22505 research outputs found
Sort by
Human-Aware Reinforcement Learning for Adaptive Human-Robot Teaming
Get PDFThe rise of Industry 5.0 has shifted robotic systems toward human-centric design, emphasizing collaboration between humans and machines. While collaborative robots (cobots) are equipped with sensors for safety, they typically overlook human internal states such as fatigue, stress, and workload—factors that significantly impact performance. This lack of awareness limits their ability to adapt and collaborate effectively, particularly in dynamic, real-time environments where human conditions fluctuate. This thesis introduces a novel framework that integrates human internal states into robot decision-making to enhance human-robot collaboration. This research leverages reinforcement learning (RL) to develop methods that enable robots to adapt their behavior based on human physiological signals, enhancing team performance and interaction fluency. A key contribution is the exploration of offline RL techniques to address challenges in real-time human data collection, facilitating the training of human-aware RL agents in a safe and resource-efficient manner. Additionally, this work introduces Modality Utilization (MU) and State Utilization (SU) metrics—tools designed to quantify an RL agent’s reliance on different input features, including human data. These metrics enhance explainability and provide mechanisms to detect and mitigate over-reliance on any single information modality. By advancing human-aware robotic decision-making, this research contributes to the development of adaptive, collaborative systems aligned with Industry 5.0 principles
Enhancing Third-Party Risk Management in Aviation: A Comprehensive Framework for Modern Challenges
Get PDFThis thesis is focused on improving the way airlines have managed risks from third-party vendors. Emirates and Etihad have been used as case studies. The thesis identifies major issues, including cybersecurity threats, slow manual processes, and the challenges in the following global regulations. To To solve such problems, the research proposes a new framework that integrates advanced tools including machine learning, blockchain, and real-time updates about threats. The framework includes different features like automatic risk scoring and detection of unusual vendor behavior and keeping the secure digital records through the blockchain. This will highlight the need for the staff to train across departments for creating a strong culture aware of risks. In addition, it includes the plans for handling the regulatory disruption and ensures a smooth operation during the issues related to the vendors. The research has used interviews, observations, and document reviews for understanding the current systems and the gaps present. The proposed framework gives a theoretical analysis that suggested that the proposed framework for the management of third-party risks, ensures safety and compliance, and enhances efficiency. This study gives practical solutions that aid airlines and other industries in dealing with third-party risks in a more effective manner
The Use of Robotic Surveillance in the UAE in 2035: Balancing, Innovation, Regulation, and Privacy
Get PDFThis research explores the future of robotic surveillance in the United Arab Emirates (UAE) by 2035, aiming to balance innovation, regulatory oversight, and privacy protection. It investigates the intersection of national security objectives, ethical AI governance, and public trust within the UAE’s rapidly advancing smart city and law enforcement environments. The focus areas include regulatory gaps in current AI laws, ethical concerns such as data privacy and algorithmic bias, and the societal perception of AI-powered surveillance systems. The background of this study stems from the UAE’s ambition to lead in AI deployment, especially within public security initiatives such as autonomous patrol robots and biometric surveillance systems. However, the lack of clear, AI-specific governance policies in the UAE has created significant legal and ethical challenges, especially when compared with mature frameworks like the EU AI Act or Singapore’s AI Ethics Guidelines. The central research question guiding this study is: How can the UAE develop an AI governance framework for robotic surveillance that ensures national security while upholding privacy rights and ethical standards? Supporting questions explore regulatory deficiencies, ethical risks, public perception, and the formation of practical policy recommendations. Methodologically, the research employs a scenario planning framework within a future foresight approach. No primary data collection was conducted; instead, the study uses only secondary data such as policy reports, academic publications, foresight documents, and case studies from global examples. The analysis involved thematic coding of secondary sources and mapping these themes to four future scenarios of robotic surveillance in the UAE. These scenarios considered legal frameworks, ethical principles, public sentiment, and technology adoption trends to outline plausible policy pathways. Key findings reveal that the UAE’s existing Data Protection Law No. 5 of 2020 provides general privacy guidance but lacks provisions tailored to AI-powered surveillance systems. Ethical risks, especially around mass data collection and algorithmic discrimination, remain largely unregulated. Furthermore, public perception studies in similar regions indicate that trust and transparency are critical to citizen acceptance, yet the UAE lacks localized research in this area. These findings suggest the urgent need for UAE-specific AI governance frameworks that are both adaptive and context-sensitive. The study concludes that a strategic, foresight AI governance model is necessary to prevent misuse, ensure transparency, and build public confidence in robotic surveillance technologies. It also contributes to the broader academic discourse on AI ethics, surveillance governance, and foresight-driven policy design in emerging economies. Practical recommendations include the development of legal frameworks that mirror global best practices while respecting the UAE’s unique socio-political context. This includes the establishment of an independent AI oversight authority, guidelines for ethical AI deployment, transparency audits for surveillance algorithms, and public awareness campaigns to boost trust in AI systems. For future research, scholars should explore experimental studies on UAE citizen attitudes toward surveillance, the effectiveness of AI auditing mechanisms, and comparative impacts of surveillance technologies in different smart cities. Policymakers are encouraged to adopt a flexible, scenario-based approach to technology regulation, allowing the UAE to adapt to rapid innovation while preserving individual freedoms
Determinants of Profitability in UAE Banks (2019-2023): An Analysis and Forecast for the Next Five Years
Get PDFThis study aims to analyze the factors that led to the banks\u27 profitability in UAE from 2019 to 2023 based on factors such as capital adequacy, asset quality, and liquidity. With these factors and Return on Assets, the study seeks to determine the link between the variables affecting financial performance in the banking sector. The study relies on quantitative data from UAE banks to include statistical analysis and construct predictive models. The relationships between the variables are established using Multiple Linear Regression. At the same time, Random Forest Algorithms explore Non-Linear Interactions and Time Series Analysis using ARIMA models to forecast profitability trends. This analytical framework helps to understand the current dynamics and project scenarios for the next five years. It will accustom financial institutions and policymakers in the UAE with information relating to factors that express banks\u27 profitability. The trend analysis will add a dynamic viewpoint on the ongoing changes in the banking sector, possible threats, and opportunities in the future. The study\u27s findings offer valuable insights into the determinants of banking profitability in emerging markets. Thus, the proposal focuses on the UAE as an important financial center, allowing for observing local processes and identifying general tendencies applicable to other areas. The results will help UAE banks improve their financial performance efficiency and offer helpful information for global financial institutions and researchers intending to study the related markets. Thus, this study closes the gap between academic research and practical application by presenting recommendations for the banking sector to increase its resilience and profitability in the face of constant economic disturbances and regulatory modifications
Real-Time Fraud Detection using Big Data
Get PDFIn today’s digital world, fraud detection has become an important part of financial security. This study explores and compares the performance of different machine learning models in identifying fraudulent transactions using the IEEE-CIS Fraud Detection dataset. Techniques such as Random Forest, Gradient Boosting, Deep Neural Networks, and Logistic Regression were evaluated. The dataset was pre-processed using SMOTE to balance the classes and improve model sensitivity to fraud cases. Each performance of the model was assessed using accuracy, precision, recall, and F1-score. The Random Forest model achieved the highest overall performance with an F1-score of 99.2
Development of Nanopocket Membranes for the Isolation and Purification of Extracellular Vesicles
Get PDFExtracellular vesicles (EVs) play a crucial role in intercellular communication and serve as significant biomarkers for various diseases. EVs carry biomolecules such as lipids, proteins, DNA, and RNA, which reflect the physiological state of their cells of origin, making them promising tools for non-invasive diagnostics and therapeutic applications. Effective isolation of EVs is essential for advancing scientific understanding of their biological roles and unlocking their clinical potential. Porous membranes have been widely used for the purification of various biological species from biological fluids. While membranes have prove highly effective for general size-based separation, recent innovations have focused on structurally refined designs. One such innovation is the nanopocket membrane, in which the pores gradually narrow from one side to the other. This engineered geometry enhances precision in particle capture and controlled release because particles within a specific size range enter and become trapped within the nanopocket pores until they are released by reversing the pressure. The first aim of this work is to fabricate nanopocket membranes. These nanopocket structures are specifically designed to control pore size, wall tilt, and uniformity. The fabrication process is optimized to achieve well-defined pore characteristics, ensuring consistent and effective isolation of EVs. The second aim is to integrate the nanopocket membranes into a Tangential Flow Filtration (TFF) microfluidic device called the tangential flow analyte capture device. This integration aims to enable efficient capture and controlled release of beads, liposomes, and EVs based on size, utilizing TFF to reduce pore clogging and enhance the filtration process by using surface modification. The combination of the TFF system and nanopocket membranes presents a scalable and efficient method for isolating EVs
Future Unleashed: Reimagining Marketing Strategies with Future Foresight
Get PDFIn today’s fast-paced and dynamic business landscape, marketing plays a crucial role in influencing business strategy and decision-making in an organisation. Marketing not only helps organisations identify and anticipate trends, threats, and opportunities, but also enables them to formulate and implement effective strategies to capitalise on these insights and achieve the organisation’s vision and goals. Future Foresight in marketing empowers organisations to stay ahead of the curve and gain a deeper understanding of the target audience and industry landscape. The evolving nature of marketing driven by advancements in technology, particularly artificial intelligence, machine learning, and digital marketing channels, makes it imperative for organisations to integrate Future Foresight into their strategic planning. This practice allows the organisation to better prepare for the future, increasing its chances of long-term success and of achieving greater sustainability in a competitive marketplace. However, the adoption and implementation of Future Foresight practices within organisations is still limited. There is a lack of understanding and awareness of the value of strategic Future Foresight activities, which leads to the hesitancy in implementing these practices across any organisation. In this paper, the research will focus on exploring the use of Future Foresight tools and techniques in marketing to enhance business strategy and decision-making. The case study used in this research paper is Expo City Dubai. The aim is to assess and evaluate the outcomes and value added from employing Future Foresight tools and techniques in the marketing department of Expo City Dubai, and its impact on the overall success and competitiveness of Expo City Dubai in the marketplace
Determining SPHINCS+ Readiness for Standardization of SLH-DSA Signature
Get PDFAs quantum computing advances, public-key cryptographic algorithms risk becoming obsolete, requiring the development and implementation of quantum-resistant alternatives. This thesis evaluates SPHINCS+, a stateless hash-based digital signature scheme recently standardized by NIST under the FIPS 205 standard named Stateless Hash-Based Digital Signature Algorithm (SLH-DSA), which was selected for being a conservative and robust choice due to its reliance solely on well-understood cryptographic primitives. In the context of the growing need for quantum-resistant cryptographic solutions, determining the readiness of SPHINCS+ involves assessing its practical viability across different application domains and evaluating how well it meets today’s and future needs. To achieve this, we analyze SPHINCS+ in the following areas. Our analysis begins by benchmarking SPHINCS+ against classical and post-quantum signature schemes such as the Elliptic Curve Digital Signature Algorithm (ECDSA) and ML-DSA, measuring key generation, signing, and verification times, as well as signature sizes across multiple variants. The scheme’s performance is evaluated in constrained environments such as Vehicle-to-Vehicle (V2V) communication, simulating its integration into mock blockchain transactions and TLS-like protocols to assess feasibility and scalability. In addition, this work investigates hybrid cryptographic models that combine SPHINCS+ with classical schemes to ensure backward compatibility and cryptographic agility, as a transitional solution during this post-quantum migration period. The findings suggest that, while SPHINCS+ offers strong security against quantum computers and has been standardized by NIST, its substantial performance drawbacks raise critical concerns about its practicality. These limitations indicate that SPHINCS+, despite its conservative design, may not be suitable for many real-world applications. As such, we believe it is imperative that an alternative signature scheme be developed and standardized alongside the current selection to ensure feasibility across all types of applications