Applied Science and Engineering Journal for Advanced Research
Not a member yet
146 research outputs found
Sort by
EWWW - Eco Way to a Waste Free World Smart Waste Management using Deep Learning Models
No full textThe urgency for efficient waste segregation has increased as a result of accelerated urbanisation and consumption levels. Manual sorting is still the norm in many urban areas, but it is often slow and erratic. Deep learning algorithms have shown great promise in the automated categorisation of waste. Yet, most prior work has been hampered by issues like limited dataset size, uneven class distribution, computationally intensive architectures, and a lack of generalisation to uncontrolled, real-world settings. In this research, a comparative analysis is presented, covering classical machine learning and custom deep learning algorithms, transfer learning, and transformer-based models, evaluated on the widely used TrashNet benchmark dataset. The methods covered include feature-based neural networks and custom convolutional models such as ResNet-style variants , as well as traditional classifiers including SVM, KNN, Random Forest, Logistic Regression, and Naïve Bayes. In addition, numerous pretrained ImageNet models are included such as ResNet50, DenseNet121, MobileNetV2, InceptionV3, Xception, and EfficientNet-B0. Hyperparameter tuning is applied to EfficientNet-B0 using Optuna. An ensemble performance study is conducted on soft-voting networks combining EfficientNet-B0, Xception, and ResNet50, along with transformer models such as ViT, ConvNeXt, and Swin Transformers. Model performance is evaluated based on classification accuracy, robustness, and feasibility for deployment. This study contributes as a benchmarked comparative work in the field of intelligent, data-driven waste segregation research
Transforming Education with Large Language Models: Opportunities, Challenges, and Ethical Considerations
No full textLarge Language Models (LLMs), such as OpenAI’s GPT-4, significantly advance artificial intelligence, offering transformative potential in education. This paper examines how LLMs can enhance personalized learning, content creation, and real-time tutoring by generating diverse, high-quality educational materials and adapting to individual student needs. While LLMs present considerable opportunities, they also pose challenges related to technology dependency, content accuracy, data privacy, and inherent biases. By reviewing current implementations and case studies, this paper highlights the benefits and ethical considerations of LLMs in education. Recommendations for educators and policymakers include balanced integration, robust content verification, stringent data privacy measures, and bias mitigation strategies. Future research should focus on improving LLM accuracy, emotional intelligence, and ethical frameworks to advance personalized, adaptive learning in an equitable and ethical manner
Eco-Hybrid Aluminum AA5042 Composites Reinforced with Soda-Lime Waste Glass and Fly Ash: Mechanical, Thermal, and Wear Performance for Pulley Applications
No full textThis study investigates the fabrication and performance of an eco-hybrid aluminum AA5042 composite reinforced with waste soda–lime glass and fly ash to enhance its mechanical, thermal, and tribological behavior for pulley applications. The composite was produced via stir casting, and its mechanical properties—hardness, toughness, and tensile strength—were optimized using Response Surface Methodology (RSM) based on three variables: soda-lime glass waste content, fly ash content, and preheat temperature. EDXRF, and SEM analyses confirmed the presence and uniform dispersion of reinforcements within the aluminum matrix, showing good interfacial bonding and minimal porosity. The RSM models revealed that both reinforcements significantly improved hardness and strength due to the introduction of hard ceramic phases, while excessive additions led to agglomeration and interface weakening. Preheat temperature exhibited a mild negative linear but a positive quadratic effect, indicating that moderate heating enhances bonding and porosity reduction. Toughness increased with moderate reinforcement levels and controlled preheating, though excessive fly ash and temperature reduced ductility due to embrittlement. Tensile strength improved with soda–lime glass addition and moderate preheating, but declined beyond optimal reinforcement levels from poor dispersion and matrix imbalance. Thermogravimetric analysis (TGA) showed excellent thermal stability with less than 2% weight loss up to 1000 °C, while wear analysis indicated low wear depth (≈0.266 µm) and steady wear behavior. Overall, integrating waste soda–lime glass and fly ash into aluminum AA5042 yields a sustainable, lightweight composite with superior strength, hardness, wear resistance, and thermal stability—ideal for pulley and other high-performance engineering applications
Streamlining Network Operations: Combining Meraki MX with Cisco DNA Center for Automation and Assurance
Get PDFThis research explores the integration of Meraki MX with Cisco DNA Center to better operate networks, automate management processes, and ensure network performance and reliability. Modern network environments are becoming increasingly complex, and organizations are seeking solutions that make it possible to enhance automation, reduce manual work, and improve operating efficiency. The study primarily evaluates the level of automation through the Meraki MX device, scrutinizes the implementation and performance assured through Cisco DNA Center, and investigates the positive impacts of merging them on performance efficiency and efficacy of the operations in the networks. This exploratory qualitative piece synthesizes qualitative case studies based on secondary research on expert views and technical manuals regarding the aspects and best practice that could or are being accomplished in this mergence. The results indicate major network performance improvement, which include a 6.5% increase in uptime, a 62.5% reduction in troubleshooting time, and a 20% increase in network health score. On the other hand, the issues included compatibility with legacy systems, initial setup costs, and training of staff were identified. Based on the research, the conclusion is that integration offers great advantages in terms of automation and efficiency in operation; however, it has to address these challenges to be successfully implemented. It seems the research helps develop a better understanding of how the union of Meraki MX with Cisco DNA Center could optimize network management by giving practical insights into surmounting integration hurdles and maximum performance
Driving Digital Transformation: Leveraging Site Reliability Engineering and Platform Engineering for Scalable and Resilient Systems
Get PDFIn today\u27s competitive landscape, achieving scalability, resilience, and rapid innovation is important for organizations seeking digital transformation. This paper describes how Site Reliability Engineering (SRE) and Platform Engineering can be used to help drive these transformations. Integrating SRE practices with robust platform engineering methodologies allows organizations to develop the tools they need to build scalable, high-performing, and resilient systems. The paper discusses methodologies used, a mixed-method approach combining qualitative case studies and quantitative performance metrics, to evaluate the impact of SRE and Platform Engineering. Results from case studies across multiple organizations indicate important improvements in uptime, recovery time, scalability, and overall efficiency of the systems. This work highlights the crucial role that these engineering practices play in enabling digital transformation and operational excellence
Performance Assessment of Concrete Using Paper and Wastewater Sludge to Replace Part of the Cement
Get PDFAlong with deforestation and the use of fossil fuels, the cement manufacturing sector contributes significantly to carbon dioxide (CO₂) emissions. Additionally, the concrete industry is one of the major consumers of natural raw resources, which has an impact on environmental sustainability. In order to tackle these issues, this study examines the effects of partially substituting paper mill and wastewater sludge for cement in weight percentages of 5%, 10%, and 15% on the compressive, split tensile, and flexural strengths of concrete at 7 and 28 days of curing. According to experimental data, the 5% replacement mix showed better mechanical qualities than the control mix (0% replacement), suggesting that it could be a sustainable option in the manufacturing of concrete, even though higher replacement levels resulted in a decrease in strength. Strength was shown to decrease after 5%, underscoring the drawbacks of adding too much sludge. The viability of using industrial by-products in concrete to lessen reliance on cement and CO₂ emissions while preserving structural integrity is clarified by this study. By encouraging the use of waste materials in cement-based composites, the findings support the continuous efforts towards sustainable construction methods
Automating Scalable and Secure Enterprise Applications with Full-Stack Java: CI/CD Integration with Canary Testing
Get PDFFocusing on the integration of Continuous Integration and Continuous Deployment (CI/CD) pipelines with canary testing techniques, this paper investigated the automation of scalable and secure enterprise systems created with full-stack Java. The study looked at changes in deployment frequency, system performance, dependability, and security posture by means of a thorough DevOps architecture. Apart from a notable drop in security vulnerabilities, the results showed notable improvements in deployment efficiency, less downtime, and quicker recovery times. By allowing incremental rollouts and early problem detection, canary testing showed efficacy in risk reduction, hence guaranteeing better system stability. The combination of security automation and compliance and vulnerability monitoring was made even stronger by it. The research confirms that for modern enterprise application delivery, combining CI/CD automation with canary testing is a strong strategy since it balances agility with operational resilience
The Role of AI in Automating Cyber Incident Response: Challenges and Opportunity
Get PDFCyber threats are becoming more common and more complex, therefore we need faster and smarter ways to respond to them. This study looked into the function of Artificial Intelligence (AI) in automating the response to cyber incidents, focusing on how well it works, what problems it might face, and what opportunities it might create. A mixed-methods approach was used, which included testing how well AI-based tools worked in fake cyber-attack situations and talking to cybersecurity experts. The results showed that AI tools cut down on detection and response times by a lot while still being quite accurate at finding and stopping threats. However, concerns regarding trust, explainability, and integration with legacy systems emerged as key barriers to adoption. The results imply that AI has the ability to change cybersecurity for the better, but it won\u27t be successful unless systems that are clear and easy to understand are made that can work with human experience. These insights are very helpful for companies who want to use AI to improve their ability to respond to incidents
Magnetic Levitation and Superconductors
No full textThis paper explores how stable magnetic levitation can be achieved using a combination of permanent magnets and superconductors. When these materials interact, magnetic fields create a lifting force due to the Lorentz effect. By designing a proper magnetic field pattern and adjusting the magnetic strength and material hardness, the system can reach a point of stable balance, or equilibrium. This study also dies into the difference between active and passive reputation and their responses affecting levitation stability, The results suggest that with suitable magnetic geometry and alignment, a permanent magnet and a superconductor can produce a reliable levitation effect
Blockchain and AI Amalgamation for IoT Applications beyond 5G: A Case Study of Enhancing Data Security and Privacy in 5G-IoT Ecosystems
No full textThe Internet of Things (IoT) in connection with blockchain technology and artificial intelligence (AI) concerning beyond 5G (B5G) networks promises a revolutionary means of coping with the most topical issues of data privacy and existing security concerns. With 5G technology indicating a new age of a higher connectivity, high bandwidth and ultra-low latency, it will also result in increasing the vulnerabilities that have been alleged with IoT ecosystems. In this research paper, we address the issue of how blockchain and AI can be integrated to enhance data security, data privacy within the 5G-IOT systems. The main directions are the use of the decentralized and immutable ledger with blockchain to securely conduct data transactions, applying the AI to intelligent processing of the data and real-time threats detection, the combination of the technologies to build robust, autonomous, and safe IoT systems