Simon Ghyselincks, Valeriia Okhmak, Stefano Zampini
et al.
Abstract Reconstructing the structural geology and mineral composition of the first few kilometers of the Earth's subsurface from sparse or indirect surface observations remains a long‐standing challenge with critical applications in mineral exploration, geohazard assessment, and geotechnical engineering. This inherently ill‐posed problem is often addressed by classical geophysical inversion methods, which typically yield a single maximum‐likelihood model that fails to capture the full range of plausible geology. The adoption of modern deep learning methods has been limited by the lack of large 3D training data sets. We address this gap with StructuralGeo, a geological simulation engine that mimics eons of tectonic, magmatic, and sedimentary processes to generate a virtually limitless supply of realistic synthetic 3D lithological models. Using this data set, we train both unconditional and conditional generative flow‐matching models with a 3D attention U‐Net architecture. The resulting foundation model can reconstruct multiple plausible 3D scenarios from surface topography and sparse borehole data, depicting structures such as layers, faults, folds, and dikes. By sampling many reconstructions from the same observations, we introduce a probabilistic framework for estimating the size and extent of subsurface features. While the realism of the output is bounded by the fidelity of the training data to true geology, this combination of simulation and generative AI functions offers a flexible prior for probabilistic modeling, regional fine‐tuning, and use as an AI‐based regularizer in traditional geophysical inversion workflows.
Geophysics. Cosmic physics, Information technology
Giuliano Bernard, Lukas Budde, Roman Hänggi
et al.
Background: The integration of lean management and Industry 4.0 offers great potential for improving the efficiency and effectiveness of production. However, there is still minimal empirical research on how these two approaches interact in practice, including possible synergies and conflicts. In this paper, we use empirical observations to examine how and why Industry 4.0 technologies can support and enhance the continuous improvement process, a fundamental aspect of lean management. We discuss how these technologies specifically contribute to continuous improvement and their overall impact on operational efficiency. Methods: Our research drew on five case studies, capturing the firsthand experiences of industry professionals—specifically, digitization managers and production managers from manufacturing companies across Europe. We analyzed data using qualitative content analysis, applying a structured approach of coding, summarizing, and cross-case comparison to identify key patterns and insights. Conclusions: Our study has shown that digitalization can contribute to an advanced continuous improvement process in several ways. Through a literature review as well as the analysis of use cases, we were able to develop a new model that describes how I4.0 technologies improve the continuous improvement process, a key lean principle. The model highlights five key areas of impacts: increasing flexibility, transparency, and reliability, as well as improved decision-making and acceptance of change processes.
Abstract General type-2 fuzzy logic systems (GT2 FLSs) based on the $$\alpha$$ α -planes representation of general T2 fuzzy sets (FSs) have become more accessible to FL investigators in recent years. Type reduction (TR) is the most important block for GT2 FLSs. Here the weighted type-reduction algorithms based on the Newton and Cotes quadrature formulas of numerical methods of integration technique are first given, and the searching spaces are divided. Then a type of weighted divided search enhanced Karnik–Mendel (WDEKM) algorithms is shown to complete the centroid TR. In contrast to the WEKM algorithms, four simulation instances show that the WDEKM algorithms get lesser absolute errors and faster calculational speeds, which may offer the potentially application values for applying T2 FLSs.
Electronic computers. Computer science, Information technology
Model-driven engineering (MDE) is believed to have a significant impact in software quality. However, researchers and practitioners may have a hard time locating consolidated evidence on this impact, as the available information is scattered in several different publications. Our goal is to aggregate consolidated findings on quality in MDE, facilitating the work of researchers and practitioners in learning about the coverage and main findings of existing work as well as identifying relatively unexplored niches of research that need further attention. We performed a tertiary study on quality in MDE, in order to gain a better understanding of its most prominent findings and existing challenges, as reported in the literature. We identified 22 systematic literature reviews and mapping studies and the most relevant quality attributes addressed by each of those studies, in the context of MDE. Maintainability is clearly the most often studied and reported quality attribute impacted by MDE. Eighty out of 83 research questions in the selected secondary studies have a structure that is more often associated with mapping existing research than with answering more concrete research questions (e.g., comparing two alternative MDE approaches with respect to their impact on a specific quality attribute). We briefly outline the main contributions of each of the selected literature reviews. In the collected studies, we observed a broad coverage of software product quality, although frequently accompanied by notes on how much more empirical research is needed to further validate existing claims. Relatively, little attention seems to be devoted to the impact of MDE on the quality in use of products developed using MDE.
Reliable aero-engine anomaly detection is crucial for ensuring aircraft safety and operational efficiency. This research explores the application of the Fisher autoencoder as an unsupervised deep learning method for detecting anomalies in aero-engine multivariate sensor data, using a Gaussian mixture as the prior distribution of the latent space. The proposed method aims to minimize the Fisher divergence between the true and the modeled data distribution in order to train an autoencoder that can capture the normal patterns of aero-engine behavior. The Fisher divergence is robust to model uncertainty, meaning it can handle noisy or incomplete data. The Fisher autoencoder also has well-defined latent space regions, which makes it more generalizable and regularized for various types of aero-engines as well as facilitates diagnostic purposes. The proposed approach improves the accuracy of anomaly detection and reduces false alarms. Simulations using the CMAPSS dataset demonstrate the model's efficacy in achieving timely anomaly detection, even in the case of an unbalanced dataset.
Over more than two decades, The University of Glasgow has co-designed and delivered numerous software engineering focused courses with industry partners, covering both technical and discipline specific professional skills. Such collaborations are not unique and many of the benefits are well recognised in the literature. These include enhancing the real-world relevance of curricula, developing student professional networks ahead of graduation and easing recruitment opportunities for employers. However, there is relatively little scholarship on the perspectives of industry practitioners who participate in course design and delivery. This gap is significant, since the effort invested by practitioners is often substantial and may require ongoing support from both the industry partner and academic institution. Understanding the motivations, expectations and experiences of practitioners who engage in course delivery can guide the formation of future partnerships and ensure their long-term sustainability. We begin to address this gap by reporting on the outcomes of a retrospective conducted amongst the practitioner coauthors of this paper, with the academic coauthors acting as facilitators. All coauthors have participated in the recent co-design and delivery of software engineering courses, but we choose to focus explicitly on the perspectives of the practitioners. We report on the themes that emerged from the discussions and our resulting recommendations for future collaborations.
Hybrid welding processes have gained significant attention due to their high efficiency and exceptional welding properties. However, there are still significant technological challenges in achieving consistent quality and suppressing welding defects. To overcome this challenge, researchers have focused on the integration of visual analysis techniques, numerical simulation techniques, and advanced technologies such as artificial intelligence/machine learning (AI/ML) and digital twins. This comprehensive review synthesizes current knowledge on quality monitoring in hybrid welding, encompassing an overview of hybrid welding processes, quality assurance, monitoring techniques, key performance indicators, and advancements in monitoring techniques. Furthermore, the review highlights the integration of sensor data with AI/ML algorithms and digital twin technologies, enhancing the capabilities of quality monitoring systems. Notably, the review emphasizes the incorporation of artificial intelligence (AI) and digital twin technologies into quality monitoring frameworks. Artificial intelligence/Machine learning enables real-time analysis of welding parameters and defect detection, while digital twins offer virtual representations of physical welding processes, facilitating predictive maintenance and optimization. The findings underscore the crucial role of sensor technology, AI/ML, and digital twin integration in enhancing defect detection accuracy, improving welded joint quality, and control in hybrid welding. In addition to improving the quality of welded joints, this integration paves the way for further developments in welding technology.
We study the continual pretraining recipe for scaling language models' context lengths to 128K, with a focus on data engineering. We hypothesize that long context modeling, in particular \textit{the ability to utilize information at arbitrary input locations}, is a capability that is mostly already acquired through large-scale pretraining, and that this capability can be readily extended to contexts substantially longer than seen during training~(e.g., 4K to 128K) through lightweight continual pretraining on appropriate data mixture. We investigate the \textit{quantity} and \textit{quality} of the data for continual pretraining: (1) for quantity, we show that 500 million to 5 billion tokens are enough to enable the model to retrieve information anywhere within the 128K context; (2) for quality, our results equally emphasize \textit{domain balance} and \textit{length upsampling}. Concretely, we find that naively upsampling longer data on certain domains like books, a common practice of existing work, gives suboptimal performance, and that a balanced domain mixture is important. We demonstrate that continual pretraining of the full model on 1B-5B tokens of such data is an effective and affordable strategy for scaling the context length of language models to 128K. Our recipe outperforms strong open-source long-context models and closes the gap to frontier models like GPT-4 128K.
Engineering a sustainable world requires to consider various systems that interact with each other. These systems include ecological systems, economical systems, social systems and tech-nical systems. They are loosely coupled, geographically distributed, evolve permanently and generate emergent behavior. As these are characteristics of systems of systems (SoS), we discuss the engi-neering of a sustainable world from a SoS engineering perspective. We studied SoS engineering in context of a research project, which aims at political recommendations and a research roadmap for engineering dynamic SoS. The project included an exhaustive literature review, interviews and work-shops with representatives from industry and academia from different application domains. Based on these results and observations, we will discuss how suitable the current state-of-the-art in SoS engi-neering is in order to engineer sustainability. Sustainability was a major driver for SoS engineering in all domains, but we argue that the current scope of SoS engineering is too limited in order to engineer sustainability. Further, we argue that mastering dynamics in this larger scope is essential to engineer sustainability and that this is accompanied by dynamic adaptation of technological SoS.
Active Learning (AL) is a well-known teaching method in engineering because it allows to foster learning and critical thinking of the students by employing debate, hands-on activities, and experimentation. However, most educational results of this instructional method have been achieved in face-to-face educational settings and less has been said about how to promote AL and experimentation for online engineering education. Then, the main aim of this study was to create an AL methodology to learn electronics, physical computing (PhyC), programming, and basic robotics in engineering through hands-on activities and active experimentation in online environments. N=56 students of two engineering programs (Technology in Electronics and Industrial Engineering) participated in the methodology that was conceived using the guidelines of the Integrated Course Design Model (ICDM) and in some courses combining mobile and online learning with an Android app. The methodology gathered three main components: (1) In-home laboratories performed through low-cost hardware devices, (2) Student-created videos and blogs to evidence the development of skills, and (3) Teacher support and feedback. Data in the courses were collected through surveys, evaluation rubrics, semi-structured interviews, and students grades and were analyzed through a mixed approach. The outcomes indicate a good perception of the PhyC and programming activities by the students and suggest that these influence motivation, self-efficacy, reduction of anxiety, and improvement of academic performance in the courses. The methodology and previous results can be useful for researchers and practitioners interested in developing AL methodologies or strategies in engineering with online, mobile, or blended learning modalities.
Alexandr Oblizanov, Natalya Shevskaya, Anatoliy Kazak
et al.
In recent years, artificial intelligence technologies have been developing more and more rapidly, and a lot of research is aimed at solving the problem of explainable artificial intelligence. Various XAI methods are being developed to allow the user to understand the logic of how machine learning models work, and in order to compare the methods, it is necessary to evaluate them. The paper analyzes various approaches to the evaluation of XAI methods, defines the requirements for the evaluation system and suggests metrics to determine the various technical characteristics of the methods. A study was conducted, using these metrics, which determined the degradation in the explanation quality of the SHAP and LIME methods with increasing correlation in the input data. Recommendations are also given for further research in the field of practical implementation of metrics, expanding the scope of their use.
Digital engineering transformation is a crucial process for the engineering paradigm shifts in the fourth industrial revolution (4IR), and artificial intelligence (AI) is a critical enabling technology in digital engineering transformation. This article discusses the following research questions: What are the fundamental changes in the 4IR? More specifically, what are the fundamental changes in engineering? What is digital engineering? What are the main uncertainties there? What is trustworthy AI? Why is it important today? What are emerging engineering paradigm shifts in the 4IR? What is the relationship between the data-intensive paradigm and digital engineering transformation? What should we do for digitalization? From investigating the pattern of industrial revolutions, this article argues that ubiquitous machine intelligence (uMI) is the defining power brought by the 4IR. Digitalization is a condition to leverage ubiquitous machine intelligence. Digital engineering transformation towards Industry 4.0 has three essential building blocks: digitalization of engineering, leveraging ubiquitous machine intelligence, and building digital trust and security. The engineering design community at large is facing an excellent opportunity to bring the new capabilities of ubiquitous machine intelligence and trustworthy AI principles, as well as digital trust, together in various engineering systems design to ensure the trustworthiness of systems in Industry 4.0.
Claudemir Miranda Barboza, Gladys Denise Wielewski
Este artigo apresenta um breve panorama da produção acadêmica que abarca as tecnologias digitais na perspectiva da formação inicial do professor de Matemática entre os anos de 2011 a 2021 e para isso buscou-se na Biblioteca Digital de teses e dissertações brasileira, usando o descritor “Tecnologias Digitais” AND “Formação Inicial do Professor de Matemática”. A pesquisa teve uma abordagem qualitativa por entender que esse tipo de pesquisa fornece informações mais descritivas. O trabalho buscou compreender como as tecnologias digitais contribuem para a formação inicial do professor de matemática, por meio de uma breve abordagem da utilização das tecnologias digitais na perspectiva dos conhecimentos do professor de Matemática. A discussão dos conhecimentos do professor de Matemática se respaldou nos aportes teóricos de Shulman (1986), Fiorentini (2003), Blanco (2003), Tardif (2014) e Carrillo (2014, 2018). Os resultados apontam para as potencialidades das tecnologias digitais na formação do professor e como suporte para evidenciar os conhecimentos de conteúdo matemático e os conhecimentos pedagógicos para o ensino da Matemática, a demonstram que o software GeoGebra está bem inserido nos debates de tecnologias em educação matemática, mostram também fragilidade na formação inicial do professor nesse aspecto.
Special aspects of education, Applied mathematics. Quantitative methods
Context: Processing Software Requirement Specifications (SRS) manually takes a much longer time for requirement analysts in software engineering. Researchers have been working on making an automatic approach to ease this task. Most of the existing approaches require some intervention from an analyst or are challenging to use. Some automatic and semi-automatic approaches were developed based on heuristic rules or machine learning algorithms. However, there are various constraints to the existing approaches of UML generation, such as restriction on ambiguity, length or structure, anaphora, incompleteness, atomicity of input text, requirements of domain ontology, etc. Objective: This study aims to better understand the effectiveness of existing systems and provide a conceptual framework with further improvement guidelines. Method: We performed a systematic literature review (SLR). We conducted our study selection into two phases and selected 70 papers. We conducted quantitative and qualitative analyses by manually extracting information, cross-checking, and validating our findings. Result: We described the existing approaches and revealed the issues observed in these works. We identified and clustered both the limitations and benefits of selected articles. Conclusion: This research upholds the necessity of a common dataset and evaluation framework to extend the research consistently. It also describes the significance of natural language processing obstacles researchers face. In addition, it creates a path forward for future research.
Models are fundamentally crucial to many scientific fields, including software engineering, systems engineering, enterprise modeling, and business modeling. This paper focuses on diagrammatic conceptual modeling, as opposed to mathematical or computational models, wherein a conceptual model is a translation of reality processes into an abstract mechanism that has similar structure and parallel events of the external processes. Although various modeling approaches exist, including UML (Unified Modeling Language) in software engineering and its dialect, SysML (System Modeling Language), in systems engineering, several difficulties arise in such models, including the problem of model multiplicity that is related to the lack an integrated view of structure and behavior. This paper generalizes conceptual modeling to be applied in organizations at large. According to authorities, the so-called organization theory portrays organizations as machine-like systems. As a machine, an organization coordinates its parts to transform inputs into outputs. Therefore, we synthesize the notion of an organization as a machine and apply a new modeling methodology called thinging machine (TM) to real engineering operations. The results show the viability of the TM methodology serving as a foundation for high-level modelling of systems.
Sascha El-Sharkawy, Dhar Saura Jyoti, Adam Krafczyk
et al.
Ideally, a variability model is a correct and complete representation of product line features and constraints among them. Together with a mapping between features and code, this ensures that only valid products can be configured and derived. However, in practice the modeled constraints might be neither complete nor correct, which causes problems in the configuration and product derivation phases. This paper presents an approach to reverse engineer variability constraints from the implementation, and thus improve the correctness and completeness of variability models. We extended the concept of feature effect analysis to extract variability constraints from code artifacts of the Bosch PS-EC large-scale product line. We present an industrial application of the approach and discuss its required modifications to handle non-Boolean variability and heterogeneous artifact types.
This doctoral dissertation proposes a novel approach to enhance the development of smart services for the Internet of Things (IoT) and smart Cyber-Physical Systems (CPS). The proposed approach offers abstraction and automation to the software engineering processes, as well as the Data Analytics (DA) and Machine Learning (ML) practices. This is realized in an integrated and seamless manner. We implement and validate the proposed approach by extending an open source modeling tool, called ThingML. ThingML is a domain-specific language and modeling tool with code generation for the IoT/CPS domain. Neither ThingML nor any other IoT/CPS modeling tool supports DA/ML at the modeling level. Therefore, as the primary contribution of the doctoral dissertation, we add the necessary syntax and semantics concerning DA/ML methods and techniques to the modeling language of ThingML. Moreover, we support the APIs of several ML libraries and frameworks for the automated generation of the source code of the target software in Python and Java. Our approach enables platform-independent, as well as platform-specific models. Further, we assist in carrying out semiautomated DA/ML tasks by offering Automated ML (AutoML), in the background (in expert mode), and through model-checking constraints and hints at design-time. Finally, we consider three use case scenarios from the domains of network security, smart energy systems and energy exchange markets.