Spectrum of Engineering Sciences

OPTIMIZING BOILER WATER CHEMISTRY AND CORROSION CONTROL SYSTEMS FOR ENHANCED EFFICIENCY AND RELIABILITY

2026-02-05T09:59:49+05:00

Background: Industrial boilers are vital in power generation, petrochemical processing, and large-scale heating applications. Maintaining proper water chemistry is essential to prevent scale formation, corrosion, and operational inefficiencies, as deviations can reduce heat transfer, increase fuel consumption, and shorten equipment lifespan.

Objectives: The study aimed to: (1) assess the impact of water chemistry on scale formation, corrosion, and overall boiler performance; (2) evaluate conventional and advanced chemical treatment methods; (3) investigate corrosion behavior under varying operating conditions; and (4) develop a comprehensive boiler water management and corrosion control framework to enhance system reliability and longevity.

Methods: A hybrid approach combined long-term field monitoring, laboratory analysis, and corrosion-coupon testing. Samples were collected from feedwater, deaerator outlets, boiler drums, and condensate return lines. Parameters measured included pH, alkalinity, hardness, TDS, conductivity, dissolved oxygen (DO), and silica. Scale deposition was assessed via suspended solids and microscopic examination, while corrosion rates were determined using weight-loss coupons and electrochemical techniques. Controlled laboratory tests were used to supplement field observations.

Results: Maintaining alkaline pH, moderate TDS, and low DO effectively minimized corrosion and scale deposition. Chemical treatments including oxygen scavengers, phosphate conditioning, dispersants, and controlled blowdown reduced corrosion rates in treated boiler water to 1.0–1.4 mpy, compared to 8–12 mpy in untreated water. Silica accumulation in the boiler drum and minor DO variations in condensate return lines indicated areas needing ongoing monitoring. Hardness and alkalinity levels confirmed effective softening and internal phosphate reactions.

Conclusions: Coordinated water-management strategies combining chemical treatment, mechanical pretreatment, and continuous monitoring improve boiler reliability, energy efficiency, and lifespan. Regular evaluation of water chemistry is essential to prevent scaling, corrosion, and operational disruptions.

MECHANICAL AND PHYSICAL PERFORMANCE OF LIGHTWEIGHT CONCRETE USING WASTE PUMICE AS COARSE AGGREGATE

2026-02-05T11:55:38+05:00

The increasing demand for sustainable and lightweight construction materials has encouraged the utilization of industrial waste in concrete production. This study investigates the mechanical and physical performance of lightweight concrete produced by partially and fully replacing natural coarse aggregate with waste pumice obtained from the apparel industry. Concrete mixes were prepared with pumice replacement levels of 0%, 20%, 40%, 60%, 80%, and 100% while maintaining a constant water–cement ratio. Fresh concrete properties were evaluated through slump tests, whereas hardened concrete was assessed for dry density, self-weight, and compressive strength at curing ages of 7, 14, 21, and 28 days. The results indicate that increasing pumice content leads to a reduction in workability, density, and compressive strength due to the porous nature and high-water absorption of pumice aggregate. However, a significant reduction in self-weight was achieved, demonstrating the effectiveness of waste pumice in producing lightweight concrete. Concrete mixes containing low to moderate pumice replacement levels exhibited acceptable strength performance suitable for non-load bearing and lightweight structural applications. A strong correlation was observed between dry density and compressive strength, highlighting the predictable behavior of pumice-based concrete. The findings confirm that waste pumice can be successfully utilized as an environmentally friendly coarse aggregate alternative, contributing to sustainable construction and efficient waste management.

ADVANCED ELECTROCHEMICAL CHARACTERIZATION OF REDOX-ACTIVE MATERIALS FOR HIGH-PERFORMANCE ENERGY STORAGE SYSTEMS: A COMPREHENSIVE STUDY OF CHARGE TRANSFER, ION DIFFUSION, AND ELECTRODE STABILITY

2026-02-05T12:22:43+05:00

This study presents a comprehensive electrochemical characterization of redox-active materials aimed at enhancing the performance, efficiency, and durability of high-performance energy storage systems. The research focuses on evaluating charge-transfer kinetics, ion-diffusion behavior, and electrode stability using a combination of cyclic voltammetry, electrochemical impedance spectroscopy, and galvanostatic charge–discharge analysis. Cyclic voltammetry revealed well-defined and reversible redox peaks, indicating efficient electron-transfer processes and diffusion-controlled electrochemical reactions. Electrochemical impedance spectroscopy demonstrated a marked reduction in charge-transfer resistance after activation cycles, while the linear Warburg region confirmed effective ion movement within the electrode matrix. Galvanostatic charge–discharge testing further supported these findings by showing high specific capacities, excellent Coulombic efficiency, and strong capacity retention during extended cycling.
Structural and surface characterization through advanced microscopy techniques provided additional insight into material stability. The materials maintained their morphological integrity after repeated electrochemical cycling, with minimal signs of cracking or particle degradation, confirming their mechanical robustness and uniform distribution of active components. These combined results indicate that the synergistic effect of fast charge-transfer, efficient ion diffusion, and durable electrode architecture contributes significantly to the superior performance of the studied materials.

This investigation emphasizes the importance of integrating electrochemical and structural analyses to optimize redox-active materials for next-generation batteries and supercapacitors. The findings provide a valuable foundation for the development of high-efficiency, long-lasting, and fast-charging energy storage technologies.

A CUSTOMIZED SWIN TRANSFORMER-BASED FRAMEWORK FOR CASSAVA LEAF DISEASE CLASSIFICATION

2026-02-06T10:14:19+05:00

Cassava leaf diseases present significant agricultural challenges due to visual similarity between pathological conditions and variability in field conditions, complicating timely intervention. The accuracy of disease identification in early stages has been critical in preventing crop losses; however, due to symptom overlap and environmental variations, manual monitoring has become increasingly difficult. In this paper, a deep learning approach for cassava disease diagnosis named "Modified Swin Transformer Framework" has been proposed, attempting to enhance classification capability by employing a transformer-based vision approach. In the proposed method, the hierarchical structure of Swin Transformer has been customized based on input dimensionality, adaptive patch embedding, and output targeting for cassava disease classification. In this approach, the input image has been split into adaptive non-overlapping patches and processed using shifted windows and attention within these patches. This process has helped the method link all windows efficiently by avoiding locality issues of non-overlapping regions in attention, while being computationally efficient. The framework has further developed based on Swin Transformer architecture and has included adaptive patch and position embeddings to take advantage of the transformer's global-linking capability by employing multi-head attention in these embeddings. Furthermore, the framework has developed and incorporated multi-scale feature aggregation into this method, which utilizes hierarchical feature fusion with these inclusive designs to address multi-scale symptom representation during processing. The inclusion of multi-scale aggregation has therefore facilitated this method to link global patterns as well as local patterns; hence, its integrity has helped improve disease classification capability by minimizing intra-class variability of cassava diseases and increasing inter-class differences among Cassava Bacterial Blight, Cassava Brown Streak Disease, Cassava Green Mottle, Cassava Mosaic Disease, and healthy leaves. In testing the proposed framework, an accuracy of 96.80% and an F1-score of 96.40% have been achieved on the Kaggle public dataset, which has outperformed standard CNN models and baseline Swin Transformer; the framework has thus proved its effectiveness as a computer-assisted tool for cassava disease observation and classification.

DIGITAL TWIN–DRIVEN ARTIFICIAL INTELLIGENCE MODELS FOR AUTOMATION AND OPTIMIZATION OF COMPLEX ENGINEERING SYSTEMS

2026-02-06T15:23:11+05:00

The escalating complexity of modern engineering systems, characterized by high dimensionality, stochastic dynamics, and non-linear interdependencies, has rendered traditional model-based control strategies insufficient. Static models, typically derived from ideal design parameters (CAD/CAE data), fail to account for the continuous temporal degradation, sensor drift, component fatigue, and environmental variance inherent in physical assets operational in the field. This research investigates the architectural and functional integration of Digital Twins (DT) with Artificial Intelligence (AI) to establish a paradigm of active, closed-loop intelligence. By conceptualizing the Digital Twin not merely as a passive replica or visualization tool but as a semantic mediator for bidirectional synchronization, this study demonstrates how AI models can leverage real-time high-fidelity state estimation to drive autonomous optimization. The proposed framework facilitates a fundamental transition from reactive maintenance and static control to predictive, self-optimizing system behaviors that adapt to the evolving physics of the machinery. The findings indicate that Digital Twin–driven AI significantly enhances automation capability levels and optimization responsiveness compared to conventional control methods, offering a robust, theoretically grounded pathway for the management of next-generation Cyber-Physical Systems (CPS)..

DECISION-MAKING IN CLOUD SECURITY MANAGEMENT USING AI-BASED ANOMALY DETECTION: A SOCIAL SCIENCES PERSPECTIVE

2026-02-06T15:37:20+05:00

The rapid migration of critical organizational infrastructure to cloud environments has precipitated a fundamental shift in security operations, necessitating a reliance on Artificial Intelligence (AI) and Machine Learning (ML) for anomaly detection. As cloud architectures evolve into ephemeral, microservices-based ecosystems, the volume of telemetry data surpasses human cognitive processing capabilities, positioning AI not merely as a tool but as a requisite agent of surveillance. While technical discourse predominantly prioritizes detection accuracy, latency reduction, and computational efficiency, the sociotechnical implications of these systems on human decision-making remain critically under-theorized. This study examines AI-based anomaly detection not as a neutral technical instrument, but as a potent socio-technical influence mechanism that reconfigures organizational judgment, authority, and power. Adopting an interpretive lens grounded in sociotechnical systems theory, sensemaking, and institutional theory, the research investigates how algorithmic outputs shape human interpretation of risk, renegotiate the locus of decision authority, and alter governance structures. The analysis reveals that AI-driven anomaly detection introduces a "black-box" authority that can erode human epistemic confidence, necessitating new frameworks for accountability where decision-making power is shared between human analysts and opaque algorithms. Furthermore, it identifies a phenomenon of "liability shielding," where reliance on algorithmic outputs serves as a defensive mechanism against organizational blame. This article contributes to the information systems and organizational studies literature by conceptualizing the shift from human-centric security management to a hybrid, algorithmically mediated governance model, offering a theoretical roadmap for navigating the paradoxes of automated security

RETRIEVAL-AUGMENTED GENERATION: ARCHITECTURES, ADAPTIVE RETRIEVAL, FEEDBACK-DRIVEN OPTIMIZATION, AND OPEN RESEARCH CHALLENGES

2026-02-06T17:06:07+05:00

RAG has become a core paradigm for grounding LLMs into external knowledge sources, preventing hallucinations, and allowing scalable reasoning over dynamic corpora. By integrating parametric language modeling with non-parametric retrieval mechanisms, RAG systems close the gap between fluent natural language generation and factuality. Unlike fully parametric models, RAG provides access to recent and domain specific information at the time of inference. Nevertheless, recent empirical results suggest that a majority of already-deployed RAG pipelines are still brittle because they only rely on static similarity-based retrieval techniques and simple naïve pipeline strategies for context construction and weak or even relevance-only reranking with no feedback-driven adaptivity. The following survey will be based on a very brief overview of current RAG research, including: basic architectural designs; retriever and re-ranker strategies; context construction methodologies; adaptive and reinforcement-learning-based RAGs; feedback aware models; graph based extensions; memory based extensions; long context behaviour; hallucination analysis; and new evaluation benchmarks. We base our work on over forty representative studies to introduce a single taxonomy, provide a metadata-based comparative analysis, and provide comprehensive information on open research challenges. Our argument is that the RAG systems of the future should no longer be in the paradigm of static retrieval, but rather integrate context adaptation mechanisms to feedback to improve the resilience, efficiency and effectiveness of deployment.

A COMPACT SUPER-WIDEBAND FRACTAL ANTENNA FOR NEXT-GENERATION 5G WIRELESS NETWORKS: DESIGN OPTIMIZATION AND PERFORMANCE EVALUATION FOR HIGH-DATA-RATE AND LOW-LATENCY APPLICATIONS

2026-02-07T09:40:45+05:00

The rapid evolution of fifth-generation (5G) wireless communication systems has intensified the demand for compact antenna solutions capable of supporting ultra-high data rates, low-latency transmission, and reliable broadband connectivity across diverse operating conditions. Achieving these requirements simultaneously remains a significant challenge due to the inherent trade-offs between antenna size, impedance bandwidth, radiation stability, and efficiency. In this paper, a compact super-wideband fractal antenna is proposed, designed, and optimized to address the stringent performance requirements of next-generation 5G wireless networks. The proposed antenna employs a space-filling fractal geometry that exploits self-similarity and multi-scale current paths to significantly enhance impedance bandwidth while maintaining a compact physical footprint suitable for integration into modern wireless devices. The antenna structure is realized on a low-profile dielectric substrate and fed using an optimized planar feeding technique to ensure wideband impedance matching and stable excitation. A systematic design evolution process is presented, beginning with a conventional radiator and progressively introducing fractal iterations and ground-plane modifications to achieve super-wideband performance. Key geometrical parameters, including fractal iteration scale, radiator dimensions, feed geometry, and ground configuration, are carefully optimized through extensive parametric analysis to maximize bandwidth, improve radiation efficiency, and stabilize gain across the operating frequency range. The optimization strategy is guided by well-defined performance objectives, including a reflection coefficient below −10 dB, minimal impedance fluctuation, and consistent radiation characteristics throughout the band. Comprehensive electromagnetic simulations are conducted to evaluate the antenna’s performance in terms of impedance bandwidth, voltage standing wave ratio, radiation patterns, gain, and radiation efficiency. The results demonstrate that the proposed antenna achieves super-wideband operation with stable omnidirectional radiation behavior and satisfactory gain over the entire frequency range, making it well suited for high-speed and low-latency 5G communication scenarios. Surface current distribution analysis is performed at multiple representative frequencies to elucidate the underlying physical mechanisms responsible for bandwidth enhancement and multi-resonant behavior. Furthermore, the antenna’s temporal performance is assessed through group delay analysis, confirming minimal signal distortion and suitability for broadband and low-latency applications. A detailed comparison with recently reported wideband and fractal antenna designs highlights the advantages of the proposed antenna in terms of compactness, bandwidth enhancement, and overall performance. The results confirm that the proposed compact super-wideband fractal antenna represents a promising candidate for next-generation 5G wireless systems, offering an effective balance between miniaturization, bandwidth, and radiation performance.

EDGE OF THINGS BASED DIABETES PREDICTION USING MACHINE LEARNING

2026-02-07T17:34:38+05:00

Diabetes mellitus is a fast-increasing health issue in the world which needs to be diagnosed and managed in time to minimize complications and health expenses. Recent developments in machine learning have shown good prospects towards enhancing diabetes prediction, but, the majority of those existing solutions are based on centralized cloud models which are characterized by a great latency, privacy reasons, and inability to use them in resource-limited settings. This paper attempts to deal with these issues by suggesting an Edge of Things (EoT)-based diabetes predictive model based on the integration of hybrid deep learning models with ensemble machine learning algorithms to achieve decentralized and real-time prediction of diseases. The suggested structure will include extensive data preparation, features normalization, and the class imbalance to optimize predictive accuracy. Hybrid deep learning models are used to learn the complex nonlinear association among demographic, clinical and behavioral variables and ensemble learning methods exploit the synergistic advantages of multiple classifiers to enhance their robustness and generalization. The system will greatly decrease the response latency and bandwidth consumption, and will avoid relying on constant internet connection, which will also increase data privacy and security by deploying trained models on the edge. The experimental assessment of the publicly available diabetes data base shows that the hybrid-ensemble framework proposed is more accurate, sensitive, and stable as compared to the conventional one-model frameworks. The findings indicate that imbalance-conscious learning and edge-based intelligence is effective in healthcare analytics. On the whole, this research provides a solution to detecting diabetes at an early stage saving privacy and being scalable and efficient to support the proactive and individualized healthcare delivery in resource-restricted conditions.

IDENTIFICATION OF BRAIN TUMOR ON MR IMAGES USING ENSEMBLE LEARNING MODELS

2026-02-08T18:55:54+05:00

Brain tumor have become a serious health concern for human beings worldwide. It’s began with an abnormal growth in the brain size. According to the recent statistics brain tumor caused 246,253 deaths globally. In 2019, the Pakistan Brain Tumor Epidemiology Study (PBTES) has reported 2,750 cases of brain tumor. Manual identification of brain tumors in MRI scans is difficult, time consuming, and subject to variable diagnosis. That's why automated computer-aided systems are important in ensuring accurate and early detection. In the last few years, deep learning classifiers have been used for brain tumor detection, but the individual classifiers are not always consistent. To overcome this, we propose an ensemble as a hybrid approach. This approach based on five classifiers namely CNN, RF, SVM, KNN and LR. All models of machine learning are based on hybrid feature extraction to achieve better output and we use soft voting technique to combine the output of all classifiers for more reliable decisions. In this study we use a dataset of 4600 MRI images for validation. We also include another unseen dataset. On the validation data, the top accuracy for the ensemble is 97.5%. Experimental results on the unseen data (600 MRI images) directly show that the ensemble method is better than each individual model. The individual accuracies were: CNN 91.67%, RF 90.67%, SVM 90.83%, KNN 89.67% and LR 88.83%. The ensemble accuracy jumps to 97.17 % confirming the workability of the hybrid approach. This study shows that ensemble learning can dramatically enhance the performance of brain tumor detection, so it is a promising method that could be used in clinical decision support system.

Keywords: Pakistan Brain Tumor Epidemiology Study, Convolution Neural Network, Support vector machine, K- nearest neighbors, Gray Level Co-occurrence Matrix, K-Means Clustering.

ACUTE LYMPHOBLASTIC LEUKEMIA SUBTYPE CLASSIFICATION USING CONVOLUTIONAL NEURAL NETWORKS ON PERIPHERAL BLOOD SMEAR IMAGE

2026-02-07T15:35:48+05:00

Acute Lymphoblastic Leukemia (ALL) is the most common kind of cancer in children. It is caused by too many immature lymphoblasts growing in the bone marrow. Accurate subtype identification is crucial for timely and effective treatment. This work presents a deep learning approach for the automated classification of four diagnostic categories based on microscopic peripheral blood smear images: benign (hematogones) and three malignant subtypes of acute lymphoblastic leukemia (Early Pre-B, Pre-B, and Pro-B). Transfer learning was used to improve four pre-trained convolutional neural network architectures EfficientNet-B3, VGG16, DenseNet-121, and ResNet-50—on a dataset of 3,256 images. EfficientNet-B3 achieved the highest test accuracy of 98.57%, followed by VGG16 (98.37%), DenseNet-121 (97.76%), and ResNet-50 (95.92%). The proposed strategy demonstrates enhanced diagnostic precision and has considerable potential to reduce observer variability, minimize diagnostic errors, and expedite clinical decision-making in all screening and subtype identification processes.

Keywords: Acute Lymphoblastic Leukemia, Convolutional Neural Networks, Blood Smear Classification, Deep Learning, Automated Diagnosis

CYBER SECUTIRY: IMPORTANCE OF WHITE HAT HACKER IN DIGITAL ERA

2026-02-08T19:15:52+05:00

Due to rapid development and advancement in Technologies in this digital era. The main goal of this study is to analyses the types of hackers and cyberattacks in the any organization. For this Purposes In the research paper wo focus on different types of hackers and their aim to hacked any system. There are two types of hackers. The first type includes those who act responsibly and ethically to enhance the safety of people and organizations. These are often called white hat hackers, red team, blue team, green team, and nation-sponsored hackers. The second type consists of hackers who use cyberattacks with harmful intentions, causing significant damage to public and private organizations and consumers. Furthermore, another part of this study to differentiate between ethical hacker and unethical hacker. What is his objection while hacked any system what type of tool and techniques are used by these hackers. also highlighted the importance of white hat hacker in an organization along with the limitation of these hacker.

Keywords: Hacking, Cyber Security, Ethical Hacker, Cyberattacks, Types of Hackers

BIOMECHANICAL STRESS ANALYSIS OF TIBIAL BONE PLATE FIXATION SYSTEMS UNDER DIFFERENT LOAD SCENARIOS

2026-02-09T12:32:58+05:00

Tibia fractures in humans are typically the result of accidents with vehicles or falls. Biomedical implants are often necessary for patients of such accidents to help in the healing process. By improving the physiological state, bone plates and other biomedical implants enable elderly individuals and car accident patients to enjoy normal lives. When a human bone is exposed to an extreme load that is above its maximum capacity, it frequently fractures. Bone healing occurs in two basic ways: primary bone healing and Secondary bone healing. This research aims to investigate the stress distribution on bone and plate under various physiological conditions, including bending, torsion, compression, and extension, using two distinct materials for the plate and screws. A 3D model's mechanical strength may be seen by simulating various mechanical tests, such as von Mises stress and maximal first principal stress, using FEA software like COMSOL. Create a three-dimensional model of the screws, metal plate, and tibia bone, then allocate the materials to the model. After applying certain load and boundary conditions, the model is then meshed and analyzed. The results of the COMSOL Multiphysics simulation indicate that modest deformation and stress levels are caused by bending, torsion, compression, and extension, but combined loads cause high deformation and stress levels. The maximum deformation in the combined load condition is 25.299 mm in Ti alloy material, whereas the smallest deformation in compression is 0.1500 mm in the same material. In stainless steel, the lowest von Mises stress is displayed in compression at 3.5606x107 N/m², while the highest is displayed in combined load at 9.8428x108 N/m². Results showed that the titanium alloy Ti-6Al-4V exhibited better performance under combination loading compared to stainless steel, with lower stress concentrations and deformation. This information can be valuable for designing implants or structures that will be subjected to various loading conditions in real-world applications.

ON RELATION OF ALTERNATIVE LA-SEMIGROUP

2026-02-09T15:00:49+05:00

In this paper, we establish the relation of alternative LA-semigroup with existence subclasses of LA-semigroup. We prove that under what condition a left alternative LA-semigroup becomes right alternative, flexible, left, right and middle nuclear square LA-semigroups. By the use of Modern mathematical technique GAP and Mace4 we construct a non-associative example that left alternative is neither contain in right alternative, flexible, left, right and middle nuclear square LA-semigroups. Although left(right) alternative LA-semigroup is non-associative, but we prove that the combination of T¹and T⁴− LA-semigroups it becomes associative. We also find the relation of right alternative, Jordan, stein, LA*,LC, RC, middle and right nuclear square LA-semigroups.

A DEEP LEARNING ENSEMBLE FOR REFERABLE GLAUCOMA DETECTION: AN INDEPENDENT ANALYSIS OF THE JUSTRAIGS CHALLENGE DATASET

2026-02-06T15:24:57+05:00

Fundus imaging is a non-invasive method for screening retina and is widely used for diagnosis of ophthalmological diseases. For early detection of glaucoma, clinically relevant structural changes in the optic nerve head, neuroretinal rim, and peripapillary region can be visualized in fundus images. In this study, we investigate automated referable glaucoma detection using deep learning ensemble model trained on the publicly available JustRAIGS Challenge dataset. We have proposed an independent ensemble framework combining convolutional and transformer-based architectures, specifically EfficientNet-B3 and Swin Transformer Tiny, to leverage complementary feature representations. To address class imbalance between referable and non-referable cases, generative adversarial network–based data augmentation was employed. The proposed ensemble achieved competitive performance across clinically relevant evaluation metrics, demonstrating its potential for robust and scalable glaucoma screening.

Keywords:

Glaucoma screening; Fundus imaging; Deep learning ensemble; Referable glaucoma; JustRAIGS dataset; EfficientNet; Swin Transformer; Generative adversarial networks (GANs)

INVESTIGATING INFLUENCE OF SLANT ANGLES ON AERODYNAMIC PERFORMANCE OF HEAVY-DUTY BUSES (HDB) THROUGH COMPUTATIONAL FLUID DYNAMICS (CFD)

2026-02-09T15:40:04+05:00

Heavy-duty buses (HDBs) play a vital role in public transportation systems; however, their blunt body geometry results in significant aerodynamic drag, leading to high fuel consumption. Traditional experimental methods for evaluating HDB aerodynamic performance are time-consuming, resource-intensive, costly, and limit the number of design configurations that can be studied. Therefore, this study employs computational fluid dynamics (CFD) to analyse the impact of different slant angles on HDB aerodynamic performance and assesses the reliability of CFD predictions by comparing them with published experimental data. A three-dimensional bus model was created using ANSYS Design Modeller, and numerical simulations were conducted in ANSYS Fluent under appropriate boundary conditions. The rear tilt angle was systematically varied to investigate its effect on the drag coefficient. Results indicate that the rear tilt angle significantly influences aerodynamic drag, with the drag coefficient reaching its minimum at a 12.5° tilt angle. Furthermore, CFD calculations showed good agreement with experimental results, with deviations consistently below 6%. These findings validate CFD as a reliable and cost-effective tool for predicting the aerodynamic performance of heavy-duty buses.

THE IMPROVING SOIL HEALTH AND CROP PRODUCTIVITY THROUGH THE MICROBIAL AND BIOTECHNOLOGICAL APPROACH : THE ASSESSMENT OF COMPREHENSIVE SUSTAINABLE SOIL MANAGEMENT STRATEGY

2026-02-10T09:52:26+05:00

Sustainable management and Climate change of food and soil are closely interconnected, as climate change occurs, soil resources continue to deteriorate, while increases the population also increase pressure on food supplies. Sustainable soil management strategies allow agriculture producers to preserve or improve crop productivity even under challenging condition. Farming practice that rely heavily on synthetic fertilizers along with traditional tillage practice have created a negative impact on the quality of soil due to strong dependency on inputs. Environmental damages, like ecosystem disruption and contamination caused by there traditional practice, has decreases the access to safe and nutritious food for many populations. Microbial inoculans combined with modern innovations such as PGPR and AMF, provides promising alternatives to farming methods which are based on traditional inputs. These approaches increases the ability to support and maintain crop yields and ensuring the sustainability of agricultural production over the long term.

Objective

The purpose of this study was to find how microbial inoculants and existed technologies can impact on the productivity and soil quality such as yield and enhance resistance to abiotic stressors like salinity and drought while potentially reducing or removing reliance on the synthetic fertilizers.

Methodology

This research was a brief plan that incorporated both greenhouse trial and field experiment across multiple soil types. The experiment included an untreated control, single microbial inocultants of AMF and PGRP, combined microbial inoculants, and these experiments applies either alongside biochar or with prebiotic seed coatings.The following microorganisms were the dominant populations: Bacillus, Rhizobium, Azospirillum, and Rhizophagus intraradices. Soil health was assessed using the following parameters: microbial diversity, enzyme activities, soil organic matter, nutrient utilization efficiency, and soil structure. The metrics used in evaluating crop performance included yield, total biomass, seed germination, seedling vigor, AMF colonization, and salinity and drought resistance. All statistical analyses were conducted using ANOVA.

Results

Increased soil organic matter (15 to 20%), increased microbial biomass (C, N, & P), and greater enzyme activity resulted from the application of microbial treatments. Crop yields increased up to 40%, and also there was an increase in seed germination/vigor and AMF colonization of plants (>45 percent). The AMF-inoculated plants under saline conditions maintained a better ionic balance than control plants, and drought resistance was improved by the addition of a microbial consortium and biochar. Fertiliser used was reduced by approximately 33% , and diseases caused by soil-borne pathogens decreased by approximately 30%

Conclusion

Microbial Inoculations and biotechnology improve soil and plant health, improve crop yield and stress resistance and decrease reliance on chemical fertilisers, indicating their potential for use in sustainable agriculture.

OBSERVER-BASED SLIDING MODE APPROACH FOR SYNCHRONIZING MOTION IN LARGE AIRCRAFT HYBRID MECHATRONIC ACTUATION SYSTEM

2026-02-10T14:06:08+05:00

In hybrid mechatronic actuation systems comprising electro-mechanical and servo-hydraulic actuators, precise motion synchronization is imperative for reliable performance, particularly in large-scale civil aircraft applications. A lack of synchronization between actuators gives rise to the *Force-Fighting* phenomenon, which occurs when differential forces are produced while both actuators jointly drive the same control surface. This research investigates the mitigation of Force-Fighting and the enhancement of motion coordination within hybrid actuation systems. An observer-based Sliding Mode Control (SMC) framework is proposed to achieve robust motion synchronization. To address modeling uncertainties, actuator coupling effects, external disturbances, and the unavailability of complete state information, an Extended State Observer (ESO) is integrated into the control design. The effectiveness of the proposed scheme is validated through extensive simulations conducted in MATLAB/Simulink. Comparative results with existing methods demonstrate that the proposed control approach significantly improves motion synchronization accuracy, enhances load disturbance rejection capability, and effectively suppresses the Force-Fighting phenomenon in hybrid mechatronic actuation systems.

A REVIEW OF GEOPOLYMER MORTARS INCORPORATING INDUSTRIAL ASHES, CONSTRUCTION AND DEMOLITION WASTE (CDW), AND RECLAIMED ASPHALT PAVEMENT (RAP)

2026-02-11T13:32:56+05:00

The construction sector is facing greater challenges in reducing its impact on the environment, with the rising issue of waste that Construction and Demolition waste (CDW) and Reclaimed Asphalt Pavement (RAP) are creating for construction industries worldwide. An extensive research review on the existing technology for geopolymer ash-based mortar composites using locally available fine materials, with special consideration for building demolition waste and RAP, counted as more favorable cement replacement materials for geopolymer matrices, is conducted. Based on a broad analysis of 67 scholarly articles published between 2015 and 2026, this research presents a summary of existing technology with regard to three vital parameters: (a) the mechanical characteristics of geopolymer mortar composites, including compressive strength, flexural characteristics, durability, and workability; (b) the sustainability factors concerning the impact on the environment, through geopolymer potential for carbon footprint reduction, life cycle analysis, and sustainability assessment techniques; and (c) the chemical aspects concerning geopolymerisation kinetics, microstructure formation, and alkali activation mechanisms for geopolymer composites. The research findings suggested that geopolymer-based matrices using 25-50% demolition waste and RAP potentially have compressive strength values from 15 to 64 MPa with a reduction of 40-72% in carbon footprint compared to conventional PC-based systems. Geographic context, along with various application examples, such as implementation of the method with appreciable success in a variety of climatic conditions worldwide. From pavement-quality concrete to various forms of masonry mortars, the variety of structures is appreciable. Critical analysis of the field reveals existing research gaps with regard to the standardization of mix design methodologies and scaling issues. Review of existing literature offers an exhaustive framework for researchers working on the stream of sustainable construction materials through waste management practices.

MULTIMODAL GRAPH REPRESENTATION LEARNING FOR ROBUST SURGICAL WORKFLOW RECOGNITION WITH ADVERSARIAL FEATURE DISENTANGLEMENT

2026-02-11T15:29:47+05:00

Recognizing the workflow of surgeries is really important for automating tasks and making sure patients are safe. When the data gets corrupted it becomes a big problem. This document talks about an approach that uses graphs and combines what we see and the movement of things to make things more accurate even when conditions are tough. The Multimodal Disentanglement Graph Network or MDGNet for short looks at how what we see. The movement of things work together using a special framework to make sure the features match up. The Contextual Calibrated Decoder uses information about time and context to make the system more resilient to changes and corruption of data. This helps the Surgical workflow recognition system to work. The Surgical workflow recognition system is important, for safety and the Multimodal Disentanglement Graph Network helps it to work more accurately. The model achieved accuracies of 86.87% and 92.38% on two datasets, demonstrating effectiveness in addressing data corruption issues and advancing automated surgical workflow recognition.

MACHINE LEARNING AND DEEP LEARNING FOR SUSTAINABLE AGRICULTURE

2026-02-11T15:42:26+05:00

Recent digitalization has included increasing elements of artificial intelligence and Machine Learning into agriculture and Deep Learning to address the challenges brought about by population growth, Cli- mate change (CC) and Resource Limitation (RL). The present study comprehensively deals with the areas of potential applications of AI techniques. The innovations range from upstream to downstream in agricultural production, with an emphasis on those that conform to Climate-smart (CS) agricultural practices. A review of research articles was carried out, with the Application of Machine Learning and Deep Learning in crop selection, monitoring and land management, water, soil and nutrient malabsorp- tion, management, weed control, harvest and post-harvest practices, managing pests and insects, and soil management. The results highlight that ML and DL enable the analysis of complicated datasets, thereby informing data-driven decision-making, reducing dependence on subjective expertise, and enhanc- ing farm management strategies. Machine Learning and Deep Learning also offer immense opportunities in increasing agriculture productivity, sustainability, and resilience. By highlighting data-driven insights and embracing innovative technologies, the agricultural sector can transition toward more efficient, en- vironmentally sustainable, and economically feasible approaches to farming to contribute towards food globally.

PERFORMANCE EVALUATION OF MODULAR MULTILEVEL CONVERTER ARCHITECTURES FOR HIGH-VOLTAGE DIRECT CURRENT APPLICATIONS

2026-02-12T09:50:11+05:00

The increasing demand for efficient long-distance High Voltage Direct Current (HVDC) integrated with renewable energy resources is universally has been intensified around the globe. The modular Multilevel Converters (MMC) is its component due to its modular structure design, high efficiency, higher dynamic performance and low harmonic distortion. This paper offers a design and performance assessment of MMC architectures for HVDC applications, highlighting its real-time authentication through Hardware-in-the-Loop (HIL) environments. The study delivers a thorough investigation of MMC modulation strategies, and multilevel output synthesis. Using real-time simulation platforms, the simulation results validate the converters’ improved excellent power quality across a wide operating range. Overall, the results confirm that simulation and HIL supported development expressively fortifies the dependability, performance, and practical deployment of MMCs in modern HVDC transmission lines.

ENHANCING PERSONAL WELLNESS THROUGH AI: OBJECT IQ'S IMAGE-BASED DIETARY MONITORING AND PERSONALIZED RECOMMENDATIONS

2026-02-11T19:59:48+05:00

In recent years, the intersection of artificial intelligence (AI) and health technology has yielded powerful tools for enhancing personal wellness and disease prevention. One such application is dietary monitoring via image-based food recognition, which offers a convenient alternative to manual logging of meals and nutrition. This research introduces ObjectIQ, an AI-powered mobile application that utilizes multimodal large language models (LLMs) and advanced computer vision techniques to analyze food images and provide real-time nutritional insights. Unlike traditional diet apps that rely on text-based inputs and fixed databases, ObjectIQ incorporates GPT 4o in a ReAct (Reason + Act) agent framework to interpret images, identify food items, estimate caloric content, and generate recipe suggestions with minimal user input.The system architecture is designed with modularity and scalability in mind. It uses Flutter for cross-platform frontend development, FastAPI for backend API orchestration, Supabase for cloud storage and authentication, and SQLite for offline data persistence. At its core, the AI model processes the input image using GPT 4o's vision capabilities, performs reasoning through chain-of-thought prompting, and interacts with external tools and APIs to retrieve nutritional data. The application offers features such as automated food recognition, real-time calorie estimation, personalized recipe generation, history logging, PDF export, and a user-friendly interface optimized for both fitness and general users. Performance evaluations demonstrate high levels of accuracy, with Top-1 accuracy of 89.3%, Top-3 accuracy of 94.7%, and a mean inference time of approximately 2.1 seconds per image. User testing involving 25 participants revealed high satisfaction in usability, clarity, and practical benefit, with average usability ratings exceeding 4.6/5 across various metrics.

BRIDGING HUMAN INTELLIGENCE AND MACHINE INNOVATION THROUGH THE INTEGRATION OF AI AND CYBERNETICS

2026-02-11T19:56:17+05:00

The intersection point between Artificial Intelligence (AI) and Cybernetics offers a revolutionary basis of hybrid intelligence systems and combining human cognitive flexibility with machine accuracy, automation and scalability. Cybernetics, based on the principles of feedback-based control and communication put forward by Wiener, provides the mechanisms of control required to have an adaptive behavior, whereas AI provides computational learning, reasoning and autonomous decision-making. Although much has been achieved in both areas, there still exists a great gap in creating a cohesive system that could coordinate human intelligence and machine innovation in a dynamic ethically aligned system. The fundamental issue that will be discussed in this paper is that there is no integrated and feedback-based hybrid intelligence architecture where human cognition functions as a part and parcel and not as an overseer. To remedy this, the paper suggests a new AI-Cybernetic Integration Framework that includes three layers: a Cognitive Computation Layer that uses machine learning to simulate the behavior of the human patternry and predictive reasoning a Cybernetic Feedback Regulation Layer that allows self-correction and adaptive control in real-time, and an Ethical and Human-Centered Oversight Layer that makes sure that the value is aligned and responsible decision making. So far as we know, this framework is the first systematic framework that integrates cybernetic feedback, cognitive computation and ethical governance in a single adaptive architecture. The evaluation of the system performance was conducted using a mixed-methods approach that incorporated formal theoretical modeling, computational simulations based on multi-scenario analysis and comparative analysis based on the stability, accuracy and resilience metrics. Findings suggest that there are quantifiable improvements, where decision accuracy, operational stability, and uncertainty resilience improve by up to 15, 22, and 18 percent relative to non-cybernetic AI baselines. The main problems are interpretability of the models, calibration of trust, data privacy and possible overload of feedback. The paper suggests open feedback loop, dynamic thresholds and governance systems to address these problems. Comprehensively, the results support the view that cybernetic control coupled with AI learning helps to enhance the technological strength, elevate ethical responsibility and deliver substantial socioeconomic value, making hybrid intelligence a key value generator of sustainable innovation in the Fifth Industrial Revolution.

Keywords: Artificial Intelligence, Cybernetics, Human–Machine Integration, Ethics, Automation, Technological Innovation, Governance

TRANSFORMING GALLBLADDER CANCER SCREENING THROUGH DEEP LEARNING–ENABLED ULTRASOUND IMAGING: A PROSPECTIVE DIAGNOSTIC STUDY AT AYUB TEACHING HOSPITAL, ABBOTTABAD, PAKISTAN

2026-02-12T15:28:17+05:00

Gallbladder cancer (GBC) remains one of the most aggressive hepatobiliary malignancies, largely due to late-stage diagnosis and the absence of reliable, non-invasive screening strategies, particularly in low- and middle-income countries. Conventional ultrasound imaging is widely used as a first-line diagnostic modality; however, its effectiveness is highly operator-dependent and limited in detecting early-stage malignant changes. This study aims to evaluate the clinical utility of deep learning–enabled ultrasound imaging for the early screening and diagnosis of gallbladder cancer in a real-world tertiary care setting. A prospective diagnostic study was conducted at Ayub Teaching Hospital, Abbottabad, Pakistan, involving patients presenting with suspected gallbladder pathology. Ultrasound images were acquired using standardized imaging protocols and annotated by experienced radiologists. A deep learning framework based on convolutional neural networks was developed to automatically analyze ultrasound images and classify gallbladder lesions into malignant and non-malignant categories. The model was trained, validated, and tested using institution-specific datasets to ensure clinical relevance and robustness. Diagnostic performance was assessed using accuracy, sensitivity, specificity, precision, F1-score, and area under the receiver operating characteristic curve (AUC), with histopathology and expert consensus serving as reference standards. The proposed deep learning model demonstrated strong diagnostic performance, achieving high sensitivity and specificity in differentiating gallbladder cancer from benign conditions such as cholelithiasis and chronic cholecystitis. Notably, the AI-assisted system showed improved detection of subtle morphological features that are often overlooked in conventional ultrasound interpretation. Comparative analysis revealed that the deep learning–enabled approach outperformed routine ultrasound assessment, particularly in early-stage disease identification. The model also exhibited consistent performance across varying image qualities, highlighting its potential to reduce inter-observer variability and diagnostic subjectivity. This study provides prospective clinical evidence supporting the integration of deep learning–powered ultrasound imaging into gallbladder cancer screening workflows. The findings suggest that AI-assisted ultrasound can enhance diagnostic accuracy, facilitate early detection, and support clinical decision-making in resource-constrained healthcare environments. Adoption of such intelligent diagnostic systems may significantly improve patient outcomes through timely intervention and personalized management. Future work will focus on multi-center validation, explainable AI integration, and real-time deployment to further advance AI-driven gallbladder cancer screening.

CROSS-LINGUAL CYBER-BULLYING: A SYSTEMATIC REVIEW OF DETECTION METHODS

2026-02-12T22:46:17+05:00

The growing use of social media has led to a significant rise in cyber-bullying and hate speech, creating serious social and mental health challenges worldwide.Consequently, numerous automated detection methods have been developed, but their performance varies widely across languages, datasets, and modeling strategies. This paper reviews existing literature on state-of-the-art approaches to cyber-bullying and hate speech detection, with particular emphasis on multilingual and low-resource language settings such as Roman Urdu and English. The reviewed studies are analyzed across several dimensions, including dataset characteristics, preprocessing methods, and feature engineering techniques, followed by an evaluation of machine learning, deep learning, and transformer- based models. The findings indicate that traditional machine learning models provide a strong baseline but struggle with contextual and intent-aware detection. Deep learning approaches achieve improved performance, yet these approaches are still limited by data scarcity and dependence on binary classification. While transformer-based models demonstrate state-of-the-art performance, they struggle with emoji-aware processing, slang interpretation, and differentiating playful teasing from harmful cyber-bullying. By identifying key research gaps, this review underscores the importance of multilingual, emoji-aware, and intent-sensitive cyber-bullying detection frameworks, supporting further research and practical moderation systems.

VISIBLE-LIGHT PHOTOCATALYTIC REMOVAL OF DYES FROM TEXTILE EFFLUENTS

2026-02-13T11:17:06+05:00

Textile industries release dye-contaminated wastewater which is a dangerous environmental issue because synthetic dyes are highly stable, toxic, and non-biodegradable. A visible-light-active iron oxide (Fe₂O₃) photocatalyst was prepared in this research through a green, environmentally friendly, co-precipitation technique with the use of Azadirachta indica (neem) leaf extract as a natural reducing and stabilizing agent. The structure, morphology and optical characteristics of the prepared photocatalyst were analytically described by XRD, SEM and FTIR, UV-Vis spectroscopy, and showed the establishment of a crystalline phase of iron-oxide and surface functional groups that allow adsorption of the dyes and photocatalysis. Degradation of classic textile dyes was assessed under the irradiation of visible light to measure the photocatalytic performance of degrading the following dyes; Methylene Blue, Rhodamine B, and Congo Red. The findings revealed quick and effective dye degradation with the efficiencies of removal surpassing above 95% under the ideal of conditions, and according to pseudo-first-order kinetics. Radical scavenging studies indicated that the predominant reactive species was the superoxide radicals; the other reactive species that acted in the degradation process included hydroxyl radicals and photogenerated holes. The photocatalyst had high stability and reusability even after several cycles without much loss of activity. Notably, the workability of the system in terms of real textile wastewater was also tested, and the results of the experiment were almost 90 percent degradation efficiency with the complex effluent structure. The results indicate that green-synthesized Fe₂O₃ has a great potential to be used as an effective, sustainable, and cost-effective visible-light photocatalyst in the purification of textile wastewater.

MICROPLASTICS CONTAMINATION IN FRESHWATER SYSTEMS OF PAKISTAN: ECOLOGICAL RISKS AND POLICY IMPLICATIONS

2026-02-13T12:07:55+05:00

Microplastic (MP) contamination poses escalating threats to Pakistan's freshwater ecosystems, vital for irrigation, fisheries and drinking water. This study quantifies MPs across 25 sites in Swat River, Ravi River, Rawal Lake, Thal Canal and Indus distributaries, analyzing water, sediments and fish biota. The results indicate a high level of contamination, as indicated by the average measure for water of 42.3 MPs/L with peaks in urban areas reaching 192 MPs/L, an average for sediment of 182 MPs/kg dry weight compared to 450 MPs/kg for the Ravi surface, and an average for fish of 1.4 MPs/individual, composed mainly of PE/PP fragments that are less than 1 mm in size (62%). The amount of MPs present in urban sites were 3.2 times greater than in rural sites (p < 0.001) and there was a post monsoon increase of 62% due to surface runoff from wastewater and inadequate waste management. Sediments were functioning as sinks for MPs (PLI > 1.5 and in a degraded state), and species at risk from bioaccumulation of MPs include the Indus dolphin.. PCA identified urban discharge as primary driver (67% variance). Results exceed regional benchmarks, urging Pak EPA policy enhancements: tertiary wastewater treatment, polymer bans, and monitoring standards (<50 MPs/L). Low cost methods enable scalable assessments amid Pakistan's urbanization crisis.

A ROBUST WAVELET–ANN FRAMEWORK FOR NOISE-AWARE DETECTION AND CLASSIFICATION OF POWER QUALITY DISTURBANCES

2026-02-13T14:30:14+05:00

The rapid penetration of power-electronic-based loads, renewable energy sources, and sensitive digital equipment has significantly increased the occurrence and complexity of power quality disturbances (PQDs) in modern electrical power systems. Accurate and automated detection and classification of PQDs remain challenging due to the non-stationary nature of disturbance signals and the presence of measurement noise. This paper presents a comprehensive, noise-aware hybrid framework integrating discrete wavelet transform (DWT) based multiresolution analysis (MRA) with artificial neural network (ANN) classifiers for reliable detection and classification of PQDs.

Standardized single and combined PQ disturbance signals are generated in accordance with IEEE Std. 1159 and sampled at 10 kHz. DWT–MRA is employed for denoising, decomposition, and extraction of discriminative statistical features from multiple resolution levels. A systematic evaluation of diverse mother wavelet families is conducted to identify the most suitable wavelet for PQD representation. The extracted features are classified using multilayer perceptron (MLP), radial basis function (RBF), and probabilistic neural network (PNN) classifiers. Performance is evaluated under varying signal-to-noise ratio (SNR) conditions ranging from 20 dB to 50 dB.

Simulation results demonstrate that the proposed framework achieves superior and consistent classification accuracy across all disturbance types and noise levels. Comparative evaluation with recent state-of-the-art techniques confirms that the proposed wavelet–ANN approach provides a computationally efficient, interpretable, and highly accurate solution suitable for real-time power quality monitoring applications.

EARLY DIAGNOSIS OF ALZHEIMER’S DISEASE (AD) USING DEEP LEARNING TECHNIQUES

2026-02-14T12:12:52+05:00

Alzheimer's disease is a neurodegenerative disorder that progresses slowly and affects memory, and it is therefore important to diagnose it as early as possible to ensure it does not progress. Conventional diagnostic methods often fail to identify subtle structural and functional brain changes in the initial stages. To address this challenge, this research proposes a DL-based structure that employs a CNN for automated feature extraction and classification from MRI and fMRI scans. CNN effectively captures discriminative spatial patterns associated with early AD, enabling accurate differentiation between normal, mild cognitive impairment, and Alzheimer-affected brains. The performance of the model was evaluated by employing standard metrics. It is observed that the experimental results show the proposed CNN framework’s 94.2% accuracy is better than the traditional methods. This proves the robust nature of the CNN models in the early stages of AD. Furthermore, this approach offers a practical diagnostic tool that can support clinicians in timely interventions, with potential for further improvement through integration of multimodal neuroimaging and clinical data.

MACHINE LEARNING-BASED DOWNSCALING OF CLIMATE MODELS USING REMOTE SENSING AND GIS DATA FOR HIGH-RESOLUTION ATMOSPHERIC FORECASTING

2026-02-14T12:21:34+05:00

Climate prediction and atmospheric forecasting remain critical challenges in environmental science, particularly at high spatial resolutions where computational constraints limit traditional General Circulation Models (GCMs). This paper presents a comprehensive review and methodological framework for machine learning-based downscaling of climate models, integrating remote sensing and Geographic Information System (GIS) data to achieve high-resolution atmospheric forecasting. Statistical downscaling techniques have evolved considerably with the advent of deep learning architectures, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and generative adversarial networks (GANs). This research synthesizes current approaches, evaluates their efficacy across diverse geographic and climatic contexts, and proposes an integrated framework that leverages multi-source satellite data, topographic information, and historical climate records. The methodology incorporates advanced preprocessing techniques, feature engineering from GIS datasets, and ensemble learning strategies to address the inherent uncertainties in climate projections. Performance metrics demonstrate that machine learning approaches can achieve spatial resolutions of 1-4 km with significantly reduced computational costs compared to dynamical downscaling. Key findings indicate that hybrid models combining physical constraints with data-driven learning outperform purely statistical methods, achieving correlation coefficients exceeding 0.85 for temperature and 0.72 for precipitation variables. The framework addresses critical challenges including spatial transferability, temporal stability, and extreme event prediction. This work contributes to the growing intersection of artificial intelligence and climate science, offering practical insights for operational weather services, agricultural planning, and climate adaptation strategies.

TRANSFORMING NOC OPERATIONS THROUGH AI-AUGMENTED ALERT TRIAGE AND ESCALATION AUTOMATION

2026-02-14T14:35:17+05:00

Modern Network Operations Centers (NOCs) face significant challenges in managing the large volume of alerts generated by diverse monitoring systems. Manual triage processes, delayed escalation, and the absence of contextual intelligence often lead to prolonged incident resolution times and service degradation. This research proposes an AI-powered NOC Alert Triage and Escalation System that integrates microservice architecture, automated escalation mechanisms, and Large Language Model (LLM)-based analysis to improve alert handling efficiency. The proposed system leverages a FastAPI-based webhook service for real-time alert ingestion, PostgreSQL for persistent storage, RabbitMQ for asynchronous communication, and an Ollama-based LLM service for incident summarization and contextual knowledge enrichment.

Automated escalation is managed through a persistent scheduling mechanism to ensure reliability, even during system restarts. The Experimental evaluation demonstrates a reduction in the Mean Time to Acknowledge (MTTA), improved alert reduplication accuracy, and enhanced incident understanding through AI-generated summaries. The system is scalable, fault-tolerant, and customizable, making it suitable for enterprise-level NOC environments.