This is a single portfolio entry
Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Aenean commodo ligula eget dolor. Aenean massa. Cum sociis natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Donec quam felis, ultricies nec, pellentesque eu, pretium quis, sem. Nulla consequat massa quis enim.
Donec pede justo, fringilla vel, aliquet nec, vulputate eget, arcu. In enim justo, rhoncus ut, imperdiet a, venenatis vitae, justo. Nullam dictum felis eu pede mollis pretium. Integer tincidunt. Cras dapibus. Vivamus elementum semper nisi. Aenean vulputate eleifend tellus. Aeneaellus.
Recent Entries
F127415070626
Investigation and prediction of defects in software is one of the important solutions to ensure software quality and reliability. Machine learning algorithms are used across a wide array of fields to solve real-world problems by building large, complex models. Many researchers have made significant contributions by developing predictive models for software defects using statistical and machine-learning approaches. But only a few frameworks have discussed the issue of building a universal software defect prediction model. Most existing models have been trained on limited datasets, which results in good performance on the training data but poor performance on unseen data. These limitations have motivated researchers to explore and develop more generalised and universal models for software defect prediction. Moreover, the growing complexity of contemporary software systems. Such limitations have encouraged researchers to investigate and build more generalised and universal models for software defect prediction. In addition, the increasing complexity of modern software systems and the rapid growth of software repositories have driven a demand for intelligent prediction techniques capable of handling heterogeneous data. Research is being conducted to investigate advanced machine learning and deep learning methods, including ensemble learning and transfer learning, to enhance prediction accuracy and adaptability across different software projects. These approaches aim to reduce development and maintenance costs and increase the overall reliability and performance of software products.
E127115060526
Industrial process control has relied on proprietary DCS for over five decades. The model worked. But working and working well are different things, and in 2026, the gap between what proprietary DCS delivers and what operators in manufacturing, oil and gas, chemical processing, and water treatment need has grown wide enough to drive real change. Vendor lock-in, hardware obsolescence that stretches across decades, and maintenance contracts that give a single supplier complete leverage over upgrade decisions are no longer tolerable when the alternative, open and software-defined control, has been proven at an industrial scale. The Open Process Automation Standard (O-PAS), the IEC 61499 function block model, and OPC UA connectivity together provide the technical foundation for software-defined control systems (SDCS) that break these dependencies. Documented lifecycle cost savings reach approximately 52% over twenty-five years when compared to equivalent proprietary DCS platforms [4]. This paper reviews the standards, examines real deployments, and confronts the barriers that still slow adoption, particularly the near-complete absence of IT-domain competencies in OT workforces. An original Software Defined Automation Risk Mapping Model (SD-ARMM) is proposed, providing practitioners with a five-dimensional, risk driven tool to determine which migration strategy fits their specific organisational and operational reality.
E126515060526
The study evaluates the return on investment (ROI) benefits of transitioning from a traditional reactive construction workflow to a proactive hybrid geotechnical resilience workflow for high-density urban infrastructure projects. This paper addressed the complexities often seen in dense-urban environments characterised by high-moisture basins, hydraulic instability, and substructure compromised by environmental unpredictability, often leading to systemic delays, soil collapse, and material wastage. These complexities are addressed by implementing lean construction principles that focus on mitigating Muda (waste), Mura (unevenness), and Muri (overburden), and on enhancing the project’s predictability and structural safety. The lean strategies to overcome the structural difficulties involved establishing a responsive feedback loop. Firstly, by achieving precise excavation using hydraulic machinery with 3D-GPS guidance kits synchronised with Digital Terrain Modelling (DTM). This helped eliminate the 10% standard manual over-dig typically encountered in traditional depth control. Thereby optimising excavation volumes and reducing redundant soil hauling. Secondly, a real-time monitoring network comprising vibrating-wire piezometers and inclinometers was used to monitor pore-water pressure and soil displacement. This sensor-driven approach enabled a Jidoka (built-in quality) protocol, in which automated alerts for pressure spikes triggered immediate stabilisation measures that helped prevent catastrophic failures that historically stall urban developments. In the study, a comparative performance analysis of a traditional workflow and a lean-integrated workflow demonstrates that the proactive lean-integrated workflow results in a quantifiable reduction in the construction timeline and labour volatility. Specifically, the excavation and the shoring durations were reduced by up to 40% through data-driven execution and Target Value Design (TVD). The findings validate that incorporating digital intelligence during the substructure phases helps achieve a net fiscal recovery of over ₹2.17 crores by preventing rework and resource wastage. By providing a scalable model for geotechnical resilience, this study helps optimise operations and improve ROI for projects in complex urban settings.
E125515060526
Eye diseases are becoming common in day-to-day life and affecting all age groups of people. The ratio of ophthalmologists to patients suggests the need for an automated technique to detect eye diseases. Conjunctivitis is of many types, including adenoviral conjunctivitis, ocular drug toxic conjunctivitis, pollen-allergic conjunctivitis, bacterial conjunctivitis, and many others. Conjunctivitis can be automatically detected using conventional image processing techniques, but with lower accuracy and precision, and more computational time is required compared to deep learning and AI techniques. This paper presents a novel deep-learning-assisted segmentation technique for the automatic detection of conjunctivitis that overcomes the limitations of conventional methods. The proposed method uses Attention-U-Net++ with Transformer Encoder (Global Context), Swin / ViT CNN +, Transformer Monte-Carlo Dropout Layer Enabled at inference time, Uncertainty-aware and Segmentation Mask + Uncertainty Map, which provides better results with Accuracy= 90.3%, sensitivity= 0.87, specificity= 0.93, precision=0.88, recall=0.87, F1-Score= 0.89, ROC-Auc=0.96.
