Science, Engineering and Technology
Not a member yet
105 research outputs found
Sort by
Investigation of the Properties of Waste Expanded Polystyrene (EPS) Modified Bitumen
Get PDFThe "white pollution" caused by the indiscriminate disposal of waste-expanded polystyrene (EPS) packaging foams in Nigeria not only poses a serious environmental threat but also results in significant waste of resources. Again, the poor waste management practices adopted in most of the urban and rural areas in Nigeria also compounded this situation, coupled with the known fact that polymeric materials generally do not decompose (i.e., are non-biodegradable) easily, thereby making them serve as a threat to the surrounding environment. Given resolving this problem, these discarded EPS packaging foams were used in the modification of 60/70 bitumen in this study, as it has been reported that polymer modifiers are known as a possible solution for improving road life in the face of increasing and heavy traffic loading when suitably incorporated into the virgin bitumen. The binder physical tests were carried out on the virgin bitumen and the hot mix waste EPS-modified bitumen samples at modifier contents of 2.5, 5.0, 7.5, 10.0, 12.5, and 15.0%. The results from the tests showed that the best-improved properties were achieved at 5.0 % EPS modifier content with increased softening point (53°C), flash point (275°C), and decreased penetration (50 dmm), while its computed penetration index (-0.48) value indicated that this blend falls within the category of the most acceptable road bitumen. Waste EPS is a suitable bitumen modifier, which enhanced its performance and could equally solve the problem of “white pollution” in Nigeria
Malware Detection Using a Random Forest Method Trained on a Balanced Synthetic Dataset
Get PDFThe accuracy of malware detection is closely related to the available datasets, which are often small and imbalanced. To overcome these challenges, this study proposed a new method that creates synthetic malware data and increases the size and balance by generating several data sets with a flow-based model. Subsequently, a random forest classifier is fitted on this augmented dataset. This study aimed to analyze the generation of synthetic data based on flow-based models and the impact of synthetic data generation on the performance of a random forest for malware detection. A flow-based model was used to generate a balanced synthetic dataset based on the CICMalDroid2020 dataset. The generated data was used for feature selection and engineering to optimize the Random Forest model. The experimental results demonstrate the effectiveness of the proposed approach. The flow-based model generated an additional 13,402 samples, massively increasing the dataset size, even though the original dataset had only 11,598 data entries. After training on the synthetic augmented dataset, the Random Forest model achieved better performance compared to the original dataset evaluation with metrics precision (93%), recall (100%), balanced precision (96%), and the F1 score (91%). The results show that flow-based model-generated synthetic data can significantly enhance malware detection capabilities
Spare Parts and Material Management in Electric Vehicle Maintenance: A Multidisciplinary Review
Get PDFMaintenance and management of spare parts for electric vehicles requires a specific approach due to high-voltage components, technological complexity, and limited availability of specialized parts. Key challenges include optimizing inventory, monitoring the life of battery systems, ensuring compatibility of parts across different vehicle models, and organizing efficient procurement, storage, and distribution. Of particular importance are safety protocols when working with high voltages, the use of specialized technical and protective materials, and environmentally friendly waste management and battery recycling. Digital solutions, including ERP and CMMS systems, IoT sensors, remote diagnostics, and predictive analytics based on artificial intelligence, enable better maintenance planning, cost reduction, and increased vehicle reliability. The paper provides an analysis of the technical, safety, environmental, and economic aspects of electric vehicle maintenance, with an emphasis on the digitalization and optimization of logistics processes. Based on the analysis, recommendations are proposed for improving the maintenance system, including the integration of advanced technologies, process standardization, and strengthening the education of service personnel. In addition, the paper identifies key research questions and outlines directions for future research—particularly in areas such as the digital integration of spare parts logistics, the environmental impact of material uses and disposal, and the role of artificial intelligence in predictive maintenance strategies. The findings suggest that the greatest improvement potential lies in combining predictive maintenance, sustainable practices, and operational cost reduction, thereby contributing to the long-term reliability and competitiveness of electric mobility
Suitability Assessment of Organic Carbon Additives in the Carburization of Low Carbon Steel (AISI 1020) for Engineering Applications
Get PDFThe quest to enhance the mechanical properties of low-carbon steel (LCS) has stimulated the exploration of diverse carburization techniques, with growing attention on organic additives derived from agricultural wastes as sustainable alternatives to conventional materials. This study investigated sheanut shell (SNS) and eggshell (ES) ash as eco-friendly carburizing agents for AISI 1020 steel to improve its performance for engineering applications. The objective was to evaluate their potential in enhancing hardness, strength, impact resistance, and microstructural properties of LCS. Experimental analysis compared carburized and un-carburized (UC) samples, focusing on hardness, tensile strength, impact energy, and microstructural features. The findings showed that carburization significantly increased hardness, with carburized LCS reaching 513 HB compared to 398 HB for UC LCS. However, UC LCS exhibited higher yield strength (221.3 N/mm²) and ultimate tensile strength (241.1 N/mm²), whereas carburized LCS absorbed more fracture energy (63.72 J), reflecting a trade-off between hardness and tensile strength. Microstructural examination revealed improved surface morphology, metallurgical bonding, and higher pearlite concentration due to carbon diffusion, while energy dispersive spectroscopy confirmed elevated carbon content in carburized samples. Structural analysis further identified both crystalline and amorphous carbon phases. The study concludes that SNS and ES ash are effective sustainable carburizing additives capable of enhancing surface properties of LCS, making the material suitable for high-strength and wear-resistant applications. It recommends the wider adoption of these agro-waste additives in industrial carburization processes to reduce reliance on costly conventional materials while promoting sustainable engineering practices
Fine-tuning AraGPT2 for Hierarchical Arabic Text Classification
Get PDFText classification consists in attributing a text to its corresponding category. It is a crucial task in natural language processing (NLP), with applications spanning content recommendation, spam detection, sentiment analysis, and topic categorization. While significant advancements have been made in text classification for widely spoken languages, Arabic remains underrepresented despite its large and diverse speaker base. Another challenge is that, unlike flat classification, hierarchical text classification involves categorizing texts into a multi-level taxonomy. This adds layers of complexity, particularly in distinguishing between closely related categories within the same super-class. To tackle these challenges, we propose a novel approach using AraGPT2, a variant of the Generative Pre-trained Transformer 2 (GPT-2) model adapted specifically for Arabic. Fine-tuning AraGPT2 for hierarchical text classification leverages the model\u27s pre-existing linguistic knowledge and adapts it to recognize and classify Arabic text according to hierarchical structures. Fine-tuning, in this context, refers to the process of training a pre-trained model on a specific task or dataset to improve its performance on that task. Our experiments and comparative study demonstrate the efficiency of our solution. The fine-tuned AraGPT2 classifier achieves a hierarchical HF score of 80.64%, outperforming the machine learning-based classifier, which scores 41.90%
Integrated Experimental and Analytical Investigation of Reverse Osmosis Desalination Systems
Get PDFWater scarcity is a growing global issue, necessitating innovative and sustainable solutions for freshwater generation. Among the available technologies, reverse osmosis (RO) has become the primary method for seawater desalination due to its effective salt rejection and high energy efficiency. This study presents an integrated experimental and analytical investigation of a full-scale seawater reverse osmosis (SWRO) plant with a 2 MLD capacity at the Victoria and Alfred (V&A) Waterfront in Cape Town, South Africa. Operational data were collected over six months, including feedwater temperature (13.66 –16.78 °C), pressure (50–60 bar), total dissolved solids (32,883–38,387 mg/L), and pH (6.19–7.89). The plant consistently produced high-quality permeate with TDS around 500 mg/L, achieving a 31% recovery rate at an average energy consumption of 3 kWh/m³. Machine learning models, specifically multiple linear regression and decision trees, were used to predict RO performance and to explore the relationships between operational parameters. Results show that higher feed pressure improves permeate flux but raises energy use, increased feedwater temperature boosts flux and slightly reduces energy consumption, while deviations from near-neutral pH negatively impact product quality and efficiency. The novelty of this work lies in combining real plant operational data with predictive analytics to establish parameter-based performance relationships and identify optimal operating ranges (e.g., feed pressure ~52–55 bar, pH ~7). These insights provide a strong foundation for optimizing desalination processes, improving membrane efficiency, and guiding the design and operation of future RO desalination projects.
Identification of Plasma Proteins Associated with Alzheimer\u27s Disease Using Feature Selection Techniques and Machine Learning Algorithms
Get PDFAlzheimer’s disease (AD) is a chronic, progressive neurodegenerative disorder that typically affects elderly individuals. Detecting Alzheimer’s using plasma proteins is a critical step toward improving treatment results for this disease. This study aims to use computational algorithms to explore the relationship between plasma proteins and AD progression by identifying a panel of plasma proteins that can serve as biomarkers for tracking and diagnosing AD. We applied two feature selection methods, Sequential Backward Feature Selection (SBFS) and Analysis of Variance (ANOVA) to extract significant proteins from a dataset of 146 proteins. The data was collected from the plasma of 566 individuals, comprising both Alzheimer’s patients and healthy controls. The SBFS technique generated all possible combinations of protein groups from the 146 proteins, which were then trained and tested using five machine learning models: Decision Tree, Random Forest, Extremely Randomized Trees, Extreme Gradient Boosting, and Adaptive Boosting. Subsequently, ANOVA was applied to refine and reduce the selected panel size. Finally, we used XGBoost and AdaBoost models to validate the final panel. The findings introduce a plasma protein panel consisting of A2Macro, BNP, BTC, PPP, and PYY proteins for diagnosing AD. This panel achieved a sensitivity of 88.88%, a specificity of 66.66%, and an AUC of 0.85. These results demonstrate that plasma protein biomarkers can facilitate timely interventions, potentially slowing disease progression and improving patient outcomes. This non-invasive and affordable diagnostic method has the potential to make Alzheimer’s screening accessible to a broader population
The Advanced Actuarial Data Science Based AI-Driven Solutions for Automated Loss Reserving Under IFRS 17 in Non-Life Insurance
Get PDFThis study introduces an AI-driven Automated Actuarial Loss Reserving Model (AALRM) designed to meet IFRS 17 standards for non-life insurance. The model leverages advanced machine learning techniques to improve accuracy, efficiency, and adaptability in loss reserves, with a specific focus on inflation-adjusted frequency-severity modeling. A unique aspect of this research is the integration of bancassurance services, enabling automated management for both microfinance and car insurance on a unified platform. This includes a no-claims bonus system that categorizes policyholders into four tiers—base, variable, final, and high-bonus—resulting in more precise risk assessments and enhanced customer retention. Among eight evaluated machine learning algorithms, the Random Forest (RANGER) outperformed others for estimating Aggregate Comprehensive Automated Actuarial Loss Reserves (ACAALR). The model’s effectiveness was validated through stress tests, scenario analyses, and comparisons with traditional methods like the Chain Ladder. Additionally, the study introduces a novel Robust Automated Actuarial Loss Reserve Margin (RAALRM) with adaptive bounds, addressing traditional limitations in reserve margin calculations. This AI-integrated approach significantly improves predictive accuracy, operational efficiency, and strategic decision-making, offering a scalable solution for the insurance industry
A Comparative Case Study of Two Pedagogical Approaches in Web Development Education: From a Traditional Java Environment to a Modern Ruby on Rails Ecosystem
Get PDFThis study presents a comparative case study of the evolution of a software development course at Kindai University. We analyze two distinct pedagogical ecosystems: a traditional course based on Java Servlet/JSP with a local integrated development environment (IDE), and its subsequent iteration, a modern course employing the Ruby on Rails framework (a Web Application Framework, or WAF), Git for version control, a cloud-based IDE, and Platform as a Service (PaaS) for deployment. This study was not a controlled experiment isolating the effects of a WAF but rather an exploratory analysis of how a shift in the entire toolchain impacted student outcomes and perceptions. Quantitative analyses of student projects over three years for each course revealed that the modern Ruby-based ecosystem resulted in applications with approximately 50% more screens and screen transitions, despite requiring approximately 40% less source code. Furthermore, student surveys indicated significantly higher comprehension and interest in the modern courses. However, the number of data models and user stories remained consistent, suggesting that upstream design thinking was less affected by the technology stack. These findings suggest that adopting a modern, integrated development ecosystem can foster a more productive and engaging learning experience. We conclude by discussing the implications of these findings for curriculum design, emphasizing the value of incorporating contemporary, industry-aligned toolchains into software engineering education, while acknowledging that the observed benefits stem from the synergistic effect of multiple technologies rather than from a single component
Odour, Colour and Turbidity Removal from Selected Industrial Wastewater Using Electrocoagulation Process
No full textElectrocoagulation (EC) is an efficient electrochemical method for treating water using electric charges to destabilize and coagulate pollutants. In this study, a bipolar electrocoagulation reactor with aluminum electrodes was used to treat selected industrial wastewater. Key parameters, including odor, turbidity, Color, and other physicochemical parameters, were analyzed to evaluate the performance of the reactor. Focusing on textile wastewater and two types of cassava wastewater, fufu and starch, this study assessed Odor and turbidity removal using aluminum electrodes. Textile wastewater was used to examine Color removal. The operational parameters—voltage (10V–40V), temperature (30°C–38°C), and operating time (15 minutes to 1 hour)—were systematically varied to optimize the performance. The reactor significantly improved turbidity and color removal, with moderate odor reduction. This study highlights the strong capability of the EC process in reducing turbidity using aluminum electrodes, despite challenges in Odor reduction. The electrocoagulation process effectively removed color, BOD, COD, TSS, and TDS from the wastewater. Voltage adjustments and electrolysis time are critical in optimizing pollutant removal and aligning with regulatory standards for industrial wastewater discharge. Despite its overall effectiveness, the challenges of achieving complete odor and BOD removal highlight areas for further research. This study highlights the potential of EC as a sustainable solution for industrial wastewater treatment