EN - LIGM - Linha de Investigação em Gestão da Manutenção
URI permanente para esta coleção:
Navegar
Entradas recentes
- Dynamic Maintenance on Ships: Integrating Threat DimensionPublication . Lampreia, Suzana; Vilarinhos, Valter; Lobo, Victor; Morgado, Teresa; Alves, JoãoMaintenance management on ships is a pillar for operations success. As in other industry systems and equipment’s, maintenance has evolute from a corrective maintenance to systematic preventive, to condition based maintenance. To accomplish a condition-based maintenance with high quality and a dynamic maintenance concept can be considered various non-destructive techniques, and data from sensors. After obtaining the data from the various systems, it should be registered, and statistical models should be developed and applied to enhance the human evaluation about the data. The Failure Mode and Effect Critically Analysis may be one reliability evaluation and design tech-nique for data treatments and to calculate the risk about a system or equipment. As a case study it was selected a propulsion diesel engine and it was applied a Failure Mode and Effect Critically Analysis, which was modified, FMECA-CC, to integrate various sever-ities, and to integrate data from condition control and the Threat. To limit the study presentation on the present article, it was decided to developed only the variable Threat. The threat was integrated on the study because it represents a study on a military envi-ronment. It was considered 5 levels for the threat variable, which are manually introduced by an operator who receives hierarchic input about the threat environment and introduces on a propulsion decision-making software. The results showed that FMECA-CC may be applied on diesel engine monitoring on board ships integrated data from sensors and also the variable threat.
- Relating Texts Using Intersection GraphsPublication . Vilarinhos, Valter; Lobo, Victor; Lampreia, SuzanaA corpus is a set of texts related to some topic, theme or subject This means that the texts in a corpus, although produced by distinct authors, have meanings related to that subject, topic or theme. This work addresses the problem of descriptive statistics of variables whose observed values are texts, aiming at descriptive statistics of corpus variables (CVs) and its interpretation. Specifically, the aim is to build text mining methods to compare and relate CVs, assuming texts as its values, using intersection graphs and hypergraphs as the basic mathematical representation of CVs and its relations. The data sets involved have the usual tabular structure, crossing the values assumed by p CVs, observed on n individuals, objects , cases or documents. The values assumed by these CVs are texts of any length, formed by sets of words in a specific language, expressing opinions or other content created by its authors. If T is the set of all texts (observed values) in such data set, let t ε T; t is a subset of V, formed by words of that specific language, obtained by tokenization of all texts in T. It is shown that hypergraphs and intersection graphs occur in this context as natural mathematical representation tools for texts and its relations. This mathematical representation is a powerful support for statistical mining of texts and its relations, namely text mining in the context of maintenance management and other engineering disciplines.
- Innovation in structural integrity prediction modelsPublication . Lampreia, Suzana; Alves, João; Morgado, Teresa; Martins, RuiStrong predictive models are needed for precise, reliable mechanical design in the aeronautical industry, to control the typically observed fatigue var-iability. Especially with the advances of additive manufacturing and the emer-gence of organic geometries. This paper reviews the development of structural integrity models, from early empirical formulations to advanced physics-based methods. The historical reliance on empirical models like Paris Law and the edge of its concept, The NASGRO, which, despite their utility, are often limited due to their empirical nature, is analyzed. One of the central themes of discussion is the scientific conflict between the "Global Driving Force" school, centered on extrinsic crack closure, and the “Local Cyclic Plasticity school, mainly focused on the induced plasticity at the crack tip. The limitations of both “schools” are scrutinized, including the inability of closure-based models to describe the be-havior at high stress ratios and the kinematic constrains linked to the local plas-ticity models. At the end of this research paper, the Christopher-James-Patterson (CJP) model is presented as a conciliatory approach, considering the crack as a plastic inclusion within an elastic material. This model provides a mechanistic description that accounts for plasticity-induced shielding and retardation without requiring arbitrary geometry correction factors. This review suggests the CJP model as the future of fatigue crack growth predictions.
- The Bridge Between Artificial Intelligence and Predictive Maintenance in Industry 4.0Publication . Arez, Daniel; Navas, Helena V. G.; Gaspar, Pedro; Andrea PratiThis systematic literature review explores the intersection of Artificial Intelligence (AI) and Predictive Maintenance (PdM) within Industry 4.0. Using a PRISMA-based methodology, 123 studies published between 2014 and April 2024 were analyzed to characterize technological trends, algorithmic choices, industrial applications, and evaluation practices. The review reveals a consistent growth of research interest, driven by the widespread adoption of Internet of Things (IoT) devices and increased data availability. The manufacturing sector dominates the literature, although most studies rely on standardized datasets rather than real industrial environments. Among the identified AI methods, Random Forest (RF), Support Vector Machine (SVM), Decision Tree (DT) and K-Nearest Neighbors (KNNs) represent the most frequently applied algorithms for tasks such as failure prediction, fault detection, and remaining useful life (RUL) estimation. Model performance is commonly evaluated with Accuracy (Acc), Precision, Recall, F1-Score, and Root Mean Square Error (RMSE), reflecting the prevalence of both classification and regression-based PdM analyses. Despite significant advances, this review identifies persistent gaps, including limited domain diversity, scarce long-term real-world validation, and insufficient use of eXplainable AI (XAI) techniques. The findings highlight the need for broader domain coverage, improved interpretability, and validation under realistic industrial conditions. Overall, this review consolidates current knowledge on AI-enabled PdM and outlines critical directions to enhance reliability, transparency, and industrial relevance in the context of Industry 4.0.
- Navy Ship Dynamic MaintenancePublication . Lampreia, S.; Vairinhos, V.; Lobo, V.A maintenance system complemented with condition monitoring is suggest being implemented in Navy Ships to allow its patrol on the large maritime territory of Portugal with high performance and operationality. To that an advanced maintenance system is required oriented towards Condition Monitoring. That must be done while maximizing system performance, as well as improved high level of ship availability and minimal oil consumption in the form e.g. lubricants or refrigerant gases but also marine gasoil reducing the environment impact. To meet these objectives, we propose the development of a Dynamic Maintenance Management System (DMMS) based on dynamic prioritization process for technical interventions. That is where Failure Modes, Effects and Criticality Analysis (FMECA) comes in - a structured risk analysis framework to mitigate the propensity of failures occurring with consequences on ship systems and persons. With FMECA, maintenance priorities can be changed somewhat dynamically due to components and failure modes that are most critical based on actual performance.
- Machine Learning applied to LCA of mechanical components obtained by SLMPublication . Alves, J.; Morgado, Pereira M.; Lampreia, S.Over time, industries are starting to be concern about their impact on the environment, and therefore new technologies, have arrived with the capability of improving the sustainability of the industrial paradigm. Machine Learning (ML) and Artificial Intelligence (AI) are taking over the global industry, by offering new data drive solutions capable of optimizing decision-making and operation. Meanwhile, new advanced manufacturing technologies such as Selective Laser Melting (SLM), also known as 3D printing, are spreading through the market, due to the new doors unlocked by the possibility of manufacturing high for a minimal time and use of material. Together, these new technologies of industry 4.0, promise to bring an all-new future, sustainable and resilient. The aim of this work is to analyze how can ML and AI expand each phase of the LCA of a light alloy, produced by SLM, in order to improve the sustainability of this new advanced manufacturing technologies.
- Advanced Methodologies Applied in a Dynamic Maintenance ContextPublication . Lampreia, S.; Lobo, Victor; Vairinhos, V.Maintaining military systems in a dynamic context presents unique challenges that require advanced methodologies to ensure the operational readiness and effectiveness of the Armed Forces. This article analyses and discusses advanced methodologies applied in the maintenance of military technology, highlighting their contributions to the optimization of the maintenance process in constantly evolving operational environments. Initially, we address the importance of predictive maintenance, which uses techniques such as real-time data analysis, condition monitoring and machine learning algorithms to predict failures and proactively schedule maintenance interventions. This approach minimizes downtime, and costs associated with corrective maintenance, significantly increasing the availability of military systems. Additionally, we explore the application of emerging technologies, databases and the use of smart sensors integrated into military equipment. These technologies enable the continuous collection of performance and condition data, enabling a more comprehensive understanding of the state of each component and system. This not only facilitates early detection of anomalies, but also supports data driven decision making, resulting in more efficient and accurate maintenance. The use of condition-based maintenance approaches is also discussed in this article. These methodologies consider not only the age or time of use of the equipment, but also its actual state of operation. By continuously monitoring operating conditions, it is possible to extend the life of systs, reduce unnecessary wear and tear, and plan maintenance interventions more effectively. Finally, we highlight the importance of integrating computer-aided maintenance management systems and decision support systems. The combination of these tools provides a comprehensive environment for planning, executing and analysing maintenance activities, enabling more efficient resource management and a more agile response to operational demands. In short, this article demonstrates how the adoption of advanced methodologies in the maintenance of military technology is fundamental to ensuring the operational readiness and effectiveness of the Armed Forces in a dynamic and constantly evolving context.
- Ship Dynamic MaintenancePublication . Lampreia, Suzana; Vairinhos, Valter; Lobo, Victor; Morgado, Teresa; Navas, HelenaConsidering the extent of the Portuguese sea and the naval ships necessary for its monitoring, the implementation of a maintenance system addressing those ships condition monitoring, is needed. That system must guarantee, also, a high performance of ships and systems availability, reduced pollution control, lubricants, refrigerant gases and marine gasoil consumptions. To achieve all this, a dynamic maintenance manage ment system (DMMS) based on a dynamic management of technical interventions is required. In this paper some risk analysis maintenance techniques are explored, and the most applicable ones will be applied to selected ship equipment. In the present research, a case study of a specific equipment is presented and the corresponding risk analysis in maintenance context is studied. The case study illustrates the importance of having a dynamic maintenance system allowing to act when needed, and, in case of no action, the risk of non-maintenance. To validate our theory the research results and its analysis will be highlighted, and, finally, some conclusions are presented.
- The Colombia Armed Conflict - A Review Until 2020Publication . Lampreia, S.Colombian armed conflict origins remote to the beginning of 20 cen tury, and even after the 2016 peace agreement, it is a reality. Colombia's armed conflict existed by the force of the armed hand of the FARC and other insur gents’ groups, they financed themselves in illegal acts like arms and drug traf ficking, abductions, coercion, and extortion. Till 2020, several FARC dissidents have returned to the guerrillas because they doubted Colombian political power defense their interests. This paper objective is to give a perspective of origins of the armed conflict in Colombia, and its development up to 2020. To expose the conflict, the facts and literature will be related demonstrating the conflict dy namic and involved actors. To support state-of-the-art during the 2016 peace agreement implementation, it was proceeded to interviews to some ex-observers that provided humanitarian support in the United Nations. In the end, it is pos sible to understand better the conflict, and it was made some reflections, which can contribute for possible solutions for the conflict.
- Dynamic Maintenance Based on Fuzzy LogicPublication . Lampreia Suzana; Mestre, Inês; Morgado T. & Navas H.Currently, certain maritime assets are confronted with the challenge of optimizing their performance, even when resources are limited. They strive to minimize intervention actions on equipment while upholding safety standards and acceptable performance levels. Ships, which are not yet autonomous, serve as maritime assets responsible for transporting personnel and systems. Maintaining these ships with high performance is imperative to ensuring the safety of both material and personnel. This not only prevents damage to the ships themselves but also mitigates the potential for personnel injuries and sea pollution. Organizations, scientific community, and stakeholders have been actively developing advanced systems to monitor data from ship equipment within the scope of maintenance management. This helps in preventing breakdowns and ensures real time knowledge about the equipment's condition. These systems employ various techniques for condition control using algorithms, statistical equations, and other methodologies on the collected data. In this investigation Fuzzy Logic will be applied to data from a selected equipment. For the case study, an air compressor from an ocean patrol vessel has been selected. This air compressor plays a crucial role on Navy ships and has been chosen by the Organization's Maintenance Management Centre for monitoring working hours and operational status.