ESTGV - DEMGI - Artigo em revista científica, não indexada ao WoS/Scopus
URI permanente para esta coleção:
Navegar
Entradas recentes
- Development of a Polymer Filament Extruder: Recycling 3D Printer WastePublication . Guimarães, André; Messias, Samuel; Lopes, João; Salgueiro, José; Gaspar, DanielThis article presents the development of a polymer filament extruder to recycle waste from 3D printing. As additive manufacturing grows within Industry 4.0, managing thermoplastic waste like PLA, ABS, and PET has become a key challenge. The proposed modular system includes a shredder, an extrusion unit, and a winding module to produce high-quality filaments with precise dimensions (1.75 ± 0.03 mm), ensuring compatibility with 3D printers. Aligned with circular economy principles, the system promotes material reuse and reduces environmental impact. Results confirm its technical and environmental feasibility, with potential for large-scale use. Future improvements may include recycling other polymers and using smart sensors and algorithms to optimize the process.
- Industry 4.0 Readiness Assessment – A Comparative Analysis of Portuguese and Brazilian CompaniesPublication . Guimarães, André; Moura, Luciano Raizer; Cardoso, Antonio J. MarquesPurpose: This paper aims to present a comparative analysis of the Industry 4.0 (I4.0) maturity levels between Portuguese and Brazilian industrial companies. This study focuses on identifying signi¯cant di®erences across various evaluation dimensions using a standardized maturity model (MM). Methodology: The same evaluation model, developed by the German Mechanical Engineering Industry Association (VDMA), was applied to Portuguese and Brazilian companies, speci¯cally in the State of Esp {rito Santo (ES). The research encompassed 370 Portuguese industrial companies and 46 Brazilian ones. The VDMA platform was used to process individual results, indicating the levels across six model dimensions and providing an overall score on a scale of 0–5. The data collected were then tabulated to enable a comparative analysis between the two countries. Findings: The study revealed that, on average, Brazilian companies have a lower maturity level (0.95) than Portuguese companies (1.22) on the 0–5 scale for I4.0 readiness. Notably, signi¯cant di®erences were observed in the dimensions of Smart Operations and Employees. Based on these di®erences, this study outlines potential pathways for these companies to enhance their I4.0 maturity levels. Originality/value: This research provides a unique comparative perspective on industrial companies' I4.0 maturity levels in Portugal and Brazil, using a standardized and widely recognized MM. The ¯ndings o®er valuable insights into the speci¯c areas where companies in these countries can focus their e®orts to advance their readiness for I4.0, highlighting the importance of tailored strategies for di®erent national contexts.
- AI-Powered Data Management to Optimize Data Collection and Processing in a Painting LaboratoryPublication . Pereira, Teresa; Oliveira, Marisa; Tavares, Miguel; Guimarães, AndréIndustrial laboratories often remain under-digitized compared to production lines, creating a gap between data acquisition and analytical intelligence, critical for advanced quality control. This study addresses this gap by proposing and validating a novel framework that combines Low-Code digitalisation tools with Machine Learning (ML) and Causal Inference to optimise data collection and analysis in an automotive painting laboratory. A Microsoft Power Apps-based platform was developed in order to digitalise all measurement records, eliminating manual transcription errors (previously ≈ 40.01%) and reducing data-handling time by up to 34% of an operator’s shift, while enabling centralised, traceable storage and Power BI integration. Four datasets were used to assess predictive capacity with Random Forest, XGBoost and Neural Networks; Random Forest consistently provided the most stable results—Mean Absolute Error (MAE) of 0.972, Mean Absolute Percentage Error (MAPE) of 16.45%, and Root Mean Square Error (RMSE) of 1.307. Causal models (Linear Regression, DoWhy, Causal Forest, Double Machine Learning) consistently identified ultrafiltrate I solid content of the electrodeposition process as a dominant causal factor for defects. This study provides a novel framework that bridges digitalisation and ML-based causal reasoning in laboratory settings, offering a scalable approach that can be extended and replicated in other industrial sectors, aiming to develop smart, data-driven quality control systems.
- Business Productivity and the Adoption of Lean and Industry 4.0 Tools: A Regional StudyPublication . Guimarães, André; Rosivalda Pereira; Pereira, Marisa; Pereira,Maria TeresaThis study investigates the relationship between the adoption levels of Lean and Industry 4.0 (I4.0) tools and business productivity among 140 industrial companies in the Central Region of Portugal. Lean and I4.0 adoption indices were constructed and categorized into tertiles. Productivity data were retrieved from the SABI database. Statistical analysis using non-parametric methods revealed a marginally significant association between Lean tool adoption and productivity (Kruskal-Wallis H(2) = 5.30, p = 0.071), indicating a positive trend. In contrast, a statistically significant relationship was found for Industry 4.0 adoption (H(2) = 8.39, p = 0.015, Dunn’s p = 0.039), with companies with low adoption levels underperforming those with medium and high levels. No significant productivity differences were observed by firm size (p = 0.154). These findings highlight the relevance of Lean and I4.0 tools as productivity drivers, regardless of company size, and underscore the importance of promoting structured digital and operational transformation strategies in low digital maturity regions.
- Modified Delphi Method for Insights on Digital Transformation in Asset ManagementPublication . Messias, Samuel; Guimarães, André; Raposo, Hugo; Estácio Marques Mendes Gaspar, Daniel AugustoA utilização de um método de avaliação e de validação é de extrema importância para credibilizar e fortalecer a aceitação de conclusões sobre um determinado tema. Este artigo tem como objetivo descrever a adaptação da técnica Delphi no processo de validação do conteúdo de um questionário sobre a Transformação Digital na Gestão de Ativos. No estudo, participaram 25 especialistas da indústria portuguesa com experiência em Gestão de Ativos e Transformação Digital. O questionário abrangeu os seguintes temas: Gestão de Ativos, Transformação Digital, norma NP EN 55001, sistemas de informação e o ciclo de vida dos ativos. Adicionalmente, foram abordados os modelos de maturidade relacionados com a Gestão de Ativos e a Transformação Digital. Para a avaliação do nível de maturidade na Gestão de Ativos, foi utilizado o modelo do IAM (Institute of Asset Management), enquanto para a avaliação do nível de maturidade digital foi utilizado o modelo VDMA/IMPULS. A técnica Delphi aplicada foi modificada para se adequar ao formato de um questionário, sendo realizada numa única ronda pelos especialistas. Os resultados mostraram uma taxa de resposta de 72%. Os especialistas chegaram a um consenso sobre as 15 perguntas apresentadas, com 10 questões validadas por unanimidade e bem classificadas, juntamente com algumas sugestões de melhoria, enquanto 5 questões não obtiveram tanto consenso. Para a classificação e avaliação, foi utilizada a escala de Likert de 1 a 5, onde “1 — Nada relevante” e “5 — Extremamente relevante”. Esta técnica Delphi modificada permitiu perceber tendências e vetores futuros na Gestão de Ativos e na Transformação Digital. O método Delphi contribuiu para o aprimoramento da formulação do questionário, sendo esta uma etapa intermediária num processo de validação iterativo. A opinião dos especialistas possibilitou a obtenção de feedbacks construtivos para a validação das questões e da estrutura do questionário a ser utilizado.
- Digital Maturity In Industry 4.0: A Bibliometric Review of Evaluation Models for Small and Medium EnterprisesPublication . Guimarães, André; Reis, Pedro; Charrua-Santos, FernandoThe present study, utilizing bibliometric techniques, aimed to structure and analyze the existing scientific literature on Digital Maturity Assessment Models for Industry 4.0, explicitly focusing on small and medium-sized enterprises (SMEs). Despite digital transformation not being a new phenomenon, it continues to be a crucial and relevant concept, significantly impacting SMEs and presenting a valuable opportunity for their integration into the global economy. This research involved a comprehensive review of the scientific literature and a bibliometric analysis to uncover key trends and insights in this field. The study identified the primary research trends and highlighted the ten most influential articles based on citation count and publication frequency. Furthermore, it pinpointed the most cited authors, offering greater clarity on the development level of SMEs in their integration and preparation for the Industry 4.0 paradigm. This analysis provides valuable insights into SMEs’ current state and future directions of digital maturity, facilitating their transition into a more technologically advanced and competitive environment.
- AI-Powered Data Management to Optimize Data Collection and Processing in a Painting LaboratoryPublication . Pereira, Maria Teresa Ribeiro; Pereira, Marisa João Guerra; Tavares, Miguel Guedes; Guimarães, André; Vilarinho, HermilioIndustrial laboratories often remain under-digitized compared to production lines, creating a gap between data acquisition and analytical intelligence, critical for advanced quality control. This study addresses this gap by proposing and validating a novel framework that combines Low-Code digitalisation tools with Machine Learning (ML) and Causal Inference to optimise data collection and analysis in an automotive painting laboratory. A Microsoft Power Apps-based platform was developed in order to digitalise all measurement records, eliminating manual transcription errors (previously ≈ 40.01%) and reducing data-handling time by up to 34% of an operator’s shift, while enabling centralised, traceable storage and Power BI integration. Four datasets were used to assess predictive capacity with Random Forest, XGBoost and Neural Networks; Random Forest consistently provided the most stable results—Mean Absolute Error (MAE) of 0.972, Mean Absolute Percentage Error (MAPE) of 16.45%, and Root Mean Square Error (RMSE) of 1.307. Causal models (Linear Regression, DoWhy, Causal Forest, Double Machine Learning) consistently identified ultrafiltrate I solid content of the electrodeposition process as a dominant causal factor for defects. This study provides a novel framework that bridges digitalisation and ML-based causal reasoning in laboratory settings, offering a scalable approach that can be extended and replicated in other industrial sectors, aiming to develop smart, data-driven quality control systems.
- Development of a Polymer Shredder: Recycling Waste from 3D PrintersPublication . Lopes, João; Samuel Messias; Guimarães, André; salgueiro marques, José Manuel Neto; Estácio Marques Mendes Gaspar, Daniel AugustoThe increasing use of 3D printers has heightened the need for sustainable solutions to manage the polymeric waste generated, such as PLA. This study presents the development of an innovative shredder capable of processing 3D printing waste and transforming it into reusable granules for the production process. The project involved the design of a knife mill, adapted for PLA, with blades offset by 30°, ensuring efficient distribution of cutting forces and reducing stress on the components. Critical components, such as blades, shafts, and motor, were dimensioned using precise calculations to determine cutting force, torque, and material resistance. The system was validated through finite element analysis (FEA), ensuring structural robustness and an adequate safety factor. The system allows for the adjustment of granule size using sieves with different calibers, making it adaptable to process requirements. It also includes safety devices that ensure reliable operation and protect the operators. The equipment proved versatile and capable of processing common polymers such as ABS, PETG, and PA. The results confirm that the shredder is a practical, efficient, and sustainable solution, contributing to the circular economy and reducing the environmental impact of 3D printing. In the future, improvements in design and automation could enhance its scalability and facilitate integration into industrial processes.
- Implementation of Autonomous Mobile Robots in Intralogistics: Simulations in a Case StudyPublication . Guimarães, André; Silva A.; Teixeira J.; Gomes F.; Martins S.Autonomous Mobile Robots (AMRs) are increasingly being integrated into intralogistics operations, playing a pivotal role in material handling within the Industry 4.0 framework. In this context, a central unit oversees programming and planning decisions, while AMRs independently communicate and negotiate with other resources, such as machines and systems, decentralizing decision-making processes. These advancements have significantly impacted traditional methods and decision-making processes for planning and control. This study analyzes the use of simulation to evaluate AMRs in supplying XYZ factory assembly lines, with a view toward future implementation on the factory floor. Beyond comparisons, the study aims to assess AMR performance to determine their feasibility and provide analytical validation. Various simulation models were developed using SIMIO software to achieve these goals, followed by iterative adaptations and testing through simulation. The findings provide valuable insights into the role of AMRs in Industry 4.0- based production networks. They offer production managers practical guidance for determining optimal configurations and evaluating the performance impacts of AMR-based production networks compared to traditional assembly lines.
- Mixed mode interlaminar fracture of carbon nanotubes enhanced epoxy/glass fiber compositesPublication . Silva, H.; Ferreira, J. A. M.; Costa, J.D.M.; Capela, C.Present paper studied the improvement of the fracture toughness under mixed mode loading obtained by using carbon nanotubes reinforcement in fiber glass mats/epoxy laminates. Mixed-mode bending tests were performed considering different loading ratios GII/GI. Laminates were manufactured using the epoxy resin Biresin® CR120 reinforced with fiber glass triaxial mats ETXT 450 and multiwalled carbon nanotubes with 98% of carbon. It was observed that the total fracture toughness increases linearly with the mode II loading component and that linear mixed-mode fracture criteria reproduces the GI versus GII relationship. The incorporation of small quantity, up to 0.5%, of carbon nanotubes into matrix improves significantly mixed-mode fracture toughness.