20505 research outputs found
Sort by
Brief review of vibrothermography and optical thermography for defect quantification in CFRP material
Get PDFThis article belongs to the Special Issue Feature Review Papers in Physical SensorsQuantifying defects in carbon-fiber-reinforced polymer (CFRP) composites is crucial for ensuring quality control and structural integrity. Among non-destructive evaluation techniques, thermography has emerged as a promising solution for defect detection and characterization. This literature review synthesizes current advancements in active thermography methods, with a particular focus on vibrothermography and optical thermography, in identifying defects such as delaminations and BVID in CFRP composites. The review evaluates state-of-the-art techniques, highlighting the advanced applications of optical thermography. It identifies a critical research gap in the integration of vibrothermography with advanced image-processing methods, such as computer vision, which is more commonly applied in optical thermography. Addressing this gap holds significant potential to enhance defect quantification accuracy, improve maintenance practices, and ensure the safety of composite structures.This research was funded by Lembaga Pengelola Dana Pendidikan (LPDP) of the Ministry of Finance of Indonesia grant number 20210222226064. H.F. is grateful for the support provided by the Brazilian National Council for Scientific and Technological Development (CNPq) through grant number 312530/2023-4.Sensor
Effect of printing parameters on the dynamic characteristics of additively manufactured ABS beams: an experimental modal analysis and response surface methodology
Get PDFThis study investigates the dynamic characteristics of three-dimensional (3D) printed acrylonitrile butadiene styrene (ABS) cantilever beams using Experimental Modal Analysis (EMA). The effects of Fused Deposition Modelling (FDM) process parameters—specifically infill pattern, infill density, nozzle size, and raster angle—on the natural frequency, mode shapes, and damping ratio were examined. Although numerous studies have addressed the static mechanical behaviour of FDM parts, there remains a significant gap in understanding how internal structural features and porosity influence their vibrational response. To address this, a total of seventy-two specimens were fabricated with varying parameter combinations, and their dynamic responses were evaluated through frequency response functions (FRFs) obtained via the impact hammer test. Damping characteristics were extracted using the peak-picking (half power) method. Additionally, the influence of internal porosity on damping behaviour was assessed by comparing the actual and theoretical masses of the specimens. The findings indicate that both natural frequencies and damping ratios are strongly influenced by the internal structure of the printed components. In particular, gyroid and cubic infill patterns increased structural stiffness and resulted in higher resonant frequencies, while low infill densities and triangle patterns contributed to enhanced damping capacity. Response Surface Methodology (RSM) was employed to develop mathematical models describing the parameter effects, providing predictive tools for applications sensitive to vibration. The high R² values obtained in the RSM models based on the input variables show that these variables explain the effects of these variables on both natural frequency and damping ratio with high accuracy. The models developed (with R2 values up to 0.98) enable the prediction of modal behaviour, providing a valuable design tool for engineers optimizing vibration-sensitive components in fields such as aerospace, automotive, and electronics.Polymer
Unveiling synergies: the mediating role of China's OFDI and intelligent transformation in advancing green total factor productivity
Get PDFThis research focuses on the impact of intelligent transformation on green total factor productivity (GTFP) within China's manufacturing enterprises from 2010 to 2022. To assess the extent of intelligent transformation across the sector, the study adopts natural language processing and the non‐radial SBM‐ML index to measure GTFP. Utilizing threshold and mediation analysis, we explore how outward foreign direct investment (OFDI) affects intelligent transformation and GTFP while additional robustness tests supplement the analysis. The study also considers the variability in the impact of intelligent transformation at enterprise, industry, and regional levels. The results reveal a significant positive relationship between intelligent transformation and GTFP, notably stronger in non‐OFDI firms and more pronounced in heavily polluting industries. In particular, enterprises with smaller scale of FDI and those in China's middle region see more substantial benefits.Thunderbird International Business Revie
Safety voice concept clean-up: examining the voice that challenges us to be safer
Get PDFSafety voice, the act of speaking up about safety concerns, is essential for preventing accidents and fostering an engaged safety culture. This study systematically reviewed 86 empirical studies of safety voice by operationalising and applying Podsakoff et al.’s (2016) four-stage framework for developing good conceptual definitions to assess its conceptual clarity, triggers, contextual variations, and measurement. This identified opportunities to refine definitions, theory, and hazard categorisation to enable proactive risk management. Current research on communication scope, directionality, and dyadic sender-receiver dynamics is fragmented which limits potential insights. Contextual disparities and Western culture biases affect generalisability. While senior leadership is key to a positive safety culture this focus is lacking. Addressing these areas through improved conceptual frameworks, hazard-voice models, and cross-industry comparisons will enhance proactive safety management, engagement, and resilience in high-risk industries.Safety Scienc
Explainable artificial intelligence for time series modelling and causal inference
Get PDFDue to the concern over the trust issue of black-box artificial intelligence (AI) in daily
human life, “right to an explanation” requirement is hence proposed for any algorithm in
Europe general data protection regulation (GDPR). However, the framework of explain-
able AI (XAI) is still in its infancy. Thus, this thesis aims to propose a framework of
explainable AI for time series (X-AI4TS) in neural modelling and causal inference.
In order to achieve the trustworthy usage of algorithm, at the first stage, this thesis
embed the priori knowledge in time series and break it down into neuron-based dynamic
system with domain expertise for explainable modelling. Then secondly, this thesis ad-
dress the explainabilty in causal inference steps of the neuron-based model — (i) Identifying the “hard to learn” features for neurons; (ii) Revealing and visualizing the inner
operations within neurons; (iii) Inferring the original posteriori multimodal distribution
via neurons.
In addition, the aforementioned methodologies are formulated into an X-AI4TS frame-
work and applied to a real business case in semiconductor fabrication for an integrated
implementation. A data-driven neuron-based time-series forecasting digital twin is generated, under my proposed framework, for a dynamic manufacturing system. The results
offer the diverse explainability to different users (i.e., end-users, data scientist and AI
expert) and show the landing potential in dynamic engineering time series application.PhD in Aerospac
Multi-modal analysis of barely visible impact damage in carbon fibre composites through the fusion of pulsed thermography and phased array ultrasonic testing
No full textThe increasing use of composite materials in modern aircraft structures has necessitated more efficient and reliable inspection techniques to ensure structural integrity and operational safety. Barely visible impact damage (BVID) poses a significant challenge in composite maintenance due to its subtle nature, requiring advanced non-destructive testing and evaluation (NDT&E) methods for accurate detection and characterisation. This study explores a multi-modal inspection approach that integrates phased array ultrasonic testing (PAUT) and pulsed thermography (PT) to enhance BVID detection in carbon fibre-reinforced polymer (CFRP) composites. By leveraging complementary fusion strategies, the proposed framework improves defect localisation beyond the limitations of individual techniques. The results demonstrate that fusion increased PAUT-detected sizes by up to 7% for thin specimens and 8% for thick ones, while PT-detected sizes improved by as much as 71% and 53%, respectively. These findings highlight the synergistic advantages of multi-modal NDT&E, showcasing its potential to provide complementary defect assessment and reduce uncertainty in damage evaluation. The results of this study contribute to the development of more sophisticated inspection methodologies, which ultimately support more efficient and reliable maintenance strategies in aviation.This research was supported by the Republic of Turkiye’s Ministry of National Education through a Ph.D. scholarship awarded to the first author.Quantitative InfraRed Thermography Journa
Machine learning and artificial intelligence in type 2 diabetes prediction: a comprehensive 33-year bibliometric and literature analysis
Get PDFBackground:
Type 2 Diabetes Mellitus (T2DM) remains a critical global health challenge, necessitating robust predictive models to enable early detection and personalized interventions. This study presents a comprehensive bibliometric and systematic review of 33 years (1991-2024) of research on machine learning (ML) and artificial intelligence (AI) applications in T2DM prediction. It highlights the growing complexity of the field and identifies key trends, methodologies, and research gaps.
Methods:
A systematic methodology guided the literature selection process, starting with keyword identification using Term Frequency-Inverse Document Frequency (TF-IDF) and expert input. Based on these refined keywords, literature was systematically selected using PRISMA guidelines, resulting in a dataset of 2,351 articles from Web of Science and Scopus databases. Bibliometric analysis was performed on the entire selected dataset using tools such as VOSviewer and Bibliometrix, enabling thematic clustering, co-citation analysis, and network visualization. To assess the most impactful literature, a dual-criteria methodology combining relevance and impact scores was applied. Articles were qualitatively assessed on their alignment with T2DM prediction using a four-point relevance scale and quantitatively evaluated based on citation metrics normalized within subject, journal, and publication year. Articles scoring above a predefined threshold were selected for detailed review. The selected literature spans four time periods: 1991–2000, 2001–2010, 2011–2020, and 2021–2024.
Results:
The bibliometric findings reveal exponential growth in publications since 2010, with the USA and UK leading contributions, followed by emerging players like Singapore and India. Key thematic clusters include foundational ML techniques, epidemiological forecasting, predictive modelling, and clinical applications. Ensemble methods (e.g., Random Forest, Gradient Boosting) and deep learning models (e.g., Convolutional Neural Networks) dominate recent advancements. Literature analysis reveals that, early studies primarily used demographic and clinical variables, while recent efforts integrate genetic, lifestyle, and environmental predictors. Additionally, literature analysis highlights advances in integrating real-world datasets, emerging trends like federated learning, and explainability tools such as SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations).
Conclusion:
Future work should address gaps in generalizability, interdisciplinary T2DM prediction research, and psychosocial integration, while also focusing on clinically actionable solutions and real-world applicability to combat the growing diabetes epidemic effectively.Frontiers in Digital Healt
Foraging supply chains: investigating disaster for improved food provisioning
Get PDFDisasters such as COVID-19 and the Russia–Ukraine war are drawing attention to the provisioning of food during crises. The main concern has been quickly establishing a stable food supply. However, climate change and public health concerns are shifting attention to the critical gap in identifying the minimal considerations that would adequately address ecological disaster food provisioning. A meta-ethnography of 16 disasters in 12 different countries is employed to identify the activities and their supporting strategies that provide benefits to existing actors within food networks. Analysis suggests that public health, resilience, and sustainability stand to benefit from the identified practices. A conceptual model of an ecologically embedded minimum viable ecosystem for disaster food provisioning is proposed. Exemplar applications are provided for Tigray, Gaza, and Ukraine. The findings may be applied to disaster settings for the development of policy for culturally sensitive, equitable, and nutritious food provisioning strategies.This research was supported by a Research England grant administered by Universities UK International (project reference number 11155). Sandeep Jagtap acknowledges the support of FORCE (Centre for Food Preparedness and Competitiveness) at Lund University, Sweden.Ambi
Bioactivity screening of selected Moroccan medicinal and aromatic plants, and the chemical basis of the phytotoxicity of caper, Capparis spinosa L.
Get PDFPlant natural products are potential sources of biostimulants that can help plants overcome the effects of stress. The adverse effects of soil salinity on wheat growth necessitate the exploration of alternative sustainable solutions, such as biostimulants from medicinal and aromatic plants, to enhance crop resilience and productivity. This study aimed to screen nine Moroccan medicinal and aromatic plant extracts for their effect on wheat growth under saline and non-saline conditions using a seed soaking treatment, in a completely randomised experiment. Except for Marrubium vulgare leaf and Origanum compactum extracts, which averagely improved root length by 25 % and 14 %, respectively, none of the other extracts had significant positive effects on wheat seedling growth. Capparis spinosa (caper) extracts inhibited wheat emergence and growth, with leaf extracts being more phytotoxic than the stem extracts. The leaf extracts of C. spinosa caused an average reduction of the leaf length, root length, shoot dry weight and root dry weight of the wheat seedlings by 31 %, 21 %, 92 % and 94 %, respectively, compared with the control. Further fractionation of the leaf crude extract and follow-up screening revealed that the phytotoxicity likely resulted from a synergy between compounds in different fractions. Chemical analysis of the most active fraction by UHPLC-MS and NMR revealed loliolide as the major compound, alongside oxylipins and indole alkaloid derivatives. Additionally, a previously undescribed compound, 8-(1H-indol-3-yl-methyl)rutin, was also identified. These compounds potentially contribute to the phytotoxicity. The results of this experiment show that although two extracts enhanced root length, overall biostimulant effects were minimal, with C. spinosa extracts being significantly toxic, indicating the need to prevent their application on wheat.FP05 project (Bioproducts for African Agriculture), which is funded by OCP Morocco. FP05 is a collaboration between Mohammed VI Polytechnic University, Rothamsted Research and Cranfield University.Industrial Crops and Product
Development of air-to-ground connectivity for 5G UAV networks
Get PDFThis thesis explores novel methodologies aimed at addressing critical challenges
in the integration of Unmanned Aerial Vehicles (UAVs) within heterogeneous
networks, particularly in the context of fifth-generation (5G) communication systems. The primary focus revolves around achieving seamless connectivity and
mitigating interference to enhance the performance and reliability of UAV operations. As such, this thesis contributes three novel components across these do-
mains. The first novel component of this thesis investigates the development of
an algorithmic framework grounded in the Received Signal Strength (RSS) criterion for network selection, augmented by the Entropy Weighted Method (EWM)
decision-making process. Through comprehensive simulation-level analyses, the
efficacy of this approach in facilitating seamless handover (HO) between 5G and
long-term evolution (LTE) networks for UAVs is demonstrated. This component
represents a significant contribution to addressing the mobility challenges inherent in heterogeneous networks, thus advancing the state-of-the-art in UAV-5G
communication. However, ensuring uninterrupted connectivity for UAVs operating in remote or dynamic environments remains a significant hurdle. The second component introduces a novel approach to achieving seamless HO for UAVs
transitioning between terrestrial and satellite communication networks, leveraging graph theory to develop a decision-making algorithm aimed at optimising
HO decisions. Evaluation through extensive simulations and comparison with
existing solutions underscores significant improvements in various performance
metrics, such as RSS, Signal-to-Noise Ratio (SNR), throughput, latency, and overall UAV connectivity. The proposed graph-method (GM)-based seamless HO solution represents a pivotal advancement in enabling reliable and uninterrupted
communication for UAVs operating in remote and challenging environments,
thereby advancing the state-of-the-art in UAV technology. The third component
delves into interference mitigation strategies to ensure optimal UAV performance
within 5G networks. A novel deep Q learning (DQL) algorithm is proposed to
address interference from neighbouring 5G base stations (gNBs). By formulating
and solving a Signal-to-Interference and Noise Ratio (SINR) optimisation problem using the DQL algorithm, interference is effectively mitigated, resulting in
improved link performance. This Ph.D. research contributes to the advancement
of knowledge in the field of UAV-5G integration by presenting innovative solutions to key challenges. The findings pave the way for the seamless incorporation
of UAVs into heterogeneous networks, unlocking their full potential across a diverse range of applications, from surveillance to communication infrastructure
maintenance and beyond.Engineering and Physical Sciences Research Council (EPSRC)PhD in Aerospac