Abstract
[No abstract available]

Abstract
According to the Portuguese Roadmap for Carbon Neutrality 2050 (RNC2050), Portugal aims to achieve carbon neutrality by 2050. To achieve this goal, it is necessary to decrease the consumption of primary energy from non-renewable sources and increase the consumption of energy from renewable sources. Portugal has a high potential for energy production through solar energy, and the country has a large solar potential that can be used. Thus, this work focuses on the study of the reliability of charging electric vehicles through photovoltaic energy, being sized electric vehicles charging stations, with different topologies, for domestic consumption, for different types of user profiles. At the same time this study evaluated technically and economically the proposed solutions. The research concluded that this type of technology proves to be a viable solution, especially if storage systems do not need to be implemented, as the limited useful lifetime of batteries substantially increases investment amortization times. Key words. Photovoltaic Systems, Electric Vehicle, Charging Stations, Energy Efficiency, Techno-Economic Study.

Abstract
This study explores the prediction and mitigation of pallet collapse during transportation within the glass packaging industry, employing a machine learning approach to reduce cargo loss and enhance logistics efficiency. Using the CRoss-Industry Standard Process for Data Mining (CRISP-DM) framework, data were systematically collected from a leading glass manufacturer and analysed. A comparative analysis between the Decision Tree and Random Forest machine learning algorithms, evaluated using performance metrics such as F1-score, revealed that the latter is more effective at predicting pallet collapse. This study is pioneering in identifying new critical predictive variables, particularly geometry-related and temperature-related features, which significantly influence the stability of pallets. Based on these findings, several strategies to prevent pallet collapse are proposed, including optimizing pallet stacking patterns, enhancing packaging materials, implementing temperature control measures, and developing more robust handling protocols. These insights demonstrate the utility of machine learning in generating actionable recommendations to optimize supply chain operations and offer a foundation for further academic and practical advancements in cargo handling within the glass industry.

Abstract
The underwater domain presents a myriad of challenges for perception systems that must be overcome to achieve accurate object detection and recognition. To augment the performance and safety of existing solutions for intricate O&M (Operations and Maintenance) procedures, AUVs must perceive the surroundings and locate potential objects of interest based on the perceived information. A depth gradient methodology is employed to survey the seabed using a multibeam sonar to perform a coarse reconstruction of the scenario that it later used to locate and identify foreign objects. This could include rocks, debris, wreckage, or other objects that may pose potential exploratory interest. First results show that the proposed method was able to detect 100 % of the objects present in the scenario with an average chamfer distance error of 0.0238m between models and respective reconstruction. © 2024 IEEE.

Abstract
Traffic forecasting problems involve jointly modeling the non-linear spatio-temporal dependencies at different scales. While graph neural network models have been effectively used to capture the non-linear spatial dependencies, capturing the dynamic spatial dependencies between the locations remains a major challenge. The errors in capturing such dependencies propagate in modeling the temporal dependencies between the locations, thereby severely affecting the performance of long-term predictions. While transformer-based mechanisms have been recently proposed for capturing the dynamic spatial dependencies, these methods are susceptible to fluctuations in data brought on by unforeseen events like traffic congestion and accidents. To mitigate these issues we propose an improvised spatio-temporal parallel transformer (STPT) based model for traffic prediction that uses multiple adjacency graphs passed through a pair of coupled graph transformer- convolution network units, operating in parallel, to generate more noise-resilient embeddings. We conduct extensive experiments on 4 real-world traffic datasets and compare the performance of STPT with several state-of-the-art baselines, in terms of measures like RMSE, MAE, and MAPE. We find that using STPT improves the performance by around 10 - 34% as compared to the baselines. We also investigate the applicability of the model on other spatio-temporal data in other domains. We use a Covid-19 dataset to predict the number of future occurrences in different regions from a given set of historical occurrences. The results demonstrate the superiority of our model for such datasets.

Abstract
One of the principles of digital democracy is to actively inform citizens and mobilize them to participate in the political debate. This paper introduces a tool that processes public political documents to make information accessible to citizens and specific professional groups. In particular, we investigate and develop artificial intelligence techniques for text mining from the Portuguese Diário da Assembleia da República to partition, analyze, extract and synthesize information contained in the minutes of parliamentary sessions. We also developed dashboards to show the extracted information in a simple and visual way, such as summaries of speeches and topics discussed. Our main objective is to increase transparency and accountability between elected officials and voters, rather than characterizing political behavior. © (2024), (SciELO-Scientific Electronic Library Online). All Rights Reserved.