Abstract
Low Power Wide Area Networks (LP-WAN) are receiving a lot of attention because of their ability to communicate using radio frequency in long distances, with low-power consumption and low-cost devices. In this paper, we provide a comparison between the two LP-WAN platforms that are leading the market, the Sigfox and the LoRaWAN, based on the literature. Both platforms are analyzed considering the context of the forest fire detection and verification systems. Many aspects are being considered to identify which LP-WAN is more adequate to be used in this kind of systems, such as battery lifetime, coverage range, business model and costs. The comparison shows that both platforms are very similar in most of the aspects, although LoRaWAN is more flexible than Sigfox on the deployment and management of the network infrastructure. LoRaWAN allows customers to implement and manage their own infrastructure network, which is essential in systems which monitor vast forest areas.

Abstract
The Machine Learning approach is used in several application domains, and its exploitation in predicting accidents in occupational safety is relatively recent. The present study aims to apply different Machine Learning algorithms for classifying the occurrence or non-occurrence of accidents at work in the retail sector. The approach consists of obtaining an impact score for each store and work unit, considering two databases of a retail company, the preventive safety actions, and the action plans. Subsequently, each score is associated with the occurrence or non-occurrence of accidents during January and May 2023. Of the five classification algorithms applied, the Support Vector Machine was the one that obtained the best accuracy and precision values for the preventive safety actions. As for the set of actions plan, the Logistic Regression reached the best results in all calculated metrics. With this study, estimating the impact score of the study variables makes it possible to identify the occurrence of accidents at work in the retail sector with high precision and accuracy.

Abstract
Weather change plays an important role in work-related accidents, it impairs people's cognitive abilities, increasing the risk of injuries and accidents. Furthermore, weather conditions can cause an increase or decrease in daily sales in the retail sector by influencing individual behaviors. The increase in transactions, in turn, leads employees to fatigue and overload, which can also increase the risk of injuries and accidents. This work aims to conduct a case study in a company in the retail sector to verify whether the transactions records in stores and the weather conditions of each district in mainland Portugal impact the occurrence of work accidents, as well as to perform predictive analysis of the occurrence or non-occurrence of work accidents in each district using these data and comparing different machine learning techniques. The correlation analysis of the occurrence or non-occurrence of work accidents with weather conditions and some transactions pointed out the nonexistence of correlation between the data. Evaluating the precision and the confusion matrix of the predictive models, the study indicates a predisposition of the models to predict the non-occurrence of work accidents to the detriment of the ability to predict the occurrence of work accidents.

Abstract
Research have been driven by the increased demand for delivery and pick-up services to develop new formulations and algorithms for solving Vehicle Routing Problems (VRP). The main objective is to create algorithms that can identify paths considering execution time in real-world scenarios. This study focused on using the Guided Local Search (GLS) metaheuristic available in OR-Tools to solve the Capacitated Vehicle Routing Problem with Time Windows using the Solomons instances. The execution time was used as a stop criterion, with short runs ranging from 1 to 10 s and a long run of 360 s for comparison. The results showed that the GLS metaheuristic from OR-Tools is applicable for achieving high performance in finding the shortest path and optimizing routes within constrained execution times. It outperformed the best-known solutions from the literature in longer execution times and even provided a close-to-optimal solution within 10 s. These findings suggest the potential application of this tool for dynamic VRP scenarios that require faster algorithms.

Abstract
Deep learning networks, more specifically convolutional neural networks, have shown a notable distinction when it comes to computer vision problems. Their versatility spans various domains, where they are applied for tasks such as classification and regression, contingent primarily on the availability of a representative dataset. This work explores the feasibility of employing this approach in the domain of agriculture, particularly within the context of olive growing. The objective is to enhance and facilitate cultivar identification techniques by using images of olive tree leaves. To achieve this, a comparative analysis involving ten distinct convolutional networks (VGG16, VGG19, ResNet50, ResNet152-V2, Inception V3, Inception ResNetV2, XCeption, MobileNet, MobileNetV2, EfficientNetB7) was conducted, all initiated with transfer learning as a common starting point. Also, the impact of adjusting network hyperparameters and structural elements was explored. For the training and evaluation of the networks, a dedicated dataset was created and made available, consisting of approximately 4200 images from the four most representative categories of the region. The findings of this study provide compelling evidence that the majority of the examined methods offer a robust foundation for cultivar identification, ensuring a high level of accuracy. Notably, the first nine methods consistently attain accuracy rates surpassing 95%, with the top three methods achieving an impressive 98% accuracy (ResNet50, EfficientNetB7). In practical terms, out of approximately 2016 images, 1976 were accurately classified. These results signify a substantial advancement in olive cultivar identification through computer vision techniques.

Abstract
Unmanned Aerial Vehicles (UAVs) are being deployed in different applications due to their reduced time execution to perform tasks, more extensive coverage area, and more risk minimization to humans. In the Oil & Gas industry, its use for inspection activities is even more attractive due to the large structures in these facilities. Therefore, this research proposes deploying an autonomous UAV system to inspect unburied pipelines of onshore O&G facilities. The proposed UAV guiding system is based on image processing techniques Canny edge detection and Hough Transform to detect the line and on a path follower algorithm to generate the trajectory. The proposed strategy was developed in Robot Operating System (ROS) and tested in a simulated environment considering the practical oper-ational. The same controller was tested on a physical UAV to validate the results obtained in previous simulations. The results demonstrated the effectiveness and feasibility of deploying the proposed strategy for this specific task and the cost reduction potential for real-life operations, as well as reduced potential risks to the physical integrity of the workers.