63477 research outputs found
Sort by
ONDS: Optimum Node and Data Selection From Constrained IoT for Efficient Online Learning
No full textSupervised Machine Learning (ML) models require large amounts of labeled data for training. However, this becomes challenging when dealing with resource- and network-constrained Internet of Things (IoT) devices that collect data. Furthermore, in scenarios where the acquired data is fast-changing and highly temporal, continuous and online learning becomes necessary. In this paper, we address the problem of efficiently training ML models using data from IoT nodes. We specifically focus on two aspects: i) selecting the nodes that provide data for the re/training, and ii) determining the optimal amounts of data to be acquired from these nodes, considering network and time constraints, while minimizing learning errors. To tackle this optimization problem, we propose ONDS: an Optimum Node and Data Selection algorithm with linear complexity in the worst-case. ONDS offers a model-agnostic solution applicable to different data modalities and ML architectures. To evaluate the performance of ONDS, we conduct experiments using various models and real-world datasets. The results demonstrate the effectiveness of ONDS, as it outperforms existing alternatives in both classification and regression tasks.This work was supported by the PDRA under Grant PDRA7-0410-21004 from the Qatar National Research Fund (a member of Qatar Foundation).Scopu
A Robust Ensemble Model for Plant Disease Detection Using Deep Learning Architectures
No full textThis study explores advanced methods for plant disease classification by integrating pre-trained deep learning models and leveraging ensemble learning. After a comprehensive review of deep learning methods in this area, the InceptionResNetV2, MobileNetV2, and EfficientNetB3 architectures were identified as promising candidates, as they have been shown to achieve high accuracy and efficiency in various applications. The proposed approach strategically combines these architectures to leverage their unique strengths: the advanced feature extraction capabilities of InceptionResNetV2, the lightweight and efficient design of MobileNetV2, and the scalable, performance-optimized structure of EfficientNetB3. By integrating these models, the approach aims to improve classification accuracy and robustness and overcome the multiple challenges of plant disease detection. Comprehensive experiments were conducted on three datasets-PlantVillage, PlantDoc, and FieldPlant-representing a mix of laboratory and real-world conditions. Advanced data augmentation techniques were employed to improve model generalization, while a systematic ablation study validated the efficacy of key architectural choices. The ensemble model achieved state-of-the-art performance, with classification accuracies of 99.69% on PlantVillage, 60% on PlantDoc, and 83% on FieldPlant. These findings highlight the potential of ensemble learning and transfer learning in advancing plant disease detection, offering a robust solution for real-world agricultural applications.The research reported in this publication was supported by the Qatar Research Development and Innovation Council [ARG01-0513-230141]. The Qatar National Library provides Open Access funding.Scopu
Empirical analysis of car-following behavior: Impacts of driver demographics, leading vehicle types, and speed limits on driver behavior and safety
No full textCar-following behavior is the most fundamental and common driving behavior and is crucial for road safety and traffic efficiency. Traffic flow dynamics are greatly affected by this behavior, and driver-related factors in car-following behavior have been identified as a key cause of rear-end crashes. Despite extensive research on car-following behavior, a gap remains in understanding how this behavior manifests itself in culturally diverse driver populations and heterogeneous driving conditions. The aim of this study was to empirically investigate the effects of individual characteristics, leading vehicle types, posted speed limits and deceleration rates of the leading vehicle on car-following behavior. To this end, the car-following behavior of 61 participants was investigated in eight different scenarios involving a motorbike, sedan, SUV, and truck as the leading vehicle under two different speed limits, i.e., 50 km/h and 80 km/h in a driving simulator environment. The results showed that considerable variations in car-following behaviors exist depending on gender, age, driving experience, educational levels, and the type of leading vehicle, as well as speed limits and deceleration rates. The risk of rear-end crash was found to be higher at the lower speed limit and with a two-wheeler (motorbike) as the leading vehicle. Additionally, females were identified as a having higher crash risk than males. In summary, this study provides valuable insights that could be applied to enhance road safety, such as tailoring targeted training materials for high-risk groups and informing policy decisions. Incorporating these insights into model calibration can lead to more accurate and realistic simulations that capture the complexities of real-world driving scenarios.Research reported in this publication was supported by the Qatar Research Development and Innovation Council [UREP29-029-5-002]. Open Access funding provided by the Qatar National Library. The content is solely the responsibility of the authors and does not necessarily represent the official views of Qatar Research Development and Innovation Council
Cooperative offloading multi-access edge computing (COMEC) for cell-edge users in heterogeneous dense networks
No full textMulti-access edge computing (MEC) addresses the rising computational demands of advanced applications by bringing processing closer to users. Yet, cell-edge users often face high latency and low throughput—challenges that can be mitigated by deploying multiple MEC servers for simultaneous task offloading in dense heterogeneous networks. This paper investigates the performance gains of collaborative computing and presents a novel Cooperative Offloading Multi-access Edge Computing (COMEC) scheme. The COMEC aims to optimize resource allocation for cell-edge users by reducing latency and maximizing energy efficiency (EE). In this way, cell-edge users with limited battery and computational power can sustain low-latency applications for a longer time. A bi-objective optimization problem is formulated to maximize the EE of edge users while simultaneously minimizing the latency. We propose an iterative algorithm named ORA-ETO to solve the mixed integer non-linear fractional (MINLF) problem. The proposed scheme has been evaluated using both the Rayleigh and WINNER-II propagation models within an asymmetric cell configuration. The obtained results validate the efficacy of the proposed COMEC scheme for cell-edge users, achieving performance gains of over 55% compared to dense multi-server-assisted MEC and CoMP-assisted MEC architectures. The COMEC scheme is statistically more significant (p<0.01) and has more stable performance with standard deviation less than 0.082 kbps/J, making it a superior choice for cell edge users. The effect size (η2=0.71) confirms that the choice of scheme has a considerable impact on EE of cell-edge users
Shipping economic forecasting: recent developments, applications, and future directions
No full textForecasting is vital in shipping economics and directly affects the business decisions of shipping companies and the quality development of the shipping markets. This study critically reviews variables, methods, and results used for shipping economic forecasting. This study provides an extensive review of the development of the shipping market forecasting models, which can be broadly categorised into artificial intelligence and classical economic models. Our review identifies forecasting applications in the following areas: freight markets, newbuilding and second-hand ship markets, and ship-demolition markets. We review the evolution of the forecasting methods over time and distinguish six types of feature engineering (i.e. the process of preparing and transforming input data) that improve model generalisation performance (i.e. ability for the model to work outside training data) in the existing literature. We further discuss the improvement, input determination, evaluation metrics, and hyper-parameter optimisation of models. Our analysis shows that support vector regression and artificial neural networks are the commonly used techniques; Grid search and evolutionary optimisation are popular for hyperparameter optimisation in current research. Finally, we discuss the achievements and limitations of the existing literature. The survey concludes with the identification of existing gaps and recommendations for future research
A multi-faceted approach for leveraging AI and intellectual capital for enhanced supply chain decision-making
No full textPurpose: This study aims to investigate the mediating role of different types of intellectual capital (human, structural and relational) in the relationship between artificial intelligence-driven supply chain analytics capability (AI-SCAC) and various supply chain decision-making processes, specifically rational, bounded and tacit decision-making. Design/methodology/approach: The study used a quantitative survey strategy to collect the data. A total of 320 valid questionnaires were received from manufacturing companies. The data were analysed using structural equation modeling with partial least squares (PLS-SEM) approach through SmartPLS software. Findings: The results indicate that human and structural capital significantly mediate the relationship between AI-SCAC and rational and bounded decision-making processes. However, structural capital does not mediate the relationship between AI-SCAC and the tacit decision-making process. Moreover, relational capital does not show a significant mediating effect on all of the decision-making processes. Notably, structural capital has the strongest impact on rational and bounded decision-making, while human capital plays a critical role across all three decision-making processes, including tacit decision-making. Originality/value: This study contributes to the literature by providing a nuanced understanding of the differentiated impact of intellectual capital components on various decision-making processes within the context of AI-SCAC. While previous studies have broadly acknowledged the role of intellectual capital in decision-making, this research provides more understanding of how specific types of intellectual capital interact with AI to influence distinct decision-making processes. Notably, the differential impact of structural capital on rational and bounded decision-making versus tacit decision-making highlights the need for organisations to adopt a more tailored approach in leveraging their intellectual capital.Scopu
Lipid Subclasses Differentiate Insulin Resistance by Triglyceride-Glucose Index
No full textBackground: Insulin resistance is a key driver of metabolic syndrome and related disorders, yet its underlying metabolic alterations remain incompletely understood. The Triglyceride-Glucose (TyG) index is an emerging, accessible marker for insulin resistance, with growing evidence supporting its clinical utility. This study aimed to characterize the metabolic profiles associated with insulin resistance using the TyG index in a large, population-based cohort, and to identify metabolic pathways potentially implicated in insulin resistance. Methods: Here, we conducted a cross-sectional study using data from the Qatar Biobank, including 1255 participants without diabetes classified as insulin-sensitive or insulin-resistant based on TyG index tertiles. Untargeted serum metabolomics profiling was performed using high-resolution mass spectrometry. Our statistical analyses included orthogonal partial least squares discriminate analysis and linear models. Results: Distinct metabolic signatures differentiated insulin-resistant from insulin-sensitive participants. Phosphatidylethanolamines, phosphatidylinositols, and phosphatidylcholines, were strongly associated with insulin resistance, while plasmalogens and sphingomyelins were consistently linked to insulin sensitivity. Conclusions: Lipid-centric pathways emerge as potential biomarkers and therapeutic targets for the early detection and personalized management of insulin resistance and related metabolic disorders. Longitudinal studies are warranted to validate causal relationships.This research was funded by the Qatar Research Development and Innovation (QRDI) Council, grant number ARG01-0420-230007.Scopu
Performance evaluation of 3D-printed PLA composites doped with WE43 magnesium alloy for bone tissue engineering applications
No full textThis study presents a comprehensive performance evaluation of 3D-printed PLA/WE43 magnesium alloy composites, offering novel insights into composition–structure–function relationships for bone tissue engineering. PLA and WE43 composite filaments, containing various magnesium concentrations (5 %, 10 %, and 15 %), were produced using solvent evaporation and extrusion methods. The 3D-printed scaffolds were assessed for mechanical strength, porosity, biodegradability, bioactivity, and biocompatibility. Incorporating magnesium into PLA for 3D printing significantly affected the composites' dimensional stability and formation quality. While a low Mg content (5 %) only slightly impacted the print quality and dimensions, higher Mg concentrations (10 % and 15 %) led to increased weight, rougher surfaces, dimensional shrinkage in height, and overall poorer formation quality. Adding WE43 alloy to PLA decreased the average pore sizes of the composites. The results from the compression study showed that increasing magnesium content showed improved mechanical properties, with 15 % WE43 showing the highest strength value of 740.6 MPa and elastic modulus of 6.45 GPa. Also, significant calcium phosphate deposition was observed in the bioactivity study and higher degradation was observed for higher magnesium content scaffolds. In vitro studies revealed that the 10 % WE43 scaffolds showed good interaction with cells, forming clusters and higher viability. These findings suggest that 3DP of 10 % PLA/WE43 composites offers a promising potential for bone tissue engineering, balancing print quality, mechanical strength, bioactivity, and biodegradability.The authors acknowledge the partial support by QNRF and HMC through project NPRP13S-0126-200172 (Additive Manufacturing of Mg-based Porous Tissue Scaffolds)
Improving corrosion inhibition of steel using polyurethane based composite coatings by incorporating zirconia nanoparticles and novel urea-based inhibitor
No full textCorrosion of steel in aggressive environments poses a major threat to structural integrity and operational efficiency across various industries. To address this, the present study introduces a novel polyurethane (PU)-based composite coating reinforced with zirconia (ZrO₂) nanoparticles integrated with an in-house synthesized urea-based organic corrosion inhibitor, 1,1′-(methylenebis(4,1-phenylene))bis(3-(pyridin-2-ylmethyl)urea) (MS31). This multifunctional coating system leverages the mechanical durability of PU, the barrier-enhancing role of ZrO₂ nanoparticles, and the smart, pH-responsive release characteristics of MS31 to provide active and passive corrosion protection. The composite coatings were applied to steel substrates and evaluated under simulated saline conditions (3.5 wt% NaCl). Characterization techniques including FTIR, TGA, UV–Vis spectroscopy, and contact angle measurements confirmed the successful integration of MS31, thermal stability up to 800 °C, pH-triggered inhibitor release, and hydrophobic surface behavior. Adhesion testing further demonstrated improved mechanical interlocking with the steel surface. Electrochemical analyses, including Tafel polarization and electrochemical impedance spectroscopy (EIS), revealed significantly enhanced corrosion protection relative to unmodified PU coatings, while salt spray testing validated long-term performance under corrosive exposure. This work demonstrates a robust and environmentally adaptive coating system that combines physical barrier properties with active corrosion inhibition. The integration of MS31-loaded ZrO₂ nanoparticles into the PU matrix not only improves the structural and protective features of the coating but also introduces a smart, responsive mechanism for corrosion control. These findings offer a promising pathway toward the development of sustainable, high-performance coatings for steel protection in harsh environments.This research was funded by the Qatar National Research Fund (a member of the Qatar Foundation), grant number ARG01-0516-230189, Qatar University internal grant QUHI-CENG-24/25–436 and QUCGCAS-24/25-386
Lean Six Sigma, ISO 9001, and organizational performance: An integrated approach
No full textOrganizations endeavor to enhance their market positions and gain competitive advantages by either continuously improving or radically transforming their systems and processes. Despite the worldwide popularity of the ISO 9001 as a quality management platform designed to meet the demands of customers and stakeholders while adhering statutory and regulatory requirements, they still exhibits various business issues and shortfalls. Applying the theories of process management and quality management, this study seeks to develop an effective framework that addresses these challenges by integrating Lean Six Sigma (LSS) with ISO 9001 standards. Our findings demonstrate that the implementation of LSS can overcome many of the ISO 9001 shortfalls and significantly improve process efficiency, reduces waste, promote continuous improvement, and enhances customer satisfaction. Furthermore, our research shows how LSS complements ISO 9001 by fostering positive organizational change and innovation, contributing to the development of a more sustainable business environment. Through two empirical case studies, this paper illustrates the tangible impact of LSS in addressing ISO 9001 shortfalls and boosting overall organizational performance. This study makes several contributions by offering practical insights into how the integration of LSS with ISO 9001 can advance both the theory and practice of quality management, ultimately driving business excellence.Scopu