Authors: Gunjan and Rishita Vohra, Ms. Preeti Kumari

Abstract: The AI-Based Food Expiry Tracker and Smart Recipe Suggestion System is a web-based platform developed tointelligentlymonitorfooditems,predictexpirydates,andsuggestsuitablerecipesbasedonavailableingredients.Thesystemoperates through three major components — user, business, and admin — each offering specializedfunctionalities for foodtracking,recipegeneration,andsystemmanagement.Fooditemscanbeaddedmanuallyorthroughsmartlogging,andusersreceive timely notifications before items expire. The integrated AI module enhances usability by analyzing ingredientcombinations and recommending region-specific Indianrecipes to prevent wastage. For efficient and reliable performance, thesystemarchitectureincorporatestechnologiessuchasPHP, Laravel, HTML, CSS, JavaScript, MySQL, andPythonforAIandmachinelearningintegration.Thisprojectpresentsascalableandsustainablemodelthataddressesthegrowingissueoffoodwastage by promoting intelligent kitchen management and mindful consumption.

Published by: vikaspatanker

Authors: Madhusudan Kumar, Abhinav Kumar Singh, Prof. Vishal Mehtre

Abstract: Optimal Power Flow (OPF) is a crucial problem in power systems that involves generating power at minimum cost while ensuring safe and feasible operation. This problem is normally solved by using mathematical approaches but they are not always effective due to the problem's complexity. In this context, researchers have started to apply smart, nature-inspired algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO). These techniques help find better solutions, even for complex problems. In this paper, we analyse these algorithms in terms of speed, accuracy, computational effort and robustness. After analysing results of various research papers, we find what algorithm works best under various power system conditions.

DOI: https://doi.org/10.5281/zenodo.20198493

Published by: vikaspatanker

Authors: Sagar C Saner, Shivam Dubey, Mrs.Supriya kamareddy

Abstract: The rapid growth of digital music streaming platforms has transformed the way individuals access, consume, and discover music. This study examines the impact of digital music streaming platforms on listeners’ music consumption behavior in comparison with traditional music formats such as CDs and cassettes. The research specifically investigates changes in listening behavior, the influence of personalized recommendation systems, the role of streaming platforms in music discovery, and listeners’ perceptions toward AI-generated music. A descriptive cross-sectional research design was adopted, and primary data were collected through a structured questionnaire from 102 respondents using a convenience sampling technique. Data analysis was conducted using descriptive statistics and Chi-square tests with the help of SPSS. The findings indicate that digital music streaming platforms have significantly influenced music consumption behavior by providing greater convenience, wider music accessibility, and increased listening time. Personalized recommendation systems and algorithm-based playlists were found to strongly influence listeners’ music preferences and choices. Streaming platforms also emerged as powerful tools for discovering new songs, artists, and music genres. However, respondents showed only moderate acceptance toward AI-generated music, suggesting cautious openness toward this emerging technology. Statistical analysis further revealed that demographic variables such as education, occupation, and gender showed selective influence on certain aspects of music behavior, whereas age had no significant impact on perceptions of AI-generated music. Overall, the study concludes that digital music streaming platforms have become a dominant force shaping modern music consumption patterns and listener experiences.

Published by: vikaspatanker

Authors: Tejaswini Taware

Abstract: This seminar work presents a reinforcement learn-ing based field-oriented control strategy for Permanent Magnet Synchronous Motor (PMSM) drives. A Twin Delayed Deep Deterministic Policy Gradient (TD3) agent is used to replace the conventional PI current controller in the dq-axis current loop. The controller is validated using a 10 s staircase per-unit speed profile with repeated acceleration and braking transitions. The obtained results show fast tracking, low overshoot, stable dq current regulation, and improved robustness for practical intelligent drive applications.

DOI: https://doi.org/10.5281/zenodo.20180838

Published by: vikaspatanker

Authors: Akashdeep Singh, Ansh Gupta, Anmol Goyal

Abstract: This research paper presents a comprehensive analytical study on strategic campaign restructuring and multi-level segmentation within a revenue intelligence ecosystem. The research was conducted during an industry-integrated business analytics engagement at Reviniti, a revenue intelligence platform developed by 1DigitalStack. The study investigates KPI-driven dashboard optimization, marketing attribution analysis, campaign ROI evaluation, cohort analysis methodologies, data validation procedures, and automated reporting frameworks. The implementation integrated Microsoft Excel, Google Sheets, Metabase, and the Reviniti platform to support analytical processing, visualization, and stakeholder reporting. The project identified major gaps in last-click attribution models and introduced structured multi-touch attribution methodologies for improved revenue allocation. Significant business outcomes included an 18% reduction in cost per acquisition, a 40% increase in dashboard adoption among non-technical stakeholders, a 31% improvement in lead-to-close ratio, and an over 80% reduction in reporting cycle duration. The paper demonstrates the practical significance of structured business intelligence systems, dashboard-centric architectures, and KPI-driven decision-making frameworks in optimizing marketing performance and operational efficiency.

DOI: https://doi.org/10.5281/zenodo.20179426

Published by: vikaspatanker

Authors: Dr. Jermiah Anand Jupalli, Dr. Kiran Koduru

Abstract: The rapid evolution of artificial intelligence (AI) and digital transformation has significantly influenced the domain of human resource management (HRM), enabling the development of intelligent and data-driven systems. This paper proposes an Intelligent Human Resource Management System (IHRMS) framework designed to enhance organizational efficiency and decision-making through AI-driven analytics, automation, and predictive modeling. The study integrates multiple HR functions, including recruitment, performance evaluation, employee engagement, and attrition prediction, into a unified intelligent system. A synthetic dataset is utilized to evaluate the performance of the proposed model, and comparative analysis is conducted with traditional machine learning approaches such as Support Vector Machine and Decision Tree. The results demonstrate that the proposed IHRMS model achieves higher accuracy, improved prediction consistency, and better decision support capabilities. Furthermore, the study addresses ethical considerations such as fairness, transparency, and data privacy in AI-based HR systems. The findings indicate that intelligent HR systems can significantly contribute to organizational excellence by improving workforce management, enhancing employee experience, and enabling strategic decision-making.

DOI: http://doi.org/10.5281/zenodo.20179023

Published by: vikaspatanker

Authors: Ms.Pharande Harshada Sudhir, Dr.Dhaigude.N.B

Abstract: In working environment, the toxic gas leakage accidents are the main reason for workers health and also causes death. The Toxic gas can be detected and monitored by recent technologies using Internet of things. This project is mainly used to reduce the industrial accidents and hazardous. This process is monitored by Internet of things. Arduino Micro controller board is connected with gas sensor, Flame sensor and Temperature senor. The alert message is display by LCD through Arduino. The alert signal arises when the gas level increases above the normal gas level. This can be done by internet receiver channel. The sensor will receive the information about the gas level and it is stored in internet. This will used for analyzing and processing the safety regulations in industrial environment.

DOI: https://doi.org/10.5281/zenodo.20177739

Published by: vikaspatanker

Authors: Vivek Mishra, Om Prakash Sondhiya

Abstract: Rice husk is a common lignocellulosic agricultural by-product produced in huge amounts across the world, with nearly 150 million tons generated every year. This work examines the preparation and assessment of bioethanol obtained from rice husk waste as an eco-friendly second-generation biofuel. The rice husk was collected, dried, powdered, and treated with 4% NaOH at 90°C for 2 h, followed by steam explosion at 121°C for 30 min to remove lignin and hemicellulose components. Enzymatic saccharification was carried out using cellulase (30 FPU/g) and xylanase (10 FPU/g) at pH 5.0 and 50°C for 72 h, producing 68.4 g/L reducing sugars. Fermentation was performed with Saccharomyces cerevisiae (MTCC 178) at 32°C for 96 h and resulted in 32.6 g/L bioethanol with 95.2% fermentation efficiency. The produced bioethanol was purified through double distillation and molecular sieve dehydration to reach 99.5% purity, and the product was analysed using GC-MS, FTIR, and NMR techniques. The physicochemical parameters, including density (789 kg/m³), calorific value (26.8 MJ/kg), and octane number (108), matched ASTM D4806 requirements. Engine testing on a 4-stroke, single-cylinder SI engine (5.2 kW, 1500 rpm) with E10, E20, E50, and E85 blends revealed that E20 decreased CO emissions by 38% and HC emissions by 32% relative to gasoline, with only a 3.5% decline in brake thermal efficiency. CFD analysis using ANSYS Fluent confirmed the experimental findings with an error lower than 6%. The results demonstrate that rice husk can serve as an effective feedstock for large-scale bioethanol manufacturing while supporting waste utilisation and renewable energy production.

Published by: vikaspatanker

Authors: Shubhangi Baghel, Om Prakash Sondhiya

Abstract: Bioethanol has emerged as one of the most promising renewable energy sources for reducing greenhouse gas emissions and decreasing dependence on fossil fuels. However, conventional bioethanol production systems face significant challenges, including low conversion efficiency, process instability, high operational costs, and limitations in feedstock utilization. Recent developments in artificial intelligence (AI) and machine learning (ML) have introduced advanced computational approaches capable of transforming industrial bioethanol production through predictive analytics, process automation, and intelligent optimization. This paper examines the role of AI and ML technologies in enhancing fermentation efficiency, optimizing biomass pretreatment, predicting ethanol yield, and improving overall sustainability in bioethanol production systems. The study also discusses key machine learning algorithms, including artificial neural networks, support vector machines, random forests, and deep learning frameworks, alongside their industrial applications. Furthermore, the paper evaluates challenges associated with data quality, computational complexity, scalability, and ethical considerations. The findings indicate that AI-driven systems significantly improve process accuracy, reduce waste generation, and enhance economic feasibility. Future research directions involving digital twins, autonomous biorefineries, and explainable AI are also explored.

Published by: vikaspatanker

Authors: Sahil Arun Bodke, Devika Deepak More, Samruddhi Mahendra Pansare, Prof. P. A. Mande, Prof. Bangar A.P., Prof. Bhosale S.B.

Abstract: Industrial workplaces continue to pose significant hazards to workers, including toxic gas exposure, thermal stress, mechanical injuries, and fatigue-related accidents. Conventional safety systems have largely remained reactive, responding to incidents after they occur rather than preventing them proactively. The convergence of Artificial Intelligence (AI), the Internet of Things (IoT), and advanced wearable sensor technologies has opened transformative opportunities for proactive occupational safety. This paper presents a comprehensive literature review of existing research on AI-integrated industrial safety wearable devices, covering sensor technologies, machine learning algorithms, edge computing strategies, cloud-based analytics, and alert mechanisms. We synthesize findings from over 25 peer-reviewed studies published in IEEE, Springer, and Web of Science indexed journals between 2019 and 2025. Key research gaps identified include the lack of multi-modal sensor fusion with real-time edge AI, insufficient datasets for industrial fatigue prediction, limited ergonomic wearable designs for harsh environments, and the absence of Explainable AI (XAI) in safety-critical decision making. Based on the review, we propose an integrated four-layer system architecture combining physiological and environmental sensing, edge-level AI inference, MQTT-based cloud communication, and a multi-level alert mechanism.

DOI: https://doi.org/10.5281/zenodo.20176697

Published by: vikaspatanker