Vol. 8 No. 4 (2026): Volume 08 Issue 04

Cybersecurity Testing and Evaluation (T&E) comprise of a foundational resilience component, moving beyond simple quality assurance to become a critical process for continuous organizational hardening. Effective T&E enhances emergency plans, policies, attacks resistance, filtration, firewall strengthening procedures, promoting the efficient utilization of capabilities required to respond to sophisticated cyber-attacks. In this paper, we are performing comparative analysis of traditional security and cyber evaluation metrics with upcoming AI driven enhanced secure measures. AI-LLM security techniques like early defect prediction, defect clustering, Secure cloud and Automated Incident Response measures are weighted against traditional security techniques in terms of – speed, velocity, criticality and depth of coverage scope. These metrics point towards countless advantages of combining the two for greater holistic impact.

Deep learning–based welding defect classification often faces challenges such as limited training data, class imbalance, and high model complexity, which restrict real-time industrial applications. To address these issues, this paper proposes a hybrid CNN–Vision Transformer framework with GAN-based data augmentation for welding defect classification. First, welding images acquired using a wide dynamic range visual sensor are preprocessed through binarization, median filtering, morphological dilation, and cropping to enhance defect features. A Generative Adversarial Network (GAN) is then employed to generate synthetic samples and alleviate dataset imbalance. A lightweight CNN extracts local spatial features and reduces feature dimensionality, after which the resulting feature maps are converted into tokens and processed by a Vision Transformer encoder to capture global contextual relationships via self-attention. The proposed model classifies welding images into four categories: normal, burn-through, undercut, and welding collapse. Experimental results demonstrate that the hybrid architecture achieves improved classification accuracy and computational efficiency compared with conventional lightweight CNN models. In addition, the model attains 98.25% accuracy on the MNIST dataset, validating the effectiveness of the proposed framework.

This study presents a numerical analysis of the deflection of a reinforced concrete building founded on sandy clay soil in the city of Douala. To conduct this research, two numerical simulations were performed. The first consisted of a simulation in MATLAB based on mathematical and mechanical models of the building and the soil. The second consisted of a finite element simulation of the building and the soil using COMSOL Multiphysics. The result obtained for the structural deflection after the MATLAB simulation was 0.0026 mm for elastic soil and 4.1 nanometers when the soil plasticity was taken into account. These results show that the rigidity of the structure and the elastoplastic nature of certain sandy clay soils play a very significant role in the stability of the building. Furthermore, the numerical simulation in COMSOL Multiphysics yielded a deflection of 2.2 mm for both elastic and elastoplastic soils. This is explained by the fact that, under steady-state conditions, the structure settles under its self-weight and live load, which, combined with the rigidity of the structure, resists large displacements and considerably reduces deflection. In addition, during the COMSOL simulation, the threshold stresses of the building were 0.16 MPa for elastic soil and 0.20 MPa for elastoplastic soil. These values represent limits beyond which structural deflection could cause damage to the frame.

This scientific article investigates the impact of data augmentation methods on the effectiveness of brain tumor classification using deep learning techniques. Brain tumor diagnosis from MRI images is a critical task in medical imaging, where limited and imbalanced datasets often reduce model performance and generalization ability. Data augmentation is widely used to overcome these limitations by artificially increasing dataset diversity. In this study, various augmentation techniques such as geometric transformations, intensity adjustments, and GAN-based synthetic image generation are analyzed in terms of their influence on classification accuracy. Experimental results show that data augmentation significantly improves model performance, with accuracy gains ranging from 5% to 20% depending on the applied method. The findings confirm that carefully selected augmentation strategies enhance the robustness and reliability of brain tumor classification systems.

The article presents the results of theoretical studies on selecting the optimal angle of inclination of the grate to the cleaned lint flow, which creates conditions for the unimpeded flow of material along its surface into the working zone of the saw cylinder of the developed pneumomechanical cleaner.

The convergence of Generative Artificial Intelligence (Generative AI) and Digital Twin (DT) technologies is redefining the operational dynamics of healthcare ecosystems and industrial cyber-physical infrastructures. Digital twins provide synchronized virtual representations of physical assets, systems, and processes, while Generative AI enables intelligent synthesis, predictive reasoning, adaptive decision-making, and automated optimization. Despite rapid advancements in both domains, existing frameworks remain fragmented, domain-specific, and inadequately aligned with interoperability, cybersecurity, scalability, and standardization requirements. This research proposes a unified Generative AI-enabled Digital Twin framework designed to support adaptive healthcare analytics and industrial cyber-physical infrastructure optimization. The study synthesizes findings from healthcare digital twins, federated learning, edge intelligence, anomaly detection, predictive maintenance, cybersecurity orchestration, and IoT-driven industrial automation. A layered methodological architecture integrating sensor fusion, data orchestration, federated intelligence, digital twin synchronization, explainable AI, and adaptive optimization mechanisms is introduced. The framework addresses critical operational challenges including real-time synchronization, secure edge analytics, multi-domain interoperability, predictive diagnostics, personalized healthcare adaptation, and resilient cyber-physical monitoring.

The research further evaluates the applicability of the proposed architecture across healthcare diagnostics, intelligent manufacturing, smart monitoring systems, battery management systems, and industrial process optimization. Analytical findings indicate that unified AI-DT ecosystems can significantly enhance predictive accuracy, operational resilience, adaptive resource allocation, and decentralized decision intelligence while reducing latency, infrastructure inefficiencies, and cybersecurity vulnerabilities. However, substantial limitations persist regarding data heterogeneity, explainability, privacy preservation, synchronization complexity, and computational overhead. The study contributes a standardization-aligned conceptual and methodological model suitable for next-generation cyber-physical infrastructures and precision healthcare systems. The proposed framework establishes a research foundation for scalable, interpretable, and autonomous AI-integrated digital twin ecosystems.

This article examines the problem of premature failure of wireless sensor networks (WSNs) in agriculture due to the ‘Energy Hole Problem’, where nodes near the base station run out of power faster than the others [1]. To address this issue, an adaptive solar-aware energy-efficient routing and clustering model (ASAR-EC) is proposed. A key feature of the model is the integration of energy harvesting parameters, which is particularly relevant for regions with high solar radiation, such as Uzbekistan. In ASAR-EC, the selection of cluster heads (CHs) is based on a non-linear function that takes into account residual energy, current solar recharge, and an optional penalty for proximity to the sink. Unlike classical protocols such as LEACH, the proposed approach uses inter-node mesh routing and adaptive sleep mode control based on solar activity forecasts. Theoretical analysis and preliminary evaluations show that the model is capable of increasing the network’s lifetime by 40–60% compared to existing solutions.

The article examines the comparative properties of Retrieval-Augmented Generation algorithms and large-language-model fine-tuning methods in the context of domain-specific search tasks with a high cost of error. The aim is to identify operating regimes in which RAG and fine-tuning differentially affect the accuracy of top-ranked results, the evidential quality of answers, and the safety of handling sensitive data. The relevance of the study is driven by the rapid growth of industrial domain-specific search systems that must simultaneously ensure knowledge updatability, strict citation-based verifiability, and regulatory discipline. The novelty lies in the fact that the comparison is conducted not in the abstract form of RAG versus fine-tuning, but at the level of individual pipeline components and from the perspective of operational trade-offs: it is shown that retrieval and ranking form a truth scaffold and a channel for knowledge refresh, whereas fine-tuning acts as a delicate regulator of format, terminology, and epistemic precision without resolving the problem of obsolescence in parametric representations. The article concludes in favor of hybrid schemes that combine hybrid retrieval, reranking, and strict citation rules with lightweight, parameter-efficient model adaptations, thereby enabling reproducible, controllable, and scalable operation of domain-specific search systems. The article is intended for researchers in information retrieval, engineers of applied RAG systems, and practitioners deploying generative models in high-risk domains.

The article presents a method of purification of obtaining betonies clay for use in catalysis. It was found that the cleaning process includes the stages of centrifugation, drying and mechanical activation. Clay cleaning frees it from impurities and increases the proportion of the useful component - the sorption-active mineral montmorillonite. The porosity of the BET method was determined (measurement of specific surface area and porosity). The accuracy and standardization of the method for determining the adsorption activity have been carried out. An exponential dependence of the adsorption activity on the weight of the sample, (1.0 g), sorption time (35 min) and the equilibrium concentration of the dye solution (39 ml 0.15%) were established. The physicochemical and textural properties of betonies clay for the use of catalysis have been studied. It was found that bentonite clay for use in catalysis is a light gray powder, odorless, practically insoluble in water and organic solvents, with a particle size of less than 0.3 mm.

We present a systematic Geant4 Monte Carlo investigation of neutron transport and attenuation in high-density polyethylene (HDPE, ρ = 0.961 g/cm³) over energies from 25 meV to 14 MeV and slab thicknesses from 2 to 50 cm. Using the QGSP_BIC_HP physics list with the ENDF/B-VIII.0 evaluated nuclear data library, we extract macroscopic attenuation coefficients, half-value layers (HVL), and tenth-value layers (TVL) from exponential fits to the Monte Carlo transmission data, and validate them against reference cross-section data. Spectral hardening of a ²³⁵U Watt fission source is quantified, and HDPE performance is benchmarked against water, paraffin, and ordinary concrete. For 1 MeV neutrons, we obtain μ = 0.475 ± 0.002 cm⁻¹, HVL = 1.46 cm, and TVL = 4.84 cm, confirming HDPE as one of the most effective compact solid neutron shields across the energy range studied.

Industry 4.0 has increased the demand for intelligent predictive maintenance systems capable of supporting real-time monitoring, early fault detection, and efficient decision-making in industrial environments. In this context, accurate prediction of machine failures has become essential for minimizing downtime, reducing maintenance costs, and improving operational reliability. This study employs the Predictive Maintenance Dataset from the UCI repository to develop and evaluate data-driven models for machine failure prediction and classification. The research pursues two primary objectives: first, to compare the performance of several machine learning algorithms in classifying machine failures, and second, to assess the effectiveness of deep learning approaches in achieving higher predictive accuracy. Among the machine learning models examined, the XGBoost classifier demonstrates the strongest performance. To further enhance prediction capability, this study adopts a hybrid Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM) model, which integrates CNN’s strength in automatic feature extraction with LSTM’s ability to learn temporal dependencies from sequential data. Experimental results show that the proposed CNN–LSTM model outperforms traditional machine learning models as well as Artificial Neural Networks (ANN) in predicting machine failures. The main contribution of this study lies in the comparative evaluation of machine learning and hybrid deep learning techniques on an imbalanced predictive maintenance dataset. The findings confirm the potential of hybrid deep learning models for predictive maintenance applications and highlight their practical value in enabling proactive maintenance strategies, optimizing resource allocation, and enhancing the reliability of smart industrial systems.

The increasing adoption of Continuous Integration and Continuous Delivery (CI/CD) workflows has significantly accelerated software development cycles, but it has simultaneously introduced complex security challenges. Traditional post-development security testing approaches often fail to detect vulnerabilities early, resulting in higher remediation costs and increased exposure to threats. This research investigates the impact of proactive safeguard validation—commonly associated with shift-left security practices—on the early identification of system weaknesses within CI/CD environments. The study synthesizes theoretical constructs from workflow management systems, intrusion detection methodologies, and distributed system security models to propose a structured validation framework integrated within early development stages.

Drawing upon prior work in workflow orchestration (Aalst, 2000; Liu et al., 2001) and intrusion detection systems (Li et al., 2019; Sadotra et al., 2019), this research develops a multi-layered validation model that embeds automated security testing mechanisms within CI/CD pipelines. The framework evaluates static, dynamic, and behavioral validation strategies across development phases. Empirical and conceptual analysis demonstrates that early-stage validation significantly improves vulnerability detection rates, reduces false positives, and enhances system resilience. Furthermore, the study incorporates findings from Thanvi et al. (2026) to highlight the measurable benefits of integrating security testing earlier in the software lifecycle, including reduced mean time to detection and improved pipeline efficiency.

The findings indicate that proactive validation not only strengthens security posture but also contributes to operational efficiency by minimizing rework and deployment delays. However, challenges such as integration complexity, performance overhead, and toolchain compatibility remain critical considerations. The research concludes by emphasizing the necessity of adaptive validation frameworks that align with evolving DevOps practices and recommends future exploration of AI-driven security automation.

This paper presents a mathematical model in the form of complex differential equations describing the mechanical cutting of white cast irons. Heat conduction PDE, similarity criteria theory (Peclet, F, D), and mechanical cutting theory were applied. A system of six interrelated equations was developed covering geometry, temperature, forces, plasticity, wear, and heat balance. Conclusion: The model enables justification of tool material requirements for machining white cast irons.

The article examines the ensemble of causes of premature wear of sealing elements in the Wankel engine under high-load operating regimes from the standpoint of the coupled evolution of sealing performance, thermal state, and tribological contact conditions. The aim of the study is to formalize the causal cascade of pressure, heat release, deformation, lubrication, and wear, and to interpret premature failure as the assembly's exit from its functional window rather than as a simple acceleration of material loss. The relevance of the work stems from the growing interest in highly loaded rotary power plants, where seal durability remains a key limitation to reliability and efficiency. Based on a synthesis of contemporary computational and experimental data on leakage dynamics, coating contact fatigue, and lubrication regimes, an integrated degradation framework is proposed that links local overheating, vibrational mobility of seals, and the transformation of wear mechanisms from abrasive and adhesive to fatigue and corrosion–chemical. The scientific novelty lies in the systematic description of failure scenarios for apex, side, and corner seals under load as self-sustaining loops, as well as in the emphasis on diagnostic indicators and engineering strategies for breaking the cascade through control of the thermal regime, lubrication, detonation, and the selection of tribological pairs. The article is intended for designers and researchers of rotary engines, tribology specialists, engine manufacturers, and operating engineers dealing with highly loaded units.

This study examines the operational principles of classical encryption algorithms, which form the foundation of modern cryptographic systems. Classical encryption techniques, including substitution ciphers, transposition ciphers, and symmetric key algorithms, are analyzed in terms of their structure, functionality, and security properties. The research focuses on understanding how plaintext is transformed into ciphertext through systematic encryption procedures and how decryption restores the original information. Special attention is given to the historical development of classical cryptography and its role in the evolution of information security. The study also evaluates the strengths and limitations of classical algorithms, particularly their vulnerability to frequency analysis and brute-force attacks. By comparing different encryption methods, the paper highlights the importance of algorithmic design in ensuring data confidentiality and integrity. The findings demonstrate that while classical encryption algorithms are relatively simple, they provide essential conceptual foundations for modern cryptographic approaches and continue to be valuable for educational and introductory purposes in cybersecurity.

This article compares the MSVC and Clang/LLVM compiler stacks for native ARM64 builds in the Windows on Arm environment on Qualcomm Kryo and Oryon platforms. The study's relevance lies in the transition of the Windows ecosystem toward native ARM64 execution and the growing role of the compiler in determining software efficiency on new Snapdragon X systems. The aim of the study is to conduct a quantitative and qualitative assessment of code-generation differences between the two toolchains for computational workloads with different profiles. The novelty of the work is defined by a systematic cross-microarchitectural comparison of new MSVC and Clang/LLVM versions in the specific Windows on Arm environment, with separation of compiler influence from hardware platform influence. It was established that Clang/LLVM delivers higher performance in most tests in audio and video encoding, sorting, and interpreted code execution, whereas MSVC reveals a pronounced advantage in the isolated NumPy sqrt mathematical kernel on the Oryon architecture. The results confirm that code-generation efficiency depends on workload characteristics and processor microarchitecture. The article will be useful for developers of C/C++ systems, compiler researchers, and software architects targeting Windows on Arm.

This study provides a comprehensive legal and economic analysis of Uzbekistan’s accession process to the World Trade Organization. The article examines the stages of the accession process and the adaptation of national legislation to international agreements such as GATT, GATS, and others. The methodology is based on legal analysis, a comparative approach, and the application of TRIST and partial equilibrium models. The findings indicate that WTO membership is expected to enhance GDP growth, attract foreign direct investment, and expand exports. At the same time, the study identifies potential short-term adjustment challenges in certain protected sectors. In conclusion, WTO accession is considered a key driver of economic modernization and deeper integration into the global economy.

Despite the transformative potential of artificial intelligence (AI) across multiple sectors, its deployment generates significant risks for human rights, exacerbates structural inequality, and produces outcomes of limited predictability. These dynamics make the development of robust ethical and legal governance frameworks both urgent and complex. Uzbekistan adopted Ethical Rules for the development, deployment, and use of AI-based solutions in 2026, marking an important step in national regulatory development. Yet questions of implementation, understood here as the formal incorporation of normative principles into binding legal instruments and the establishment of institutional enforcement mechanisms, remain largely unresolved. This paper employs a comparative legal analysis method, grounded in soft law theory and legal transplant theory, to examine how Uzbekistan's emerging AI governance framework aligns with international standards, including UNESCO's 2021 Recommendations, the OECD Principles on AI, and the European Union's Artificial Intelligence Act. The analysis identifies six structural gaps in Uzbekistan's current framework and proposes a pathway from declaratory principles toward enforceable, institutionally supported standards. The paper contributes to the emerging literature on AI governance in transitional legal systems and to broader debates on the conditions under which soft law instruments can effectively constrain technological risk.

Modern healthcare, research, and data center facilities require highly reliable HVAC systems capable of maintaining strict environmental conditions and pressure relationships. Inadequate airflow distribution, hydronic imbalance, and poor pressure control can compromise system performance and safety in mission-critical environments. Testing, Adjusting, and Balancing (TAB) provides a structured methodology to validate and optimize system performance.

This paper presents an integrated TAB framework that includes air distribution balancing, chilled water hydronic balancing, liquid cooling system balancing, and pressurization verification for critical rooms. Field measurements were evaluated using descriptive statistics and regression analysis to quantify performance improvements.

Results indicate airflow deviation reductions of up to 74%, chilled water flow variation reductions of 72%, liquid cooling efficiency improvements of approximately 12%, and pressure stability improvements within ±0.01 in. w.g. A case study conducted at Sidra Medicine in Doha demonstrates the application of this methodology in a large healthcare and research facility. The findings highlight TAB as an essential component of commissioning and performance validation in complex HVAC systems. 

This paper presents an inspection-based civil case study of drainage vulnerability during active construction. Using two civil field inspection reports, a coordination log, response emails, and supporting site photographs from a mid-construction residential project in Florida, the study examines how drainage-related deficiencies developed after stormwater infrastructure was already materially installed, how those deficiencies were addressed, and what those events reveal about preserving drainage function before final site stabilization. The record shows a progression from control slippage to direct system vulnerability, including temporary exposure of active drainage elements, sediment intrusion into collection features, uncertainty in an internal manhole control detail, breakdown of work-area sediment containment, and localized loss of drainage effectiveness caused by sediment accumulation. These conditions are interpreted as linked preservation failures within a live drainage network rather than as isolated field defects. The response record further shows that different conditions required different closure pathways, including immediate field protection, correction and reconciliation of a questioned internal detail, and reinforcement of site-scale sediment controls. On that basis, the paper distinguishes between observed deficiency, documented corrective action, and construction-control closure. The study concludes that the principal construction-stage risk lies not in incomplete installation alone, but in loss of protection around drainage assets that have already become operational, and it proposes a practical framework for preserving those assets until the site reaches stable completion. 

Urbanization increases impervious cover and accelerates stormwater delivery to nearby streams and rivers, often resulting in higher runoff volumes, shorter response times, flashier hydrographs, channel instability, and degraded water quality. Low-impact development (LID) practices aim to manage runoff close to its source by enhancing infiltration, detention, storage, evapotranspiration, and pollutant removal. This short review summarizes how common LID practices can reduce urban runoff near river corridors and support more stable hydrologic and hydraulic behavior in receiving waters. The paper first outlines the concept of LID and its relevance to urban river systems. It then reviews the main practices commonly applied in built environments, including bioretention cells, permeable pavements, green roofs, vegetated swales, and riparian buffer measures. Their primary hydrologic and hydraulic benefits are discussed in terms of runoff volume reduction, peak-flow attenuation, delayed runoff response, and reduced erosive stress on urban channels. The review also highlights key implementation challenges, including maintenance needs, clogging, limited land availability, groundwater constraints, and reduced effectiveness during extreme storms. The paper concludes that LID is most effective near river corridors when implemented as a distributed source-control strategy that is integrated with riparian protection and broader catchment-scale flood and water-quality management. 

Advanced ceramics have become foundational materials in high precision manufacturing equipment used across semiconductor, display, photovoltaic, and advanced materials industries [1], [3]. Their exceptional thermal stability, chemical inertness, mechanical rigidity, and plasma compatibility enable performance levels that metals and polymers cannot achieve. As device architectures become more complex and process windows narrower, ceramic components play a critical role in ensuring uniformity, reliability, and contamination control. However, ceramics also introduce significant challenges in design, manufacturability, cost, and long-term stability [9], [10], [15]. This paper examines the strategic importance of ceramics in high precision equipment and outlines the technical and operational challenges that must be addressed to fully leverage their capabilities.