This bibliometric study analyses 435 English-language articles from Scopus and Web of Science to map the evolution of Fintech in risk management. Research output surged significantly, with the USA (13 articles) and China (15 articles) emerging as dominant contributors. Key themes include Financial Technology & Credit Scoring, Artificial Intelligence in Peer-to-Peer Lending, and Machine Learning & Credit Risk. Leong, Carmen (TC=171) and Bartlett, Robert (TC=152) are the most influential authors, while the Journal of Financial Economics (TC=181, ABDC A*) and International Journal of Information Management (TC=171, ABDC A*) lead in scholarly impact. Local citation analysis highlights foundational works like Gomber et al. (2018, LC=27) and Bartlett (2022, LC=7). The findings reveal a shift toward AI/ML-driven solutions for credit risk, fraud detection, and operational risk, underscoring Fintech’s transformative role in enhancing financial stability and inclusion. Emerging gaps in liquidity and market risk applications warrant further exploration.

Artificial Intelligence (AI) has grown from just an idea into a powerful technology that is changing almost every part of our lives. This paper reviews the journey of AI—from its early beginnings to where it stands today and where it might go in the future. First, we look at the history of AI, including its important achievements and the difficult times known as “AI winters,” when progress slowed down. We then explain the main building blocks of modern AI, such as Machine Learning (ML), Deep Learning (DL), Natural Language Processing (NLP), Computer Vision (CV), and Reinforcement Learning (RL), describing how they work and why they matter. The paper also shows how these technologies are used in real life, especially in areas like healthcare, transport, finance, and education. After that, we discuss new research trends such as Explainable AI (XAI), Federated Learning, Human–AI teamwork, and the connection of AI with quantum computing and biotechnology. Finally, we look at the future of AI, pointing out both its benefits and challenges. The main conclusion is that AI must be developed responsibly, with proper ethics and human values, so that it can truly serve humanity.

The smooth movement of people and goods is essential for modern society, but it remains a difficult and constantly changing task. This challenge, known as the "transportation problem," has expanded from basic logistics to large, complex urban and global networks. This paper reviews the computational and algorithmic methods developed to address it. First, it introduces the key mathematical models in transportation planning, such as the classic Transportation Problem and the Vehicle Routing Problem (VRP). It then highlights how solution approaches have evolved—from exact methods for small-scale cases, to metaheuristics, and now to machine learning techniques for handling large and complex real-world systems. The paper also discusses practical applications, including real-time ride-sharing, sustainable city logistics for e-commerce, and management of autonomous vehicle fleets. Finally, it outlines current challenges such as the demand for robust and real-time solutions, and emphasizes that combining traditional operations research with modern artificial intelligence will be crucial for building efficient and resilient transportation systems in the future.

The deregulation of the electric power industry has shifted market transactions from cost-based operations to price-driven mechanisms, creating a pressing need for efficient tools to ensure grid stability and economic optimization. This study evaluates the role of Unscheduled Interchange (UI) charges and the Deviation Settlement Mechanism (DSM) in managing frequency deviations and reactive power (Q) adjustments under CERC regulations based on Availability-Based Tariff (ABT). Using real-time operational data and adaptive pricing models, the research compares static and adaptive Q management strategies. MATLAB simulations demonstrate that adaptive approaches outperform static methods, with adaptive Q reaching 5080.7 compared to 4542.9 in static models, reflecting reduced variability and improved stability. Dynamic pricing strategies, incorporating real-time impacts of 0.05, ensured the grid could respond effectively to fluctuations. Penalty multipliers of 1.5 during peak hours versus 1.0 in off-peak periods further encouraged grid discipline and economic efficiency. The findings underscore the superiority of adaptive regulatory mechanisms, showing that real-time integration of pricing and operational strategies not only minimizes economic losses but also enhances reliability in competitive power markets.

This research article explores the conversion of renewable energy into electricity through an integrated system composed of multiple components that ensure efficient power delivery to the end user. Key elements, such as a DC–DC converter and a harmonic analyzer, are employed to minimize power loss and enhance voltage stability. To further optimize system performance, the Zebra Optimization Technique is applied for parameter tuning and control. A Zero-Burden Deep Belief Neural Model (ZbDBNM) is introduced to manage critical aspects of the microgrid, including power generation capacity based on source availability, energy storage regulation, and adaptive handling of power and voltage according to load requirements. Additionally, the operational limits of converter–inverter units and varying environmental conditions are considered to maintain reliable and efficient functioning. The proposed approach demonstrates the potential for improved efficiency, stability, and adaptability in renewable energy-based microgrid systems.

The microstrip patch antenna (MPA) has evolved significantly since its introduction, moving from a simple radiating element with many limitations to a versatile component essential for modern wireless systems. This paper provides an in-depth review of the evolution, design improvements, and research progress in MPA technology. It traces the development from the early 1970s to the present, highlighting the advantages that led to its wide adoption. The paper also discusses the challenges of early designs, such as narrow bandwidth and low gain, and the solutions introduced to overcome them, including modified geometries, slotting methods, multi-layer structures, and the use of metamaterials. Applications of MPAs in mobile communication, satellite links, radar, and biomedical devices are also reviewed, with emphasis on their importance in 5G. Finally, future directions such as reconfigurable intelligent surfaces (RIS), AI-assisted design, and terahertz applications are outlined, showing the continuing importance of this technology.

The global energy sector stands at a critical turning point, balancing the need to meet growing demand while reducing carbon emissions. The shift from a centralized, fossil-fuel system to a decentralized, digital, and sustainable model is being driven by disruptive technologies. This paper reviews these developments under three main pillars: (1) Decarbonization Technologies such as advanced renewables and green hydrogen, (2) Digitalization Technologies including Artificial Intelligence (AI), Internet of Things (IoT), and blockchain for smarter grid operations, and (3) Integration Technologies like advanced energy storage and Vehicle-to-Grid (V2G) systems. We examine current progress, synergies, and implementation barriers within each pillar. The review highlights how combining these pillars can create resilient, efficient, and intelligent energy networks. Finally, we outline key research needs and policy considerations, emphasizing that a system-wide perspective is essential to achieve a clean, affordable, and secure energy future.

This paper presents a comprehensive solution for the real-time identification of electronic devices and instruments using Tensor Flow, a widely adopted machine learning framework. A web-based application, developed with JavaScript and powered by TensorFlow.js, has been designed to recognize various laboratory devices in real time. This tool addresses a common issue faced by individuals who may forget the names or functions of electronic equipment after extended periods of disuse. By leveraging artificial intelligence and machine learning techniques, the application assists users in accurately identifying instruments such as multi-meters, de-soldering pumps, and batteries. The system performs all inference directly within the Chrome browser, ensuring fast and efficient processing without the need for server-side computation. Experimental validation demonstrated 100% accuracy in identifying the selected devices, highlighting the effectiveness and practicality of the proposed approach.

Machine Learning (ML) has emerged as a transformative force, driving digital innovation and reshaping industries and societies worldwide. This paper presents a comprehensive exploration of ML, beginning with its historical foundations and theoretical underpinnings, followed by an overview of its key algorithms and architectures. Various types of Artificial Neural Networks, including classification and clustering models implemented in Python, are discussed to provide practical understanding. The study illustrates ML’s versatility through applications in healthcare diagnostics, financial risk prediction, and automation calibration using IoT sensor data. These real-world examples highlight how ML integrates theory with practice to address complex problems. Beyond current applications, the paper also examines emerging trends such as distributed learning, transparency tools, and on-device personalized training, which are expected to define the next generation of ML systems. By synthesizing theoretical concepts with experimental insights, this work not only provides a holistic view of ML’s current impact but also anticipates its potential role in shaping future technologies, industries, and societal transformations. Ultimately, the study aims to encourage research and innovation in advancing ML-driven solutions for sustainable progress.

The use of Piezoelectric energy harvesters in low power requirement self-power device applications is increasing day by day. So, it’s very essential and important to improve the harvester energy conversion performance and increase the output power of the harvester so that the efficiency of the harvester also improved and increased. The way for improve the harvester generation voltage depends on many factors, among them one way of approach is the uniform use of materials and minimizing that more strain stress on the piezoelectric layer. Both approaches lead to improving converter efficiency and the requirements of the materials can be reduced by optimum use of the piezoelectric materials. The conventional cantilever arrangement with tip mass does not provide uniform stress strain to the harvester but it’s high near to tip mass and less to the rest of the harvester beam. This paper focused on the comparisons between the cantilever piezoelectric energy generations with the clamped type of configuration. The comparisons between cantilever and clamped configuration proofs that the uniform stress strain achievement is more in clamed configuration and the performance of the harvester improved, and requirement of the piezoelectric materials reduced. 

This work demonstrates the feasibility of friction stir welding (FSW) as a novel technique for joining aluminum sheets while analyzing thermal profiles and microstructural evolution. Defect-free welds were achieved on 1.6 mm thick plates at a welding speed of 160 mm/min and 450 rpm using die steel tools. Optical microscopy revealed distinct microstructural regions in the weld zone, while transverse microhardness measurements and tensile tests confirmed the welds’ strength, with failure occurring outside the weld area. The study highlights the potential of FSW for aluminum welding, particularly in automotive applications where aluminum alloys of the 5XXX/6XXX series are increasingly used for tailor welded blanks (TWBs) due to their superior formability. Despite extensive research on process parameters and weld quality, experimental studies on thin-sheet AA 5XXX/6XXX TWBs remain limited, underlining the significance of this work.

The global Gas Tungsten Arc Welding (GTAW) machine market, valued at approximately $870 million in 2025, is projected to experience robust growth, driven by a Compound Annual Growth Rate (CAGR) of 5.7% from 2025 to 2033. This expansion is fueled by several key factors. The increasing demand for high-precision welding in industries like aerospace, where lightweight and high-strength materials are crucial, is a significant driver. Advancements in technology, leading to more efficient and automated GTAW machines, are also positively impacting market expansion. America and Europe likely maintaining strong positions due to established industries and technological advancements. However, the Asia-Pacific region is expected to experience significant growth, propelled by rapid industrialization and increasing manufacturing activities in countries like China and India.

A few benefits of this welding method include improved tensile strength, finer grain hardness, excellent weld quality and appearance, micro structure improvement, and metallurgical properties over the source metal. To shed light on those sectors that have not yet realised their full potential, the primary objective of this research is to assess current advances in the TIG welding business. Thus, based on tests, parameters, and the kinds of materials used for welding, numerous TIG welding tales are compiled and investigated in this study article. The bulk of the mentioned studies state that the main parameter among them rate and angle of weld utilising modern techniques have not been used up to that point.

Tungsten Inert Gas (TIG) welding, also known as gas tungsten arc welding (GTAW), is a type of arc welding process that utilizes an inert gas shield and a non-consumable electrode. The electrodes may include a mixture of 1 to 2% thoria (thorium oxide) along with the tungsten core, or they may consist of tungsten alloyed with 0.15 to 0.40% zirconia (zirconium oxide). This method is suitable for welding both ferrous and non-ferrous metals. Advantages of this process include a reduced heat-affected zone, the ability to join dissimilar metals, the lack of slag, and a high concentration of heat, particularly when compared to other welding methods. The parameters for TIG welding are crucial factors that influence the quality, productivity, and cost of the welding proce.

Green chemistry is a science-driven approach focused on designing chemical products and processes that reduce or eliminate the use and generation of hazardous substances. Its principles promote waste minimization, safer chemical synthesis, energy efficiency, and the use of renewable materials, contributing to sustainable development across industries. This concept has significant relevance in agriculture, where it supports the development of eco-friendly fertilizers, biodegradable pesticides, and organic soil amendments. These innovations improve nutrient use efficiency, reduce environmental pollution, and enhance soil health. Studies have shown that controlled-release fertilizers made from biopolymers, and amendments like compost, vermicomposting, and bio char, can effectively suppress soil borne pathogens and improve crop productivity without compromising ecosystem integrity. Green chemistry plays a critical role in meeting global challenges such as food security, climate change, and sustainable agriculture. By shifting focus from remediation to prevention, it provides a pathway toward cleaner technologies, healthier ecosystems, and a more resilient agricultural system.

A multilevel inverter (MLI) is a power electronic device that is used for high-power and high-voltage applications and has many advantages like reduced total harmonic distortion (THD) and high-power quality waveform. Hence, the size and bulkiness of passive filters can be reduced. The output voltage of the multi-level inverter is in a staircase waveform. Therefore, the voltage stress of the switching device is lower, and the total harmonic distortion (THD) is also lower than that of the other types of inverters. As the number of DC voltage sources increases in input side, the sinusoidal like waveform can be generated at the output of inverter. In this paper, a CHB 7-level inverter is discussed using two modulation techniques. In the present work, the Sinusoidal Pulse Width Modulation (SPWM) and nearest level control (NLC) method are used. The SPWM scheme provides the best harmonic profile in high frequency switching. The nearest level control method is used in low switching frequency, which leads to reduction of switching losses. The model is simulated using SPWM and nearest level modulation techniques and the effect of the harmonic spectrum is analyzed.