Publication
Intelligent System for Miscalibration Detection and Quality Prediction in Automotive Assembly Line
| datacite.subject.fos | Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática | |
| datacite.subject.fos | Engenharia e Tecnologia::Engenharia Mecânica | |
| datacite.subject.sdg | 04:Educação de Qualidade | |
| datacite.subject.sdg | 09:Indústria, Inovação e Infraestruturas | |
| dc.contributor.advisor | Leitão, Paulo | |
| dc.contributor.advisor | Barbosa, José | |
| dc.contributor.advisor | Borges, André Pinz | |
| dc.contributor.author | Gonçalves, Waldyr Turquetti | |
| dc.date.accessioned | 2025-10-16T08:15:51Z | |
| dc.date.available | 2025-10-16T08:15:51Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract | This work applies data analysis techniques, Machine Learning and Optimization Algorithms to develop two data analysis systems in the OpenZDM project for the assembly measuring stations in the workshop of an automotive plant. These systems were designed to generate insights into the production line, helping the industry to achieve Zero Defect Manufacturing (ZDM) goals. The case study chapter presents the context for understanding the characteristics of the case study. The following chapters detail the tools created to implement the ZDM strategies. The first is the system for identifying the miscalibration of measuring stations, consisting of a microservice that provides alerts informing whether or not the stations are miscalibrated. The second system is the diagnostic system, made up of four microservices. Within the functionalities of this system, there is the functionality responsible for making quality predictions and the other for improving the hyperparameters of the prediction model. The tests carried out demonstrated theneffectiveness of the systems in dealing with the large volume of data and creating useful insights for the production line. In addition, the cases of miscalibration detection and the prediction process have been validated, affirming the usefulness of the services within the automotive factory’s production line. | por |
| dc.description.abstract | Este trabalho aplica técnicas de análise de dados, Aprendizado de Máquina e Algoritmos de Otimização para desenvolver dois sistemas de análise de dados no projeto OpenZDM para as estações de medição de montagem da oficina de uma fábrica automotiva. Esses sistemas foram projetados para gerar insights sobre a linha de produção, ajudando a indústria a atingir as metas de Manufatura com Defeito Zero (ZDM). O capítulo do caso de estudo apresenta o contexto para a compreensão das características do estudo do caso. Os capítulos seguintes detalham as ferramentas criadas para implementar as estratégias do ZDM. O primeiro abordado é o sistema para identificação da descalibração das estações de medição, composto por um microsserviço, que fornece alertas que informam se as estações estão descalibradas ou não. O segundo sistema é o sistema de diagnóstico, composto por quarto microsserviços, dentro das funcionalidades desse sistema existe a funcionalidade responsável por fazer a predição de qualidade e o outro por aprimorar os hiperparâmetros do modelo de predição. Os testes realizados demonstraram a eficácia dos sistemas em lidar com o grande volume de dados e criar insights úteis para a linha de produção. Além disso é feita validações dos casos de detecção de descalibração e do processo de predição, afirmando a utilidade dos serviços dentro da linha de produção da fabrica automotiva. | por |
| dc.description.sponsorship | This work was partially supported by the openZDM project under the HORIZON-CL4-2021-TWIN-TRANSITION01 call, Grant Agreement No. 101058673. It was also supported by national funds through FCT/MCTES (PIDDAC): CeDRI, UIDB/05757/2020 (DOI:10.54499/UIDB/05757/2020)and UIDP/05757/2020 (DOI:10.54499/UIDP/05757/2020); and SusTEC, LA/P/0007/2020 (DOI: 10.54499/LA/P/0007/2020). I would like to express my sincere gratitude to Prof. José Carlos Rufino, Prof. Rui Pedro Lopes, and the team at the Polytechnic Institute of Bragança (IPB) for their support in setting up the development environment, enabling the replication of shop-floor production machine functionalities, and for designing the microservice-based infrastructure. I also thank the openZDM team, especially the members from the vehicular sector, for sharing practical knowledge and valuable insights related to shop-floor operations. Furthermore, I would like to thank Prof. Paulo Leitão, Prof. Andre Pinz Borges, Prof. Rui Tadashi, and PhD student Victória Melo, whose guidance and teachings were fundamental to the development of this work. | |
| dc.identifier.tid | 204017700 | |
| dc.identifier.uri | http://hdl.handle.net/10198/34842 | |
| dc.language.iso | eng | |
| dc.relation | Research Centre in Digitalization and Intelligent Robotics | |
| dc.relation | Research Centre in Digitalization and Intelligent Robotics | |
| dc.relation | Associate Laboratory for Sustainability and Tecnology in Mountain Regions | |
| dc.rights.uri | http://creativecommons.org/licenses/by-sa/4.0/ | |
| dc.subject | ZDM | |
| dc.subject | Real-time monitoring | |
| dc.subject | Regression models | |
| dc.subject | Optimization algorithms | |
| dc.title | Intelligent System for Miscalibration Detection and Quality Prediction in Automotive Assembly Line | por |
| dc.type | master thesis | |
| dspace.entity.type | Publication | |
| oaire.awardTitle | Research Centre in Digitalization and Intelligent Robotics | |
| oaire.awardTitle | Research Centre in Digitalization and Intelligent Robotics | |
| oaire.awardTitle | Associate Laboratory for Sustainability and Tecnology in Mountain Regions | |
| oaire.awardURI | info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F05757%2F2020/PT | |
| oaire.awardURI | info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDP%2F05757%2F2020/PT | |
| oaire.awardURI | info:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/LA%2FP%2F0007%2F2020/PT | |
| oaire.fundingStream | 6817 - DCRRNI ID | |
| oaire.fundingStream | 6817 - DCRRNI ID | |
| oaire.fundingStream | 6817 - DCRRNI ID | |
| project.funder.identifier | http://doi.org/10.13039/501100001871 | |
| project.funder.identifier | http://doi.org/10.13039/501100001871 | |
| project.funder.identifier | http://doi.org/10.13039/501100001871 | |
| project.funder.name | Fundação para a Ciência e a Tecnologia | |
| project.funder.name | Fundação para a Ciência e a Tecnologia | |
| project.funder.name | Fundação para a Ciência e a Tecnologia | |
| relation.isProjectOfPublication | 6e01ddc8-6a82-4131-bca6-84789fa234bd | |
| relation.isProjectOfPublication | d0a17270-80a8-4985-9644-a04c2a9f2dff | |
| relation.isProjectOfPublication | 6255046e-bc79-4b82-8884-8b52074b4384 | |
| relation.isProjectOfPublication.latestForDiscovery | 6e01ddc8-6a82-4131-bca6-84789fa234bd | |
| thesis.degree.name | Informática |