediction: CERTH
I8: Real time monitoring platform that deploys Deep Learning models for automatic inspection: CERTH
I9: System Management and Optimisation: UBRUN.
I10: Discrete Event Modelling Systems (DES): UBRUN.
I11: A Defect Predictor -The Genomics of Industrial Process through Real-Time Sequential Event Modelling: UBRUN
I12: Deburring technique using robotic arms: SIR.
I13: Production Management Module: HOLONIX
I14: Components repairing techniques combining additive manufacturing and laser processing: CETRI.
Z-Fact0r is expected to have impact in multiple stages of the manufacturing sector. The Z-Fact0r as a complete solution would be responsible for the reduction of the defected parts and also the reduction of the defected parts that are throw away for recycling. This will affect positively the total production cost and also the efficiency of the production system. Further to that by utilizing the data from the prediction module will give to the manufacturers insights of their processes and therefore improve them and potentially eliminate defected parts, which has impact at the final cost of the product. Besides the cost impact, the above reasons will reduces also the material waste and the power needed for manufacturing of products. This will turn the production systems more eco-friendly something that is crucial nowadays.