Abstract
Disasters, both natural and man-made, are extreme and complex events with consequences that translate into a loss of life and/or destruction of properties. The advances in IT and Big Data analysis represent an opportunity for the development of resilient environments once the application of analytical methods allows extracting information from a significant amount of data, optimizing the decision-making processes. This research aims to apply the CRISP-DM methodology to extract information about incidents that occurred in the city of Lisbon with emphasis on occurrences that affected buildings, constituting a tool to assist in the management of the city. Through this research, it was verified that there are temporal and spatial patterns of occurrences that affected the city of Lisbon, with some types of occurrences having a higher incidence in certain periods of the year, such as floods and collapses that occur when there are high levels of precipitation. On the other hand, it was verified that the downtown area of the city is the area most affected by occurrences. Finally, machine learning models were applied to the data and the predictive model Random Forest obtained the best result with an accuracy of 58%.

Abstract
SAR Interferometry (InSAR) techniques are for detecting and monitoring ground deformation all over the planet. Natural disasters such as volcanoes and earthquakes deformations are among the main applications, and the great developments that we have witnessed in recent years suggests that near real-time monitoring will soon be possible. InSAR is developing fast - space agencies are launching more satellites, leading to exponential data growth. Consequently, conventional techniques cannot process all the acquired data. Modern deep learning methods can be a solution since they reach high accuracy in automatically detecting patterns in images and are fast to operate. In this work, we explore the contribution of deep learning vision transformer models to automatically detect seismic deformation in SAR interferograms. A VGG19 model is trained as baseline and ViT model uses 256x256 pixels patches and the full interferogram. The ViT model outperforms the state-of-the-art both for patch and full interferogram approaches, achieving 0.88 and 0.92 F1-score, respectively.

Abstract
Performance indicators, such as the SAIFI and the SAIDI, are commonly used by regulatory agencies to evaluate the performance of distribution companies (DisCos). Based on such indicators, it is common practice to apply penalties or grant rewards if the indicators are greater to or less than a given threshold. This work proposes a new multi-objective optimization model for pinpointing the critical assets involved in outage events based on past performance indicators, such as the SAIDI and the System Average Interruption Duration Exceeding Threshold (SAIDET) indexes. Our approach allows to retrieve the minimal set of assets in large historical interruption datasets that most contribute to past performance indicators. A case study using a real interruption dataset between the years 2011-2104 from a Brazilian DisCo revealed that the optimal inspection plan according to the decision maker preferences consist of 332 equipment out of a total of 5873. This subset of equipment, which contribute 61.90% and 55.76% to the observed SAIFI and SAIDET indexes in that period, can assist managerial decisions for preventive maintenance actions by prioritizing technical inspections to assets deemed as critical.

Abstract
Meta-learning is increasingly used to support the recommendation of machine learning algorithms and their configurations. These recommendations are made based on meta-data, consisting of performance evaluations of algorithms and characterizations on prior datasets. These characterizations, also called meta-features, describe properties of the data which are predictive for the performance of machine learning algorithms trained on them. Unfortunately, despite being used in many studies, meta-features are not uniformly described, organized and computed, making many empirical studies irreproducible and hard to compare. This paper aims to deal with this by systematizing and standardizing data characterization measures for classification datasets used in meta-learning. Moreover, it presents an extensive list of meta-features and characterization tools, which can be used as a guide for new practitioners. By identifying particularities and subtle issues related to the characterization measures, this survey points out possible future directions that the development of meta-features for meta-learning can assume.

Abstract
Manual vineyard yield estimation approaches are easy to use and can provide relevant information at early stages of plant development. However, such methods are subject to spatial and temporal variability as they are sample-based and dependent on historical data. The present work aims at comparing the accuracy of a new non-invasive and multicultivar, image-based yield estimation approach with a manual method. Non-disturbed grapevine images were collected from six cultivars, at three vineyard plots in Portugal, at the very beginning of veraison, in a total of 213 images. A stepwise regression model was used to select the most appropriate set of variables to predict the yield. A combination of derived variables was obtained that included visible bunch area, estimated total bunch area, perimeter, visible berry number and bunch compactness. The model achieved an R-2 = 0.86 on the validation set. The image-based yield estimates outperformed manual ones on five out of six cultivar data sets, with most estimates achieving absolute errors below 10%. Higher errors were observed on vines with denser canopies. The studied approach has the potential to be fully automated and used across whole vineyards while being able to surpass most bunch occlusions by leaves.

Abstract
This study investigates the potential of dynamically adjusting WIP cap levels to maximize the throughput (TH) performance and minimize work in process (WIP), according to real-time system state arising from process variability associated with low volume and high-variety production systems. Using an innovative approach based on state-of-the-art deep reinforcement learning (proximal policy optimization algorithm), we attain WIP reductions of up to 50% and 30%, with practically no losses in throughput, against pure-push systems and the statistical throughput control method (STC), respectively. An exploratory study based on simulation experiments was performed to provide support to our research. The reinforcement learning agent's performance was shown to be robust to variability changes within the production systems.