Abstract

Abstract
Accurate demand forecasting is essential for retail operations as it directly impacts supply chain efficiency, inventory management, and financial performance. However, forecasting retail time series presents significant challenges due to their irregular patterns, hierarchical structures, and strong dependence on external factors such as promotions, pricing strategies, and socio-economic conditions. This study evaluates the effectiveness of Transformer-based architectures, specifically Vanilla Transformer, Informer, Autoformer, ETSformer, NSTransformer, and Reformer, for probabilistic time series forecasting in retail. A key focus is the integration of explanatory variables, such as calendar-related indicators, selling prices, and socio-economic factors, which play a crucial role in capturing demand fluctuations. This study assesses how incorporating these variables enhances forecast accuracy, addressing a research gap in the comprehensive evaluation of explanatory variables within multiple Transformer-based models. Empirical results, based on the M5 dataset, show that incorporating explanatory variables generally improves forecasting performance. Models leveraging these variables achieve up to 12.4% reduction in Normalized Root Mean Squared Error (NRMSE) and 2.9% improvement in Mean Absolute Scaled Error (MASE) compared to models that rely solely on past sales. Furthermore, probabilistic forecasting enhances decision making by quantifying uncertainty, providing more reliable demand predictions for risk management. These findings underscore the effectiveness of Transformer-based models in retail forecasting and emphasize the importance of integrating domain-specific explanatory variables to achieve more accurate, context-aware predictions in dynamic retail environments.

Abstract
Artificial Intelligence has recently expanded across various applications. Machine Learning, a subset of Artificial Intelligence, is a powerful technique for identifying patterns in data to support decision making and managing the increasing volume of information. Simultaneously, Digital Twins have been applied in several fields. In this context, combining Digital Twins, Machine Learning, and Smart Buildings offers significant potential to improve energy efficiency and operational effectiveness in building management. This review aims to identify and analyze studies that explore the application of Machine Learning and Digital Twins for operation and energy management in Smart Buildings, providing an updated perspective on these rapidly evolving topics. The methodology follows the PRISMA guidelines for systematic reviews, using Scopus and Web of Science databases. This review identifies the main concepts, objectives, and trends emerging from the literature. Furthermore, the findings confirm the recent growth in research combining Machine Learning and Digital Twins for building management, revealing diverse approaches, tools, methods, and challenges. Finally, this paper highlights existing research gaps and outlines opportunities for future investigation.

Abstract

Abstract
This study investigates the integration of deep learning for single-modality and multimodal data within materials science. Traditional methods for materials discovery are often resource-intensive and slow, prompting the exploration of machine learning to streamline the prediction of material properties. While single-modality models have been effective, they often miss the complexities inherent in material data. The paper explores multimodal data integration—combining text, images, and tabular data—and demonstrates its potential to improve predictive accuracy. Utilizing the Alexandria dataset, the research introduces a custom methodology involving multimodal data creation, model tuning with AutoGluon framework, and evaluation through targeted fusion techniques. Results reveal that multimodal approaches enhance predictive accuracy and efficiency, particularly when text and image data are integrated. However, challenges remain in predicting complex features like band gaps. Future directions include incorporating new data types and refining specialized models to improve materials discovery and innovation.

Abstract
As electric vehicle (EV) adoption accelerates, residential buildings-particularly multi-dwelling structures-face increasing challenges to electrical infrastructure, notably due to conservative sizing practices of electrical feeders based on maximum simultaneous demand. Current sizing methods assume all EVs charge simultaneously at maximum capacity, resulting in unnecessarily oversized and costly electrical installations. This study proposes an optimized methodology to estimate accurate coincidence factors, leveraging simulations of EV user charging behaviors in multi-dwelling residential environments. Charging scenarios considering different fleet sizes (1 to 70 EVs) were simulated under two distinct premises of charging: minimization of current allocation to achieve the desired battery state-of-charge and maximization of instantaneous power delivery. Results demonstrate significant deviations from conventional assumptions, with estimated coincidence factors decreasing non-linearly as fleet size increases. Specifically, applying the derived coincidence factors can reduce feeder section requirements by up to 86%, substantially lowering material costs. A fuzzy logic inference model is further developed to refine these estimates based on fleet characteristics and optimization preferences, providing a practical tool for infrastructure planners. The results were compared against other studies and real-life data. Finally, the proposed methodology thus contributes to more efficient, cost-effective design strategies for EV charging infrastructures in residential buildings.