Journal of Techniques

this paper presents a systemic analysis aimed at linking the causes of failures and how they led to changes in system design. The methodology employs an integrated approach using two accident causation models, the Swiss Cheese Model for mapping out the alignment of latent failures and the Systems-Theoretic Accident Model and Process (STAMP) for analyzing the hierarchical deficiencies in the control structure of the system. The analysis ascertains that systemic breakdown across engineering, regulation, and training created the pathway to the accidents. Following the accidents, fault tolerance is successfully implemented in the redesigned system, furthermore, a verified record of zero critical failures across millions of flight hours has been achieved, strengthening the system. Quantitatively, simulations utilizing Root Mean Square Error (RMSE) are carried out to examine the system’s control logics in comparison, providing numerical evidence of improved system stability from 1.68° to 0.79°. The study concludes that while the redesigned system is a necessary corrective safety response, the safety of systems depends on robust structural redundancy and mandatory fault tolerance, which must be incorporated at the design stage in line with regulatory requirements for all future flight control systems.

Machine Vision (MV) is a computational technology that can independently capture images and analyze them to recognize and interpret their content. to automate visual inspections that were previously performed manually by human inspectors. Automated target recognition is among the key industrial applications of machine vision, and quality assurance on production lines depends heavily on defect detection characterized by speed and accuracy to enhance efficiency and product reliability. This study presents a comparative analysis of two instance-segmentation models, Mask R-CNN and YOLOv8x-Seg, for real-time visual inspection of bottles on industrial conveyor systems. Both models are used to detect multiple attributes, including bottle caps, labels, and liquid levels, with a specific focus on empty bottle detection. When test results are compared, Mask R-CNN is adjudged to have perfect accuracy (100%) at detecting caps, labels, and liquid levels, but has no reliability in the detection of empty bottles. It has processing time of between 0.16 and 0.24 seconds, with the time of inspection on the conveyor belt being 0.265 seconds, which limited the system throughput to 226 bottles per minute at a conveyor speed of 45 cm/s. In contrast, the system has been developed with YOLOv8x-Seg with 100% accuracy in all inspections, including detecting an empty bottle, and has also achieved much lower processing times, from 0.02 to 0.07 seconds. The time of inspection on the conveyor belt is 0.140 seconds. This in turn enhances the performance, which allows the system to run at a higher conveyor speed of 67.5 cm/s with a throughput of 428 bottles per minute and also surpasses previous works. Furthermore, decoupling image acquisition from processing and using an index-based PLC tracking mechanism instead of timer-based synchronization to reject defective bottles from the conveyor belt improves performance and reliability, while IoT technologies enhance production line monitoring, controlling, and productivity under dynamic conditions.

The most popular messaging protocol in Internet of Things (IoT) networks is the Message Queuing Telemetry Transport (MQTT). It provides three Quality of Service (QoS) levels for reliable data delivery depending on application needs. Therefore, MQTT is widely adopted and supported by major IoT platforms like AWS IoT, Azure IoT Hub, IBM Watson IoT, and Google Cloud IoT. QoS enhancement techniques are needed in IoT networks because IoT devices and their applications operate in highly constrained and dynamic environments where standard communication protocols (like MQTT) alone cannot always guarantee reliability, timeliness, or efficiency. This study presents a comprehensive review of the recent and current literature on QoS enhancement techniques of MQTT-based IoT systems. These techniques are classified into five primary categories, which are: QoS enhancements at the protocol level, QoS enhancements at the network level, QoS enhancements at the broker level, cooperative subscriber approaches, and adaptive approaches based on context or machine learning. The related works belonging to these categories are explored and discussed in detail, focusing on the trade-offs of delivery reliability, protocol overhead, and energy efficiency. In addition, the study identifies the existing research gaps and limitations, and proposes future directions for advancing scalable, intelligent, and context-aware QoS management in MQTT-based IoT applications

Agricultural production faces challenges, such as water scarcity, climate change, and the rising demand for food to feed the growing global population. Incorporating wireless sensor networks (WSNs), Unmanned Aerial Vehicles (UAVs), and Machine Learning (ML) enables an efficient Synergistic approach for crop monitoring, predicting, and managing. WSNs prefer environmental data instantaneously, UAVs provide data collection for large-scale aerial, and ML processes this data to make actionable decisions. The primary objective of this review is to explore the comprehensive effect of WSNs, UAVs, and ML in the agriculture sector, emphasizing crop yield, environmental sustainability, and optimizing resources. In addition, this review aims to detail the benefits and limitations of ML technology and its impact on farming. Integrating state-of-the-art technologies has expressed several key areas of significant potential, such as crop health monitoring, precision irrigation, yield prediction, disease and pest detection, and resource efficiency. Hence, the collaboration of recent technologies in modern farming significantly enhances monitoring and management in real-time, improves productivity and sustainability, and addresses global food security challenges. At the same time, this would shape agriculture's future through innovative promotion and more sustainable farming practice

Malware remains a significant threat to modern computing systems and networks worldwide. Evolving malware utilizes polymorphism, metamorphism, and zero-day exploits to bypass defenses. Traditional signature-based and heuristic detection methods are now struggling with the increasing complexity of malware. In this paper, a hybrid neural network model is proposed that combines a convolutional neural network (CNN) to detect spatial malware patterns, recurrent neural networks (RNNs) to analyses temporal behaviors, and attention mechanisms to select crucial features for accurate and reliable threat classification. Multi-scale convolutional layers and residual connections improve dataset generalization and reduce overfitting. Focal loss functionality addresses the class imbalance in real-world malware detection scenarios. Experimental results on EMBER, EMBER Sim, and SoReL-20M datasets show exceptional accuracy and precision. This interpretable, scalable deep learning (DL) model bridges traditional methods with modern cybersecurity challenges. The model excels in zero-day detection and produces a few false positives, achieving 96.7% accuracy, 96.1% precision, 96.8% recall, and 96.4% F1-score. Additionally, the findings demonstrate clear improvements over previous methods, achieving a 1.1–2.6% increase in accuracy, confirming the model’s superior detection capability. This advanced deep-learning approach sets a new benchmark in cybersecurity.

In response to increasing environmental challenges and the global demand for sustainable construction materials, this study explores the potential of rubberized concrete incorporating crumb rubber as a partial replacement for fine aggregate. The primary objective is to evaluate the influence of rubber content (0% to 50%) on the thermo-mechanical properties of concrete, including compressive strength, split tensile strength, flexural strength (modulus of rupture), elastic modulus, and thermal properties such as conductivity, expansion, heat resistance, and heat capacity. A total of 128 concrete specimens were prepared using a fixed mix ratio (1:1.05:2.46) with a water-to-cement ratio of 0.47. Experimental tests are conducted to determine slump, density, mechanical strengths, and thermal characteristics. The results reveal that increasing crumb rubber content can significantly reduce compressive strength (from 42.61 MPa to 15.69 MPa), split tensile strength (from 1.81 MPa to 1.09 MPa), and flexural strength (from 4.53 MPa to 2.70 MPa), while enhancing workability and thermal insulation. Thermal conductivity decreases from 0.947 W/m·K to 0.614 W/m·K, indicating better insulation capacity. Although elastic modulus and thermal expansion are not directly tested, reduced density and improved thermal behavior suggest favorable performance under high-temperature conditions. The novelty of this work lies in its systematic investigation of high-volume crumb rubber replacement and the integration of both mechanical and thermal analyses, addressing a gap in existing research. These findings demonstrate the suitability of rubberized concrete for non-structural or thermally demanding applications such as fire-resistant walls and insulating elements, while also promoting solid waste recycling and sustainable building practices.

Damper pulleys are critical mechanical subsystems in internal combustion engines that attenuate torsional vibrations arising from cyclic combustion pressures and crankshaft torque fluctuations. Existing modelling approaches range from simple empirical or linear models to advanced numerical formulations; however, few studies systematically incorporate the combined frequency and temperature dependence of elastomer viscoelasticity with a robust experimental identification framework. The strong thermo-mechanical coupling and material nonlinearity therefore still challenge accurate prediction of torsional behavior under realistic conditions. This study proposes an interactive and novel methodology that couples analytical modelling with experimental identification to characterize the torsional dynamic response of a damper pulley in a thermos-viscoelastic context. Key contributions are: (i) identification of the elastomer complex shear modulus by dynamic mechanical analysis across an extended temperature range; (ii) implementation of these temperature-dependent constitutive parameters in a lumped-parameter torsional vibration model incorporating inertia, stiffness, and viscoelastic damping; and (iii) an iterative calibration procedure correlating analytical predictions with modal parameters extracted from instruments impact tests performed at multiple temperatures. The closed-loop identification updates model parameters systematically until the numerical predictions and experimental measurements converge. Parametric studied quantify the influence of temperature and elastomer thickness: increasing temperature or elastomer thickness reduces the first torsional natural frequency, primarily due to a drop in storage modulus and effective torsional stiffness, whereas decreasing thickness raises the resonance frequency. Predicted and measured natural frequencies differ by less than 5% across the investigated thermal domain, demonstrating the robustness formulation. The developed framework provides a validated and original basis for thermo-viscoelastic modelling, sensitivity analysis, and structural optimization of torsional vibration dampers under varying operating conditions.

Environmentally friendly materials have become an important topic in practical research due to their importance to people's lives and the preservation of the environment. The use of wood plastics is of utmost importance in this field, as it is possible to obtain materials with properties that combine the strength of polymers with the properties of wood. The goal of the current study is to enhance wood-plastic composites by incorporating high-density polyethylene (HDPE), Polycarbonate (PC), and teak wood to improve their hardness and physical properties. An injection molding machine is used to prepare the samples. The Taguchi method is adopted to test the experimental design and three independent variables including teak wood particle size, weight %, and PC content. Samples are tested for water absorption, swelling thickness, density, and Shore D hardness. The findings show that an enlarged teak wood composition can result in greater water absorption and increased swelling thickness. However, the incorporation of PC added in the ratio of 7.5–15% of PC can result in better dimensional stability and decreased water absorption. Wood and PC are also added, which enhance the hardness with respect to pure polyethylene. Through Taguchi method analysis indicate that teak wood content is the most contributing factor towards these properties, followed by PC content and particle size, being least. The results also show that the model L7 type can provide the most balanced performance against mechanical and physical properties. It can be observed from the present work that the wood-plastic composites used have good mechanical properties and good moisture resistance, and are highly suitable for sustainable engineering applications.

This work is a simulation study of cooling feasibility for a 570 x 670 mm solar cell rated at 50 W, evaluated using
COMSOL Multiphysics 6.1. Specifically, the effect of the number of flat fins on the cell temperature is examined. To
determine the optimal number of fins, the impact of three fin cross-sections—flat, triangular, and T-shaped—is
investigated. Following the selection of the optimum cross-section, the effect of adding a nano-PCM mass to the heat sink
is studied. Several masses are analyzed by varying the height of the nano-PCM mass container, with dimensions of 20, 30,
and 40 mm. Once the optimum tank height is identified, the feasibility of using water to cool the nano-PCM mass is
assessed. The results indicated that eight flat fins can yield a similar temperature despite that the lowest cell temperature
has been achieved with 10 flat fins. Therefore, the number of fins is fixed at 8, which reduces the cell temperature to 52.6
°C, a 10% improvement. Subsequently, using triangular-section fins instead of flat ones can reduce the maximum
temperature to 50.8 °C, representing an 11.8% improvement. Supporting the heatsink with a 40 mm-thick container filled
with nano-PCM can further reduce the cell temperature to 44.67 °C (by 22.5%). Finally, the lowest maximum cell
temperature is obtained as 37.9 °C when the nano-PCM described above is cooled with water, resulting in a 34.1%
improvement