https://repositori.mypolycc.edu.my/jspui/handle/123456789/6661 2026-07-17T02:31:07Z 2026-07-17T02:31:07Z Orsini, Massimiliano Giordano Ferone, Alessio Inno, Laura Maratea, Antonio Casolaro, Angelo https://repositori.mypolycc.edu.my/jspui/handle/123456789/10055 2026-06-23T03:00:12Z 2025-09-27T00:00:00Z

Title: FLOW MATCHING FOR SIMULATION-BASED INFERENCE: DESIGN CHOICES AND IMPLICATIONS Authors: Orsini, Massimiliano Giordano; Ferone, Alessio; Inno, Laura; Maratea, Antonio; Casolaro, Angelo Abstract: Inverse problems are ubiquitous across many scientific fields, and involve the determination of the causes or parameters of a system from observations of its effects or outputs. These problems have been deeply studied through the use of simulated data, thereby under the lens of simulation-based inference. Recently, the natural combination of Continuous Normalizing Flows (CNFs) and Flow Matching Posterior Estimation (FMPE) has emerged as a novel, powerful, and scalable posterior estimator, capable of inferring the distribution of free parameters in a significantly reduced computational time compared to conventional techniques. While CNFs provide substantial flexibility in designing machine learning solutions, modeling decisions during their implementation can strongly influence predictive performance. To the best of our knowledge, no prior work has systematically analyzed how such modeling choices affect the robustness of posterior estimates in this framework. The aim of this work is to address this research gap by investigating the sensitivity of CNFs trained with FMPE under different modeling decisions, including data preprocessing, noise conditioning, and noisy observations. As a case study, we consider atmospheric retrieval of exoplanets and perform an extensive experimental campaign on the Ariel Data Challenge 2023 dataset. Through a comprehensive posterior evaluation framework, we demonstrate that (i) Z-score normalization outperforms min–max scaling across tasks; (ii) noise conditioning improves accuracy, coverage, and uncertainty estimation; and (iii) noisy observations significantly degrade predictive performance, thus underscoring reduced robustness under the assumed noise conditions.

2025-09-27T00:00:00Z Fahad Mahmoud Ghabban Wad Ghaban https://repositori.mypolycc.edu.my/jspui/handle/123456789/10054 2026-06-23T03:00:28Z 2025-10-28T00:00:00Z

Title: ANALYZING NON-FUNCTIONAL REQUIREMENTS OF MOBILE APPLICATIONS: USABILITY, RELIABILITY, PERFORMANCE, AND SUPPORTABILITY Authors: Fahad Mahmoud Ghabban; Wad Ghaban Abstract: Most people use mobile apps every day for a variety of purposes, including reading the news, checking social media, and shopping. Thus, these mobile apps must be thoroughly tested so that we can trust them to behave as in tended when used in the field. The purpose of this study is to collect and analyze non-functional requirements (NFRs) for mobile apps from four perspectives to determine the success and quality of the applications from the perspective of usability, reliability, performance, and supportability. By focusing on these aspects of application development, developers can create applications that are consumer-friendly, reliable, fast, and easy to maintain, thus in creasing user satisfaction and ensuring the long-term success of their products. According to the results of this study, of the 27 mobile applications that had been collected from the literature, 16 were able to meet the usability NFR, seven were able to meet the reliability NFR, 14 covered the performance NFR, and three were able to meet the supportability NFR. At the same time, three were unable to meet any NFRs. The results of the study showed that mobile application developers focused more on usability and performance than reliability and supportability. In contrast, reliability and supportability were used by a smaller percentage of developers in their development process.

2025-10-28T00:00:00Z Peng, Yizhao Zhang, Yingmin https://repositori.mypolycc.edu.my/jspui/handle/123456789/10053 2026-06-23T03:00:29Z 2024-01-01T00:00:00Z

Title: ENERGY OPTIMIZATION OF STAND ALONE ELECTRICAL GRID CONSIDERING THE OPTIMAL PERFORMANCE OF THE HYDROGEN STORAGE SYSTEM AND CONSUMERS Authors: Peng, Yizhao; Zhang, Yingmin Abstract: The surge in energy storage systems and the increasing involvement of demand-side participation can be attributed to their favorable characteristics, including their seamless integration into electrical networks and their capacity to offer operational flexibility during critical periods. This scholarly article focuses on enhancing energy utilization in an autonomous electrical grid by incorporating hydrogen storage and demand-side participation. The optimization of the stand-alone electrical grid is based on maximizing efficiency and minimizing energy consumption costs as the main objective functions are modeled. The modeling efficiency is formulated considering the ratio of the energy not supplied (ENS) to energy generation by resources. And costs of energy consumption are modeled as consumption of fuel costs by resources. The consumers’ participation is proposed based on an incentive approach to consumers for demand shaving in peak times. Also, the hydrogen storage system is installed in the stand-alone electrical grid to improve the main objectives. The particle swarm optimization (PSO) algorithm for energy optimization and solving objective functions is applied. In the end, numerical simulation is carried out in some case studies to confirm and supremacy of energy optimization with the participation of the hydrogen storage system and consumers. The case studies based on non-participation and participation of the storage system and consumers in energy management are implemented. The implementation of case studies examines the impact of both non-participation and participation of the storage system and consumers in energy management. The findings reveal that when the storage system and consumers actively participate, there is a significant improvement in efficiency and a substantial reduction in energy costs. Specifically, the efficiency is enhanced by 3% and the energy costs are reduced by 29.5% compared to the scenario where they do not participate.

2024-01-01T00:00:00Z Shreya Rao Suresh Neethirajan https://repositori.mypolycc.edu.my/jspui/handle/123456789/10052 2026-06-23T03:00:43Z 2025-08-08T00:00:00Z

Title: COMPUTATIONAL ARCHITECTURES FOR PRECISION DAIRY NUTRITION DIGITAL TWINS: A TECHNICAL REVIEW AND IMPLEMENTATION FRAMEWORK Authors: Shreya Rao; Suresh Neethirajan Abstract: Sensor-enabled digital twins (DTs) are reshaping precision dairy nutrition by seamlessly integrating real-time barn telemetry with advanced biophysical simulations in the cloud. Drawing insights from 122 peer-reviewed studies spanning 2010–2025, this systematic review reveals how DT architectures for dairy cattle are conceptualized, validated, and deployed. We introduce a novel five-dimensional classification framework—spanning application domain, modeling paradigms, computational topology, validation protocols, and implementation maturity—to provide a coherent comparative lens across diverse DT implementations. Hybrid edge–cloud architectures emerge as optimal solutions, with lightweight CNN-LSTM models embedded in collar or rumen-bolus microcontrollers achieving over 90% accuracy in recognizing feeding and rumination behaviors. Simultaneously, remote cloud systems harness mechanistic fermentation simulations and multi-objective genetic algorithms to optimize feed composition, minimize greenhouse gas emissions, and balance amino acid nutrition. Field-tested prototypes indicate significant agronomic benefits, including 15–20% enhancements in feed conversion efficiency and water use reductions of up to 40%. Nevertheless, critical challenges remain: effectively fusing heterogeneous sensor data amid high barn noise, ensuring millisecond-level synchronization across unreliable rural networks, and rigorously verifying AI-generated nutritional recommendations across varying genotypes, lactation phases, and climates. Overcoming these gaps necessitates integrating explainable AI with biologically grounded digestion models, federated learning protocols for data privacy, and standardized PRISMA-based validation approaches. The distilled implementation roadmap offers actionable guidelines for sensor selection, middleware integration, and model lifecycle management, enabling proactive rather than reactive dairy management—an essential leap toward climate-smart, welfare-oriented, and economically resilient dairy farming.

2025-08-08T00:00:00Z