Journal of Science Engineering Technology and Management Science

The rapid advancement of digital technologies has transformed warehouse management and supply chain operations across industries. Traditional warehousing systems often face challenges related to inventory inaccuracies, operational delays, labor inefficiencies, and lack of real-time visibility. In this context, the integration of Artificial Intelligence (AI) and Internet of Things (IoT) technologies has emerged as a strategic solution for developing smart warehousing systems capable of enhancing operational efficiency and supply chain performance. The present study examines the role of AI and IoT-enabled smart warehousing practices in improving inventory management, warehouse automation, order fulfillment, and supply chain responsiveness. The research adopts a quantitative and analytical approach using primary data collected from warehouse employees, logistics professionals, and supply chain managers through structured questionnaires. The study evaluates the impact of AI-based predictive analytics, IoT-enabled tracking systems, warehouse automation, and real-time monitoring technologies on operational efficiency and supply chain performance. Statistical tools such as descriptive analysis, correlation analysis, and regression analysis are utilized to examine the relationships among the study variables. The findings reveal that smart warehousing technologies significantly improve inventory accuracy, reduce operational costs, enhance warehouse productivity, and strengthen supply chain coordination. AI-driven predictive systems and IoT-enabled real-time monitoring contribute positively to faster decision-making, efficient resource utilization, and improved customer satisfaction. The study further highlights that organizations adopting smart warehousing systems gain competitive advantages through enhanced operational flexibility and data-driven logistics management. The research contributes to the growing literature on digital supply chain transformation and intelligent warehouse management by integrating AI and IoT technologies within operational and strategic business frameworks. The study provides practical implications for logistics companies, warehouse managers, and supply chain professionals seeking to implement technology-driven warehousing solutions for sustainable operational excellence and improved supply chain performance.

Keywords : Smart Warehousing, Artificial Intelligence, Internet of Things, Supply Chain Management, Warehouse Automation, Operational Efficiency, Predictive Analytics, Logistics Management.

Author : K. Rithwik Kedari, Joshi Sriyagnya, Gunti Ganesh, Maram Yashwanth Lakshmi Narayana

Title : Smart Warehousing Using Artificial Intelligence and Internet of Things: An Empirical Analysis of Operational Efficiency and Supply Chain Performance

Volume/Issue : 2026;03(05)

Page No : 195-206

The The A new conception for Solar Base a water purification and electrical IoT base monitoring system to fulfill the wants for clean water and electricity in one integrated, autonomous and efficient system is bestowed during this project. The sublimate water and power receiver contain 2 devices one is the charge controller and the another is the voltage-current sensor. Then the battery is connected to the Arduino with another voltage sensor. The water purifier is connected to a battery to consume power for purifying the unpurified water.Using the water flow sensor for how much water purifies through the water purifierthe water flow sensor is connected to Arduino and also uses a pH sensor for the purified water pH measurement. If the water purifier level is not pure then the relay was on the solenoid valve for mixed purifier solution to purify the water which is purity by the purifier. All the sensors and relays are connected to the Arduino for processing all the data from the sensor. For the IoT monitoring system use the wifi, Blynk server, and mobile Apps. All processing data send to the web server for remote monitoring. purifier water is unsafe than having to get a notification for mixing the pH solution. Potential sites for Solar Power and water purification systems embrace both rural and urban areas. An initial analysis of this potential has been conducted for the development country case. This system work perfectly here got well power source from solar and a backup to the battery. Better to monitor pH solar voltage, battery charge, output voltage, output current, and how much water has been used

Keywords : ESP8266, pH, Turbidity, Conductivity, Solar Panel, Water Purifier, IoT.

Author : Ashish D. Barve1 , Shayarsingh L. Mangliya2 , Harshad A. Bhandekar3 , Pritam T. Somkuwar4 , Nikhil R. Vairagade5 , Rohit A. Khobragade 6 , Dr. Saurabh Lawate7 ,Prof. Manjeet Sakhare8

Title : IOT BASED SOLAR POWERED WATER QUALITY MONITORING AND ITS PURIFICATION

Volume/Issue : 2026;03(05)

Page No : 207-211

The purpose of this study is to investigate how people in the city of Chennai use Qcommerce. Quick commerce services are the subject of this study, which aims to shed light on customer tastes, habits, and the variables that drive their uptake. The research takes a quantitative tack by utilising primary data gathered from 150 participants chosen at random from a structured questionnaire. Factors including ease of use, speed of delivery, cost, and quality of service are extracted from the data. This descriptive study examines Chennai customer Q-commerce behaviour. Analysis uses primary and secondary data. A systematic questionnaire collects primary data from 150 respondents, while Q-commerce publications, websites, and reports provide secondary data. Basic statistical tools like percentage analysis and ANOVA are used to interpret the data. To evaluate Qcommerce consumer behaviour, the study examines convenience, delivery speed, cost, and service quality. The results show that customers prefer Q-commerce due to its convenience and quick delivery, but that problems like high delivery prices and a lack of product diversity are still there. Among urban customers in Chennai, Q-commerce is becoming more accepted, according to the survey.

Keywords : E-commerce trends, Q-commerce, consumer behaviour, quick delivery, customer satisfaction.

Author : Mrs. PARIMALAKANTHI.T, Dr. E. VISWANATHAN

Title : CONSUMER BEHAVIOUR PRACTICES TOWARDS Q-COMMERCE – A STUDY WITH REFERENCE TO CHENNAI CITY

Volume/Issue : 2026;03(05)

Page No : 212-217

The main objective of this research is to study the impact of team work on employee productivity in leather manufacturing industry in Vellore. The aim of the study is to gain an understanding of the impact of effective teamwork, coordination, communication and cooperation among the employees on their productivity and overall performance. It also aims at exploring the factors that facilitate or impede teamwork in the work place and to provide relevant recommendations for enhancing team effectiveness and boosting employee productivity in the leather manufacturing industry.The study is based on a descriptive research design. Primary data were collected from 300 employees working in selected leather manufacturing industries in Vellore using a structured questionnaire. Convenience sampling method was adopted for selecting respondents. The collected data were analyzed using statistical tools such as percentage analysis, mean score analysis, chi-square test, and correlation analysis.The study found that teamwork has a positive impact on employee productivity in the leather manufacturing industry. Effective communication and coordination among employees improve work efficiency and performance. Employees working in strong teams show higher motivation and job satisfaction. Poor teamwork and lack of cooperation reduce productivity levels.

Keywords : Teamwork, Employee Productivity, Leather Manufacturing Industry, Communication, Coordination, Job Satisfaction, Employee Performance, Group Dynamics, Workplace Cooperation, Organizational Efficiency

Author : DR.M.SUBRAMANIAN

Title : IMPACT OF TEAMWORK ON EMPLOYEES’ PRODUCTIVITY IN LEATHER MANUFACTURING INDUSTRY IN VELLORE

Volume/Issue : 2026;03(05)

Page No : 218-224

Digital technologies have brought an entire new level of sophistication and flexibility to Integrated Marketing Communication (IMC), which now brings all digital technologies enabling businesses to engage with consumers on an elevated and unprecedented level. The Aim of this study is to understand both the challenges faced in Integrated Marketing Communication (IMC) in the digital era and what role demographic elements play in the way consumers conceptualize digital IMC solutions. Methods: I used primary data obtained using a standardized questionnaire among 250 respondents. The data was collected via convenience sampling approach. To access the data and ascertain significant relationships with demographic variables and consumer perspective, statistical analysis, i.e. percentage and Analysis of Variance (ANOVA), was conducted. The role of digital IMC for providing greater opportunities such as enhanced customer interaction with the brand, tailoring of customization to individual customer and increased distribution is also demonstrated by the study. Of course, concerns like data privacy, information overload, and a lack of trust in digital platforms came to the surface as well. The Findings are Demographic factors like age and educational qualification are shown to heavily influence the attitude of consumers towards digital IMC (ANOVA). As such, digitalization has empowered and assisted IMC authorities, but it requires the establishment of communication strategies as well as sustainable competitive advantage.

Keywords :

Author : DR.K.KARTHIKEYAN

Title : INTEGRATED MARKETING COMMUNICATION IN THE DIGITAL AGE: OPPORTUNITIES AND CHALLENGES IN TAMIL NADU

Volume/Issue : 2026;03(05)

Page No : 225-232

Purpose;The purpose of this study is to examine how consumers in Chennai City purchase organic food items, with an emphasis on identifying the variables that affect their choices, including price, availability, brand trust, awareness, and health consciousness.Design / Methodology / Approach: The study uses a quantitative technique and a descriptive research design. Convenience sampling was used to gather primary data from 284 respondents in various Chennai zones.Consumer attitudes and behaviour were assessed using a variety of statistical methods, such as correlation, chi-square testing, Likert scale analysis, and percentage analysis.Findings: The results show that consumers typically have a favourable attitude toward organic food, primarily due to perceived safety and health benefits; 53% of customers prefer organic products, and 64% think they are safer than conventional food.However, expensive prices stand out as a significant deterrent, deterring 58% of respondents from making frequent purchases. Other factors that influence purchasing decisions include restricted availability and a partial lack of faith in certifications.Overall, the study reveals a discrepancy between consumer intention and actual purchasing behaviour, indicating the need for increased accessibility, affordability, and trust in order to increase the consumption of organic food.

Keywords :

Author : DR.A.MAHENDRAKUMAR

Title : A STUDY ON CONSUMER BUYING BEHAVIOUR TOWARDS ORGANIC FOOD PRODUCTS IN CHENNAI CITY

Volume/Issue : 2026;03(05)

Page No : 233-242

India is the world’s largest dairy producer, Indian dairy sector has grown substantively over the year. Dairy products demand in India has in india has increased dramatically in both rural and urban sectors. Tamil nadu state is the one of the ten largest milk producing states in India. In the state major milk contributor is Aavin a Tamil Nadu based milk producer’s union procures milk, processes it and sell milk and milk products to consumer. This paper analysis consumer perception over the Aavin special reference to the Nagapattinam City of Tamil Nadu state. The aim of the study is revealing consumer perception over Aavin milk and milk products based on their age, educational qualification and monthly income of the consumer’s family.

Keywords :

Author : Dr. PALANIVEL.M.D

Title : A STUDY ON CONSUMER BEHAVIOUR OF AAVIN MILK AND MILK PRODUCTS WITH SPECIAL REFERENCE TO TIRUPUR CITY

Volume/Issue : 2026;03(05)

Page No : 243-249

Technology and internet use have made online buying part of modern consumer behaviour. This study analyses Chennai customers' perceptions of internet purchasing, particularly Amazon. Understanding customer preferences, contentment, and online shopping platform perceptions is the major goal. Main data was acquired from 200 respondents utilising a standardised questionnaire. A simple random sampling method ensures unbiased participant selection. The data covers convenience, cost, product variety, delivery services, and online transaction trust. The statistics show that most customers prefer internet shopping for its ease, extensive selection, and time savings. Amazon's fast delivery, simple returns, and user-friendly interface are praised. Some respondents mention product quality discrepancy and delivery delays. The study found that Chennai district customers like online shopping, with Amazon influencing their purchases. Continuously improving service quality and customer trust may boost user satisfaction and loyalty, according to study.

Keywords :

Author : ASHWINI .S, Dr. K. MURUGAN

Title : CUSTOMERS PERCEPTION TOWARDS ONLINE SHOPPING IN CHENNAI DISTRICT – WITH REFERENCE TO AMAZON

Volume/Issue : 2026;03(05)

Page No : 250-254

The Saoras, an ancient tribal community of Odisha, occupy a vital place in the state’s cultural and spiritual history. Belonging to the proto-Australoid stock and speaking the Mundari language “Sora,” they preserve a rich heritage rooted in tradition. A deep connection links the Saoras to the Jagannath cult, as legend holds that Lord Jagannath was originally a tribal deity later adopted into mainstream worship. This association is evident in their shared medicinal and ritual practices. The Daitapatis of Puri, believed to be Saora descendants, use traditional herbal remedies during the Anasara period of Lord Jagannath—methods strikingly similar to Saora ethnomedicine. Such parallels reveal the tribe’s profound influence on Odisha’s spiritual life. The Saoras thus represent more than an indigenous community; they are custodians of ancient knowledge that continues to shape Jagannath traditions, symbolizing the harmonious coexistence of tribal and classical elements in Odisha’s cultural identity.

Keywords : Soara, Antebasi, Sagories, SabaRoye, Nabakalebara, Daitapati, Anasara , Dasamula modak, Phuluri oil . Saoras Ancient tribe, Proto-Australoid stock, Mundari language, Sora language, Jagannath cult, Indigenous knowledge, Inclusive heritage, Shared tradition

Author : Mrs. Sasmita Pati, Dr. C. Kalidass, Dr. Pratap Chandra Panda

Title : The Enigmatic Saura Culture: Its Sacred Ties to the Rituals of Lord Jagannath

Volume/Issue : 2026;03(05)

Page No : 255-260

Process Safety Management (PSM) structures that are particularly focused on mechanical systems working in highrisk industrial settings. Mechanical systems, including pressure vessels, piping networks, rotating equipment, and structural components are used in harsh environment conditions, which is both mechanical and thermal, where failure may lead to disastrous s key regulatory and functional safety models, such as mechanical integrity schemes, risk assessment schemes, Safety Integrity Level (SIL) and lifecycle-based management systems. It examines how hazard identification, risk assessment, change management, reliability engineering, and asset integrity management are integrated into structured PSM systems of safety culture, evidence-based PSM, dynamic risk analysis, and practical implementation strategies, connection between the practice of mechanical engineering and process safety principles to guarantee the active reduction of risks. that good mechanical integrity, functional safety, and monitoring integration improve reliability, avoid loss of containment and improve the performance of the industrial safety in various sectors.

Keywords : Mechanical Integrity, Risk Assessment, Process Safety Management (Reliability Engineering, Hazard Analysis.

Author : Mr. Kapil Ahir

Title : A Review of Process Safety Management Frameworks for Mechanical Systems

Volume/Issue : 2026;03(05)

Page No : 261-270

Voice-based scams have become a serious cybersecurity threat because people increasingly adopt artificial intelligence and communication technologies. Fraudsters use vishing, and robocall and deepfake voice attacks to trick victims into disclosing their personal information and authorizing fake transactions. The review presents an in-depth examination of the evolving operating environment of voice scams and the development of detection and prevention technologies. The study examines various types of voicebased fraud, with a focus on AI-generated voices and social engineering techniques. The research paper investigates methods for enhancing speech signals by removing noise and processing audio to improve sound quality and intelligibility for precise analysis. The review examines the various analytical features that detect scams through acoustic, linguistic and semantic patterns, as well as speaker verification and emotion analysis capabilities. The paper investigates how ML techniques, such as supervised learning, unsupervised learning, DL and hybrid ensemble models, detect fraudulent voice interactions. Modern voice scam detection systems use speech processing, NLP and cutting-edge machine learning techniques to detect suspicious calls in real phone conversations. The review presents information about existing obstacles, recent progress and upcoming research paths for voice-based scam detection systems.

Keywords : Voice Spoofing Detection, Telecom Fraud, Speech Analytics, Natural Language Processing (NLP), Machine Learning

Author : Mr. Ram Pratap Singh

Title : A Review on Voice-based Scam Detection Using Machine Learning and Speech Analytics

Volume/Issue : 2026;03(05)

Page No : 271-279

Different types of steel fibers can be used to reinforce concrete. Steel fibers are generally classified depending on their manufacturing method. The addition of steel fibers to concrete necessitate an alteration to the mix design to compensate for the loss of workability due to the extra paste required for coating the surface of the added steel fibers. While many technical and economical advantages are benefited from using SFRC, drawbacks can also be found. They are however not likely to cause major problems. It was thought that steel fibers will have negative implications in concrete practice (i.e. transporting, surfacing, finishing etc), but experience has shown that the influence of steel fibers on these practical aspects is negligible. This research is based on investigating the effect of crimped steel fibers on strength parameters and durability of concrete for M-40 grade using fibers crimped fiber. This study is to study the compressive strength, Split tensile strength, flexural strength, and durability of steel fiber reinforced concrete (SFRC) containing fibers of 0%, 0.5%, 1%, 1.5%, 2%, and 2.5% by volume of cement.

Keywords : Steel fibers, M-40 grade, compressive strength, Split tensile strength, flexural strength, Durability.

Author : DASANGAM SRAVANI1 D IBRAHEEM KHALEELULLA2

Title : STRENGTH AND DURABILITY ANALYSIS OF STEEL FIBRE REINFORCED FLY ASH CONCRETE OF M40 GRADE

Volume/Issue : 2026;03(05)

Page No : 280-284

High-performance concrete is defined as concrete that meets special combinations of performance and uniformity requirements that cannot always be achieved routinely using conventional constituents and normal mixing, placing, and curing practices. Ever since the term high-performance concrete was introduced into the industry, it had widely used in large-scale concrete construction that demands high strength, high flowability, and high durability. A high-strength concrete is always a highperformance concrete, but a high-performance concrete is not always a high-strength concrete. Durable concrete Specifying a high-strength concrete does not ensure that a durable concrete will be achieved. It is very difficult to get a product which simultaneously fulfill all of the properties. So the different pozzolanic materials like Ground Granulated Blast furnace Slag (GGBS), silica fume, Rice husk ash, Fly ash, High Reactive Metakaolin, are some of the pozzolanic materials which can be used in concrete as partial replacement of cement, which are very essential ingredients to produce high performance concrete. So we have performed XRD tests of these above mentioned materials to know the variation of different constituent within it. Also it is very important to maintain the water cement ratio within the minimal range, for that we have to use the water reducing admixture i.e superplasticizer, which plays an important role for the production of high performance concrete. So we herein the project have tested on different materials like rice husk ash, Ground granulated blast furnace slag, silica fume to obtain the desired needs. Also X-ray diffraction test was conducted on different pozzolanic material used to analyse their content ingredients. We used synthetic fiber (i.e Recron fibe) in different percentage i.e 0.0%, 0.1%, 0.2%, 0.3% to that of total weight of concrete and casting was done. Finally we used different percentage of silica fume with the replacement of cement keeping constant fiber content and concrete was casted. In our study it was used two types of cement, Portland slag cement and ordinary Portland cement. We prepared mortar, cubes, cylinder, prism and finally compressive test, splitting test, flexural test are conducted. Finally porosity and permeability test conducted. Also to obtain such performances that cannot be obtained from conventional concrete and by the current method, a large number of trial mixes are required to select the desired combination of materials that meets special performance.

Keywords :

Author : KARANAM SREEVATHSA1 , B FAYAZ2

Title : EXPERIMENTAL INVESTIGATION ON HIGH-PERFORMANCE FIBRE-REINFORCED CONCRETE USING POZZOLANIC MATERIALS

Volume/Issue : 2026;03(05)

Page No : 285-291

The rapid growth of cloud computing, big data and enterprise applications has made database administration an essential component of modern information systems. This review paper gives a comprehensive survey on optimization techniques in SOQL, MQL and NoSQL-based database systems to improve performance, scalability and efficient data management. The study discusses the roles of relational and non-relational databases in handling structured, semi-structured, and unstructured data in modern applications. It explores crucial optimization techniques such as query optimization, indexing, caching, replication, partitioning, aggregation, and sharding, which improve query execution and resource utilization. The paper also discusses performance characteristics of Salesforce Object Query Language (SOQL), Mongo Query Language (MQL) and NoSQL systems, highlighting their strengths, limitations and practical applications in cloud-native and enterprise environments. The current challenges of consistency, distributed transactions, interoperability and security in large-scale database systems are also discussed in the review. Emerging directions for intelligent and adaptive database management systems are also discussed, including AI- and machine-learning-based query optimization, autonomous database administration, and hybrid SQL–NoSQL architectures.

Keywords : Database Administration, SOQL, MQL, NoSQL Databases, Query Optimization

Author : Mr. Deepak Mehta

Title : Optimizing Database Administration: A Survey of SOQL, MQL, and NoSQL-Based Techniques

Volume/Issue : 2026;03(05)

Page No : 292-298

In software development, analyzing large volumes of bug reports is a challenging and time-consuming task. This paper proposes an ensemble machine learning-based model combined with Natural Language Processing (NLP) techniques to automatically predict the nature of bug reports. Multiple classifiers, including Logistic Regression, Naïve Bayes, Support Vector Machine, and Random Forest, are integrated using a voting mechanism to improve prediction accuracy. Experimental results demonstrate that the proposed system significantly enhances classification performance and reduces manual effort in software maintenance.

Keywords : Bug Reports, Ensemble Learning, Natural Language Processing (NLP), Machine Learning, Voting Classifier, Software Maintenance, Text Classification, Random Forest, Support Vector Machine (SVM), Logistic Regression

Author : Ch. Satyanarayana Reddy1, K.Pavani2, R.Mounika3

Title : Enhancing Software Maintenance Through Ensemble Learning-Based Bug Report Prediction

Volume/Issue : 2026;03(05)

Page No : 299-308

Vehicular Ad Hoc Networks (VANETs) generate massive volumes of real-time communication data, including packet size, packet rate, latency, vehicle speed, signal strength, and message transmission information. Due to the open wireless communication environment, decentralized architecture, and highly dynamic topology of VANETs, these networks are particularly vulnerable to cyberattacks, malicious intrusions, and anomalous communication activities. Traditional intrusion detection approaches primarily rely on rule-based systems and statistical threshold methods, which often lack adaptability and struggle to identify emerging attack patterns in real-time scenarios. To address these limitations, the proposed framework introduces a hybrid Machine Learning (ML) and Deep Learning (DL)-based intrusion detection system utilizing Classification and Regression Trees (CART) for enhanced attack detection and anomaly prediction. The framework incorporates multiple algorithms, including Linear-CART, Decision Tree-CART, Passive Aggressive-CART, and the K-NeuroFusion CART model. The K-NeuroFusion architecture combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to extract spatial and temporal features from vehicular communication data, while K-Nearest Neighbors (KNN) further improves classification performance. Additionally, the proposed Temporal Echo Fusion Network (TEFN) employs an Echo State Network Decision Tree Cost Complexity Pruning (ESN-DTCCP) model to perform advanced temporal pattern analysis and anomaly score prediction. The framework supports both classification and regression tasks for identifying normal and malicious communication behaviors while estimating anomaly severity levels. Furthermore, it provides single and batch prediction capabilities along with visualization and exploratory data analysis modules. Implemented using Flask, SQLite, TensorFlow, Scikit-learn, Pandas, NumPy, and Matplotlib, the proposed system offers an accurate, scalable, and efficient solution for intelligent VANET security and anomaly detection.

Keywords : Vehicular Ad Hoc Networks (VANETs), Intrusion Detection System, Deep Learning, Echo State Network (ESN), Network Security, Anomaly Detection

Author : T. Harshitha1, B. Ravikumar1, Ch. Pavani1, D. Shilpa1

Title : Context-Aware Federated Learning for Intelligent Threat Detection in Software-Defined Vehicular Networks

Volume/Issue : 2026;03(05)

Page No : 292-303

This study examines The Hungry Tide by Amitav Ghosh and Earthlings by Sayaka Murata through the perspectives of ecocriticism and posthumanism. The research focuses on the growing ecological instability, emotional alienation, and crisis of human identity reflected in contemporary literature. Although environmental concerns have become increasingly important in literary studies, comparatively little attention has been given to the ways contemporary Indian and Japanese fiction represent ecological vulnerability, survival anxiety, and the changing relationship between humans and the non-human world. The study mainly aims to explore how the selected novels portray ecological crisis and human survival while also questioning anthropocentric ideas of civilisation and identity. The research follows a qualitative and analytical approach based on close textual reading and comparative literary analysis. Ecological imagery, symbolic patterns, psychological fragmentation, and representations of survival have been examined with the support of ecocritical, posthumanist, and environmental humanities perspectives. The analysis reveals that The Hungry Tide presents the Sundarbans as a powerful ecological space where human survival depends upon coexistence, adaptation, and environmental awareness. In contrast, Earthlings explores psychological alienation and emotional instability within a mechanised and socially oppressive modern civilisation. Despite their different narrative structures and cultural settings, both novels expose the fragile condition of human existence and challenge the assumption of human superiority over nature. The study further suggests that ecological crisis extends beyond environmental destruction and increasingly affects emotional life, social relationships, and human consciousness. Through ecocritical and posthuman perspectives, the novels ultimately emphasise the need to reconsider rigid human-centred worldviews and recognise the interconnectedness between humanity and the natural world.

Keywords :

Author : Dr. Hemantkumar Babubhai Pargi

Title : Nature, Survival, and Posthuman Anxiety: An Ecocritical Comparative Study of The Hungry Tide and Earthlings

Volume/Issue : 2026;03(05)

Page No : 314-325

Due to increased incidence of obesity, diabetes, cardiovascular disease, hypertension, and other lifestyle-related diseases, health and nutrition are major concerns today. Calorie counters, activity trackers, and recipe apps are readily available, but most of them gather data but don't provide personalized advise or consider a user's medical problems. Results are a torrent of generic recommendations that normal users struggle to apply. Based on health concerns, the AI-Powered Nutrition Advisor provides recommendations online. individualized Machine meal Learning algorithms, notably K-Means Clustering, evaluate curated food databases to recommend meals for chronically ill users. Diabetes, Heart Disease, High Blood Pressure, Liver Disease, Stroke, and Kidney Disease are supported. The system gives severity based precautions (Low, Medium, High) and supports numerous health problems with consistent dietary advice. The Python Flask-based program provides real-time recommendations via a responsive web interface on any device.

Keywords : Artificial Intelligence, Personalized Nutrition, Machine Learning, K-Means Clustering, Diet Recommendation System,

Author : Ch. Satyanarayana Reddy1, K.Pavani2, T. Siva Anil3

Title : SMART NUTRITION RECOMMENDATION SYSTEM

Volume/Issue : 2026;03(05)

Page No : 326-332

Mobile Ad Hoc Networks (MANETs) are decentralized wireless communication networks where mobile nodes communicate dynamically without relying on fixed infrastructure. Due to their open medium, dynamic topology, limited bandwidth, and lack of centralized administration, MANETs are highly vulnerable to security threats such as packet dropping, black hole attacks, wormhole attacks, data interception, and unauthorized access. Ensuring robust and secure data transmission in such environments has become a major challenge in modern wireless communication systems. This project proposes a robust and secure data transmission model using Artificial Intelligence (AI) techniques in Ad Hoc Networks. The system integrates Machine Learning (ML) and intelligent routing mechanisms to improve network security, reliability, and transmission efficiency. AI algorithms are used to analyze node behavior, detect malicious activities, predict network congestion, and select optimal routing paths dynamically. The proposed approach continuously monitors network traffic patterns and identifies abnormal activities in real time, thereby reducing packet loss and improving secure communication. The system employs intelligent intrusion detection and trust-based routing techniques to enhance protection against cyberattacks. AI-based decision making enables the network to adapt quickly to topology changes and node mobility while maintaining stable communication. Encryption mechanisms and authentication protocols are also incorporated to ensure confidentiality, integrity, and secure access to transmitted data. Experimental analysis demonstrates that the proposed AI-driven approach achieves higher packet delivery ratio, lower transmission delay, improved throughput, and better attack detection accuracy compared to traditional routing methods. The system is scalable, adaptive, and suitable for applications such as military communication, disaster recovery, emergency response systems, and IoT-based wireless environments. In conclusion, the integration of Artificial Intelligence techniques with Ad Hoc Networks provides an effective solution for achieving robust, reliable, and secure data transmission. The proposed system significantly enhances network performance and security while addressing the challenges associated with dynamic and infrastructure-less wireless communication environments.

Keywords :

Author : Priyadarshni Hiremath1 C.Navya2

Title : Air Qua Net A Convolution Neural Network Model with Multi-Scale Feature Learning

Volume/Issue : 2026;03(05)

Page No : 292-297

Autonomous vehicles require intelligent decision-making systems to ensure safe, efficient, and reliable driving in dynamic traffic environments. Traditional autonomous driving approaches primarily rely on external factors such as road conditions, traffic flow, and the behavior of surrounding vehicles. However, these methods often overlook the influence of internal vehicle parameters that can significantly impact driving performance and decision quality. To address this limitation, this study proposes an Intelligent Driving Decision Strategy (DDS) based on Machine Learning for autonomous vehicles. The proposed DDS framework integrates both internal and external driving factors to generate optimal driving decisions in real time. Sensor data collected from the vehicle and its environment are processed and supplied to a Genetic Algorithm (GA), which identifies optimal parameter values and enhances prediction accuracy. The optimized features are then utilized by the DDS model to support intelligent decision-making and route selection. To evaluate its effectiveness, the proposed DDS framework is compared with widely used machine learning techniques, including Random Forest (RF) and Multi-Layer Perceptron (MLP). Experimental results demonstrate that the DDS approach achieves higher prediction accuracy, improved decision quality, and faster convergence than the benchmark models. The findings indicate that the integration of genetic optimization with machine learning significantly enhances autonomous vehicle decision-making capabilities, making the proposed DDS framework a promising solution for next generation intelligent transportation systems.

Keywords :

Author : K. Sathyanarayana Reddy ¹, K . Pavani 2, G.Likhitha3

Title : A Predictive Machine Learning Approach for Autonomous Vehicle Driving Decisions

Volume/Issue : 2026;03(05)

Page No : 314-320

The global energy landscape is undergoing a profound structural transformation, driven by the dual imperatives of mitigating climate change and ensuring national economic security. For emerging market economies, the transition away from fossil-fuel-based internal combustion engine (ICE) transport systems is no longer merely an ecological choice; it is a macroeconomic necessity As the world's third-largest consumer of energy, India stands at a critical juncture where its developmental aspirations are directly constrained by its energy dependencies. This study establishes a comprehensive macroeconomic and financial framework to evaluate the transition toward sustainable alternatives—specifically green hydrogen and electric mobility—assessing their economic viability, structural transition costs, and strategic roles in reshaping India's energy future. 1.1. The Conundrum of Fossil Fuel Dependency : The foundational vulnerability of the Indian macroeconomy lies in its deep, structurally embedded reliance on imported crude oil. India imports over 85% of its crude oil requirements to sustain its domestic transportation and industrial sectors.

Keywords :

Author : Dr M Bhaskar

Title : The Economic Viability and Transition Costs of Alternative Fuels: Evaluating Green Hydrogen and Electric Mobility as Substitutes for Diesel and Petrol in India"

Volume/Issue : 2026;03(05)

Page No : 321-335

The rising need for environmentally friendly building materials has led to interest in using industrial by-products like steel slag in concrete. This study looks at how the mechanical and workability qualities of concrete change when some of the natural fine particles are replaced with steel slag in different amounts (0%, 10%, 20%, and 30%). We did experiments on standardised concrete samples to see how they held up under different conditions, such as slump, compressive strength, split tensile strength, and flexural strength. The results show that replacing 20% of the fine aggregate with steel slag gives the best performance. It increases strength while keeping workability at an appropriate level. Sieve examination showed that the grading of steel slag is good for use in concrete. During the mixing, casting, and curing procedures, both visual and laboratory investigations confirmed that the concrete mix was consistent and strong. The results suggest the use of steel slag as a good substitute for natural sand, which helps conserve the environment and save resources. This study adds to the growing number of studies that support using circular economy methods in the building industry. It is suggested that more research be done to look at how well it lasts and how well it works in different climates and under different loads.

Keywords : Steel Slag, Sustainable Concrete, Fine Aggregate Replacement, Compressive Strength, Flexural Strength, Workability, Green Construction, Sieve Analysis.

Author : SHAIK.SYDHA, Dr.T.V.S VARALAKSHMI

Title : Experimental Investigation on the Utilization of Steel Slag as Partial Replacement of Fine Aggregate in Concrete

Volume/Issue : 2026;03(05)

Page No : 336-346

This study focuses on forecasting national-level self-harm trends by analyzing emotional patterns extracted from social media data. With the increasing influence of digital communication, individuals often express their mental states through online platforms, making them a valuable source for understanding psychological well-being. The proposed approach integrates sentiment analysis from user-generated content with official health statistics to build a predictive dataset. Various machine learning models, including ARIMA, Support Vector Machine, Bayesian Ridge, XGBoost, Random Forest, and Decision Tree, are applied to estimate injury and death trends related to self-harm. The results indicate that ensemble methods such as XGBoost and Decision Tree provide more accurate predictions with lower error rates. This framework can assist authorities in identifying at-risk populations and implementing timely intervention strategies, ultimately contributing to improved mental health awareness and prevention efforts.

Keywords : Self-Harm Prediction, Social Media Analysis, Sentiment Analysis, Machine Learning, XGBoost, Decision Tree, Time Series Forecasting, Mental Health Analytics

Author : Y. Shakuntala1 , Mohammed Zubair Ahmed2 , Syed Raheem Ahmed3 , Syed Eeshan4 , Mudassir Shareef5

Title : Data-Driven Prediction of Self-Harm Trends Using Social Media Signals and Machine Learning Techniques

Volume/Issue : 2026;03(05)

Page No : 347-353

This paper presents an automated system for detecting safety helmet usage in industrial environments using deep learning– based object detection techniques. Ensuring that workers wear helmets is essential for preventing head injuries, yet maintaining compliance remains a significant challenge. To address this, the proposed system utilizes an enhanced model based on the YOLO architecture, referred to as YOLOESCA, which is designed to improve detection accuracy under complex conditions such as varying lighting, occlusions, and diverse worker poses. The system provides a complete pipeline including dataset upload, preprocessing, model training, and evaluation. Performance is measured using metrics such as precision, recall, and mean Average Precision (mAP), along with graphical analysis of training progress. In addition, an extended YOLOv11 model is implemented to further enhance detection capability. Comparative results show that the extended model achieves better accuracy and robustness than the baseline YOLOv5 approach. The system also supports real-time helmet detection from both images and videos, making it suitable for practical deployment in surveillance systems. By automating helmet compliance monitoring, the proposed solution reduces the need for manual supervision and improves workplace safety. Overall, this work demonstrates the effectiveness of deep learning in developing scalable and efficient safety monitoring systems for hazardous environments.

Keywords : Safety Helmet Detection, YOLOESCA, YOLOv5, YOLOv11, Object Detection, Deep Learning, Computer Vision, Industrial Safety, PPE Compliance, Real-Time Monitoring, Image Processing, Video Surveillance, mAP Evaluation, Workplace Safety.

Author : Nagma Arshin1 , Omer2 , Mohammed Fateen3 , Mohammed Asad Uddin4 , Mohammed Ahsan Aali5

Title : Intelligent Vision-Based Framework for Automated Detection of Safety Helmet Usage in Dynamic Workspaces

Volume/Issue : 2026;03(05)

Page No : 354-360

The swift growth of a number of Internet of Things, or IoT, devices has improved connectivity and convenience in sectors such as healthcare, home automation, and industry. Nevertheless, this heightened connectivity renders IoT systems more susceptible to cyberattacks. Conventional security measures are inadequate in identifying contemporary threats related to denial-of-service, data exfiltration, and man-inthe-middle assaults due to the evolving nature of these risks. This project devises a cyber-assault detection system utilizing both deep learning and machine learning methods to address the issue. The model is evaluated using an extensive dataset comprising both normal & attack-related network data. Various techniques, including Random Forest, Support Vector Machine, Logistic Regression, and CNN2D, are employed to train the system. Their performance is assessed utilizing accuracy, precision, recall, and additional criteria. The findings indicate that Random Forest and SVM provide commendable performance; however, Logistic Regression and CNN2D attain the best accuracy, approximately 99.78%. This indicates that simplistic models can operate swiftly, whereas deep learning methods can yield more precise outcomes. The suggested approach offers a dependable and effective method for detecting cyber assaults in IoT networks. It enhances security by early threat identification and is beneficial for safeguarding real-world IoT systems.

Keywords : IoT Security, Cyber Attack Detection, Machine Learning Techniques, Deep Learning Models, Random Forest, Support Vector Machine (SVM), Logistic Regression, CNN2D, Network Security, Intrusion Detection, Data Processing.

Author : Fahmida Fathima1 , Ishaq Mustafa Naseer2 , Syed Jamal Ur Rahman3 , Syed Zahid Ali4

Title : An Intelligent Deep Learning Framework for Cyber Attack Detection in IoT Networks

Volume/Issue : 2026;03(05)

Page No : 361-369