CORDIS - Forschungsergebnisse der EU

Zero-defect manufacturing strategies towards on-line production management for European factories


Report on standards (a) used and (b) generated in Z-Fact0r

Standardisation is an essential issue as only if the developed technology within Z-Factor can be standardized, broad technology take-up outside of the consortium and after the end of the project can be guaranteed. As a result, exist-ing but also future consensus-based standards, focusing on open-standards, will create a firm basis for technical procurement, support communication through standardized terminology and concepts and ensure interoperability, fitness for use and market relevance. Drawing up concepts at an early stage by a consensus-based standardization process and in close cooperation is one of the central requirements for the success of innovative manufacturing approaches and for a rapid implementation in industrial practice. Z-Fact0r will establish informal contact with other relevant projects and networks in this field ensuring a good level of information flow and mutual awareness be-tween the project and the above bodies in order to assess standardisation opportunities and sharing knowledge and best-practices that foster the uptake of technologies. Representing bodies of such projects will be invited to partici-pate in workshops and other dissemination activities to further enhance cooperation levels among projects. This way Z-Fact0r will support new regulations and standards to accelerate adoption of the new manufacturing systems, and hence bring value back to the EU. In order to do so, the standardization procedures within Z-Fact0r will be done in two steps. In the first step, already existing standards will be evaluated (as exposed above) in term of their applicability to Z-Factor technologies. With the compatibility to existing standards, Z-Factor technologies can be implemented easily. Furthermore, the ac-ceptance of Z-Factor in the market will be much better. In a second step, the technologies developed within the project will be standardized. This especially holds true for the interfaces and for the communication protocol. Also, standardised UI libraries that incorporate easy-to-access symbols and buttons for aged workers will be incorporate in production systems.

Report on DURIT demo activities (M42)

In DURIT there is not a production line but production “cells”, since the production strategy is for custom order, and a large quantity of different production orders can be performed daily at each one of these machines, making this demo task extremely valuable for the project, as it enables validating the technology in a production cell as op-posed to the previous demo tasks that focused on production lines. This demo task will be divided in two main sub-tasks, each one focusing on testing different stages of the produc-tion chain: • Pre-sintered operations: green machining • Post-sintering operations: grinding and turning in Danobat-Overbeck IRD; Hauser S35 and Mori-Seiki CNC Lathes In these two stages of the production chain, the tests will focus on the production of: (1) sealing rings; (2) cut-ting/drawing dies and (3) mandrels; as these products require very narrow dimensional tolerances and very high surface quality. We are aiming to obtain tolerances down to +/ -1 micron and automatic validation, as actually, in all the manufacturing equipment, the quality control is monitored manually essentially by operator measurement, with no sensor technology.

Management update report 1

Outline document describing the achieved progress for the first 12 month reporting period

Report on MSL demo activities (M42)

The Z-Fact0r strategies will be implemented in the premises of MSL. This task will be linked with the results of the use case design in T1.4. The assembly is a multistage process starting with 1) Visual inspection of the base PCB 2) Glue dot dispense 3) Placement of the die/component 4) Glue cure 5) Wirebond die The glue dot placement process is shown in the figure below and is the most critical problem area. The placement machine (Tresky T-8000) has no optical inspection and no method to automatically correct for the amount of glue being dispensed into the cavity. Consequently visual inspection of the assemblies has to take place and manual adjustment of the amount of glue deposited.

Technology Validation Plan

"Definition of a Technology Validation Plan (methodology, Test Cases, Test Procedures, timing, performance and defect log forms) and we’ll execute it in iterations until no ""severity level"" 1 and 2 errors are left. On all detected errors or mal-formed functionality, we’ll apply corrective design and re-implementation, we’ll execute Regression Testing and we’ll apply Change Management Control (CMC) procedures."

Report on NECO demo activities (M42)

The Z-Fact0r concept and methodologies will be integrated in the tap manufacturing process of NECO. The overall system of optical sensors, signal modifiers, DAQ systems, monitoring software etc, that will be integrated in the process will be reported. All the necessary modifications in the design of the components integrated in the different parts of the manufacturing process for the implementation of the task, as well as the comparison between the monitoring of the process and the final product will be reported. The optimization of the design and integration of the sensors system along with the tap dimensional characterization data sheets will be modified and customized, in order to be stored and processed from the R&D partners accordingly, will be described in this deliverable. All the necessary hardware chosen for the task will also be stated.

Integration Discipline and Incremental Strategy

Z-Factor architecture encompasses the design and development of a diverse set of technologies with different specifications, requirements and developing methodologies. Therefore, as an early activity in the Task, we’ll define an integration discipline to be followed for the integration exercise, incl. methodology, activities to be performed, hierarchy of components / interfaces integration, incremental builds, etc.

Methodology for Z-Factor solution validation / evaluation

• Methodology for the types of evaluation activities, timeframe and expected results per activity, instruments to be used (e.g. users’ acceptance questionnaires, impact check-lists and data collection forms) to assist the different types of evaluation exercise. • Technical indicators for performance assessment (KPIs). Indicative e.g. number of correctly detected defects per time interval, precision and recall (where precision will capture positive defect-detection values, and recall will capture sensitivity), F-Measure, False Alarm Rate, Min Time Between Failures, System Up Time, etc. • User/stakeholder acceptance indicators based on 9241/10, 9241/11 and 9241/110 standards. Ergonomics, user-friendliness, usability and plant stakeholders’ satisfaction will be also measured in that respect. E.g. (a) Ful-filment of requirements: “The solution fulfils the trial requirements”, (b) Learnability: “It is easy to start to use the solution and learn functionalities”, (c) Understandability: “The solution is easy and self-clear to understand and the concepts and terminology are understandable”, (d) Efficiency: “The time and resources required to achieve the objectives of the solution are reasonable, the solution is fast enough and does not require too many steps”. • Indicators for assessing the impact of Z-Factor on the factories. Indicative e.g: a) Improvement in the number of production facilities breakdown and amount of idle time, b) machinery deterioration rate and achieved im-provement, c) reduction of production costs, of waste and scrap, d) production output quality (qualified output / total output produced), e) % of products/workpieces successfully repaired, f) improvement in prediction and prevention efficiency, improvement in detection efficiency, g) improvement in production cost, h) improvement in single-stage production defect rate, average multistage production defect rate (goal for zero-defects), i) im-provement in defect propagation to downstream stages.

Z-Fact0r Use Cases

Each use case will be developed in detail, integrating the consolidated user requirements, customizing the Z-Fact0r System architecture and developing a proper set of Key performance Indicators to evaluate the performance achieved. Each use case will be defined during its whole lifecycle identifying which performance can be measured with the demonstrators.

Techniques for product reworking

A novel approach will be used by the Z-fact0r consortium to deal with the repairing/reworking, fully automated process of defected parts in a manner that will result in quality restoration of the defect without any deviation of the non-repaired parts. In this respect, the defected areas will be patched with dispensed material of the same kind as the underlying component using ink or paste dispensing techniques.

Deburring remanufacturing approaches

A novel concept of intelligent robotic deburring cells, implementing ZF’s ZD strategies will be developed. Metrological on-line inspection tools will evaluate product quality at each deburring stage identifying defects; a parametrized model based control will interpret the sensory input and, thanks to a model based control embedding the knowledge of the deburring process, will automatically generate an optimized deburring cycle, choosing the best tools and setting the ideal working parameters. The Robot cell supervisory control will be then able to automatically generate the multi-stage deburring cycle once sensed the quality achieved. After each deburring stage, the quality achieved will be checked in order to better tune the next operation or, in the worst case, define a proper repairing action. Proper compliant tools and ZD deburring strategies will also be defined. One of the major challenges will deal with the development of the inference engine, and its encoding into the robotic cell supervisory control, in order to add the robotic cell the “intelligence” to autonomously generate ZD strategy in real-time. Integration and validation will be carried out experimentally on a prototype at TRL 5-6.

Management update report 2

Outline document describing the achieved progress from the 13th till the 24th month reporting period

Z-Fact0r Use Cases Updated

This deliverable is an update of D1.4, including the definition and description of NECO use case.

Z-PREVENT Validation and Verification of KPIs

Validation and verification of the KPI models by experiments on the line and with the help of line managers from the use-case factories. The models for the 5 KPIs (Productivity, Efficiency, Quality (Customer Satisfaction), Environmental Impact, and Inventory levels) will be measured and fine-tuned. The already installed actuators and sensors will be used to monitor the KPI over time

Report on the analysis of SoA, existing and past pro-jects/initiatives

Investigation and update on the SoA, analysing new, existing and past projects, initiatives and if new products/technologies have been introduced into the market.

Report on Solution Validation

From the execution of the demonstrations at the operational environment of the project pilot sites (WP6), we’ll analyse the collected data and feedback; and we’ll synthesize and document our validation conclusions and conclude on the “lessons-learned” from our validation.

Management update report 3

Outline document describing the overall achievements of the project for all the duration of the project

Data Management Plan (DMP)

The plan will describe ways to manage all research data, and metadata, during and after the project duration.

Report on training activities

Concept for internal and external trainings will be developed in order to provide short courses, seminars or e-learning modules which will provide understanding of the Z-Fact0r approach and concepts and support the take-up of the Z-Fact0r results by potential customers This will be achieved by the identification and analysis of training needs, the identification of the appropriate train-ing set-up and the elaboration of the training documents and training guidelines on the basis of the project continuous integration. All the training materials will follow a ‘learning by doing’ methodology, such as the ‘teaching factory’. A final version will be adapted according to the feedback of the training participants. The final training documents will compose a competence management system that may then be used for take-up activities and further exploitation of the project results.

White Paper on transferability and take-up of Z-Factor system

Documentation of identified obstacles/barriers and the methods used to overcome them. Definition of technical and non-technical requirements for transferability and take-up of the Z-Factor system (White-Paper on taking up and applying Z-Factor in other manufacturing operational environments).

Roadmap for the project results’ uptake

The innovation department of Confindustria will carry out research to lay out a roadmap enabling the adoption and takeup of the innovation, among several industrial sectors, most of which are concretely represented by well-structured business in Region Lombardy and in the other regions participating in the project. The study at the core of roadmapping activities will entail two phases: i. desk research based on technology brokerage systems available at Confindustria, to characterise demand for zero-defect production globally matching; ii. organisation of three thematic workshops to foster discussion on the project results and enable face-to-face interviews and meetings with about 20 companies from the consortium regions, in order to better identify specific needs and concrete prospects for products development that might build on project’s results. Each workshop will focus on the theme of industrial sectors represented in WP6 through the use cases carried out at project partners’ premises, and will build on the outcomes of such demonstrations, and in particular: i) Work-shop 1 will focus on electronics and robotics; ii) Workshop 2 will focus on rubber, plastics, thermoplastics, rotomoulding, etc; iii) Workshop 3 on metals, together with shaving processing, mechanical processes and metal treatment.

Z-MANAGE Prescriptive Green Optimisation Model and Solver

Mathematical modelling of respective KPIs such as customer service and energy/power consumption as functions of decision variables. The resultant will be a multi-objective mixed-integer programming (MOMIP) model. To solve the model, a metaheuristic solver will be developed.

Z-Factor system for Deployment & Demonstration

In detail definition of the overall methodological framework for the validation of the Z-Factor solution with respect to the evaluation of the level of fulfilment of the project objectives, and the value of its results in terms of their apprecia-tion by factory stakeholders and decision makers, and in terms of their impact on the factory operation 1st version of the deliverable will be submitted on M32 to accomodate the WP6 (demo). The final version on M36 - end of T5.2.

Z-Fact0r middleware platform and components

Definition of the specifications for the middleware and related components that integrates the sensors network with the proposed Z-Fact0r architecture. The middleware itself will act as service bus, based on SOA architecture, for the communication between levels. In this sense, all the elements to be attached to the middleware will be considered. The integration of the middleware with the multi sensor/actuator cloud will be achieved through the Device Managers which will allow an interoperable communication mechanism based on standardized protocols (e.g. OPC UA) with any kind of sensors and actuators on the shop floor.

Project Website

A project website will be developed and also continuously updated with project progress information.


Towards Robust Early Stage Data Knowledge-based Inference Engine to Support Zero-defect Strategies in Manufacturing Environment

Autoren: T. Vafeiadis, D. Ioannidis, C. Ziazios, I.N. Metaxa, D. Tzovaras
Veröffentlicht in: Procedia Manufacturing, Ausgabe 11, 2017, Seite(n) 679-685, ISSN 2351-9789
Herausgeber: Elsevier
DOI: 10.1016/j.promfg.2017.07.167

A framework for inspection of dies attachment on PCB utilizing machine learning techniques

Autoren: Thanasis Vafeiadis, Nikolaos Dimitriou, Dimosthenis Ioannidis, Tracy Wotherspoon, Gregory Tinker, Dimitrios Tzovaras
Veröffentlicht in: Journal of Management Analytics, Ausgabe 5/2, 2018, Seite(n) 81-94, ISSN 2327-0012
Herausgeber: Taylor & Francis
DOI: 10.1080/23270012.2018.1434425

EventiC: A Real-Time Unbiased Event-Based Learning Technique for Complex Systems

Autoren: Morad Danishvar, Alireza Mousavi, Peter Broomhead
Veröffentlicht in: IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2018, Seite(n) 1-14, ISSN 2168-2216
Herausgeber: IEEE Advancing Technology for Humanity
DOI: 10.1109/tsmc.2017.2775666

EGEP: An Event Tracker Enhanced Gene Expression Programming for Data Driven System Engineering Problems

Autoren: Zhengwen Huang, Maozhen Li, Alireza Mousavi, Morad Danishva, Zidong Wang
Veröffentlicht in: IEEE Transactions on Emerging Topics in Computational Intelligence, Ausgabe 3/2, 2019, Seite(n) 117-126, ISSN 2471-285X
Herausgeber: IEEE
DOI: 10.1109/tetci.2018.2864724

Schema Theory-Based Data Engineering in Gene Expression Programming for Big Data Analytics

Autoren: Zhengwen Huang, Maozhen Li, Christos Chousidis, Alireza Mousavi, Changjun Jiang
Veröffentlicht in: IEEE Transactions on Evolutionary Computation, Ausgabe 22/5, 2018, Seite(n) 792-804, ISSN 1089-778X
Herausgeber: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tevc.2017.2771445

Zero defect manufacturing: state-of-the-art review, shortcomings and future directions in research

Autoren: Foivos Psarommatis, Gökan May, Paul-Arthur Dreyfus, Dimitris Kiritsis
Veröffentlicht in: International Journal of Production Research, 2019, Seite(n) 1-17, ISSN 0020-7543
Herausgeber: Taylor & Francis
DOI: 10.1080/00207543.2019.1605228

A Deep Learning framework for simulation and defect prediction applied in microelectronics

Autoren: Nikolaos Dimitriou, Lampros Leontaris, Thanasis Vafeiadis, Dimosthenis Ioannidis, Tracy Wotherspoon, Gregory Tinker, Dimitrios Tzovaras
Veröffentlicht in: Simulation Modelling Practice and Theory, Ausgabe 100, 2020, Seite(n) 102063, ISSN 1569-190X
Herausgeber: Elsevier BV
DOI: 10.1016/j.simpat.2019.102063

Fault Diagnosis in Microelectronics Attachment Via Deep Learning Analysis of 3-D Laser Scans

Autoren: Nikolaos Dimitriou, Lampros Leontaris, Thanasis Vafeiadis, Dimosthenis Ioannidis, Tracy Wotherspoon, Gregory Tinker, Dimitrios Tzovaras
Veröffentlicht in: IEEE Transactions on Industrial Electronics, Ausgabe 67/7, 2020, Seite(n) 5748-5757, ISSN 0278-0046
Herausgeber: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tie.2019.2931220

An approach to development of system architecture in large collaborative projects

Autoren: Gökan May, Dimosthenis Ioannidis, Ifigeneia N. Metaxa, Dimitrios Tzovaras, Dimitris Kiritsis
Veröffentlicht in: APMS 2017: Advances in Production Management Systems. The Path to Intelligent, Collaborative and Sustainable Manufacturing, 2017, Seite(n) 67-75
Herausgeber: Springer International Publishing
DOI: 10.1007/978-3-319-66923-6_8

Zero Defect Manufacturing of Microsemiconductors – An Application of Machine Learning and Artificial Intelligence

Autoren: Zhengwen Huang, Veerendra C Angadi, Morad Danishvar, Ali Mousavi, Maozhen Li
Veröffentlicht in: 2018 5th International Conference on Systems and Informatics (ICSAI), 2018, Seite(n) 449-454, ISBN 978-1-7281-0120-0
Herausgeber: IEEE
DOI: 10.1109/icsai.2018.8599292

A Scheduling Tool for Achieving Zero Defect Manufacturing (ZDM): A Conceptual Framework

Autoren: Foivos Psarommatis, Dimitris Kiritsis
Veröffentlicht in: Advances in Production Management Systems. Smart Manufacturing for Industry 4.0 - IFIP WG 5.7 International Conference, APMS 2018, Seoul, Korea, August 26-30, 2018, Proceedings, Part II, Ausgabe 536, 2018, Seite(n) 271-278, ISBN 978-3-319-99706-3
Herausgeber: Springer International Publishing
DOI: 10.1007/978-3-319-99707-0_34

Suche nach OpenAIRE-Daten ...

Bei der Suche nach OpenAIRE-Daten ist ein Fehler aufgetreten

Es liegen keine Ergebnisse vor