Abstract
Forests worldwide have been devastated by fires. Forest fires cause incalculable damage to fauna and flora. In addition, a forest fire can lead to the death of people and financial damage in general, among other problems. To avoid wildfire catastrophes is fundamental to detect fire ignitions in the early stages, which can be achieved by monitoring ignitions through sensors. This work presents an integer programming approach to decide where to locate such sensors to maximize the coverage provided by them, taking into account different types of sensors, fire hazards, and technological and budget constraints. We tested the proposed approach in a real-world forest with around 7500 locations to be covered and about 1500 potential locations for sensors, showing that it allows obtaining optimal solutions in less than 20 min. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract
Indoor Air Quality (IAQ) pertains to the air quality within a specific space and is directly linked to the well-being and comfort of its occupants. In line with this objective, this research presents a real-time system dedicated to monitoring and predicting IAQ, encompassing both thermal comfort and gas concentration. The system initiates with a data acquisition, wherein a set of sensors captures environmental parameters and transmits this data for storage in a database. The measured parameters are analyzed by a neural network algorithm that predicts anomalies based on historical data. The neural network model generated predictions from 75.9% to 98.1% (depending on the parameter) of precision during regular situations. After that, a test with smoke in the same place was done to validate the model, and the results showed it could detect anomalies. Finally, prediction data are stored in a new database and displayed on a dashboard for monitoring in real-time measured and prediction data. © 2023 IEEE.

Abstract
Although different actions to prevent accidents at work have been implemented in companies, the number of accidents at work continues to be a problem for companies and society. In this way, companies have explored alternative solutions that have improved other business factors, such as predictive analysis, an approach that is relatively new when applied to occupational safety. Nevertheless, most reviewed studies focus on the accident dataset, i.e., the casualty’s characteristics, the accidents’ details, and the resulting consequences. This study aims to predict the occurrence of accidents in the following month through different classification algorithms of Machine Learning, namely, Decision Tree, Random Forest, Gradient Boost Model, K-nearest Neighbor, and Naive Bayes, using only organizational information, such as demographic data, absenteeism rates, action plans, and preventive safety actions. Several forecasting models were developed to achieve the best performance and accuracy of the models, based on algorithms with and without the original datasets, balanced for the minority class and balanced considering the majority class. It was concluded that only with some organizational information about the company can it predict the occurrence of accidents in the month ahead. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract
Studies dealing with route optimization have received considerable attention in recent years due to the increased demand for transportation services. For decades, scholars have developed robust algorithms designed to solve various Vehicle Routing Problems (VRP). In most cases, the focus is to present an algorithm that can overcome the shortest distances reported in other studies. On the other hand, execution time is also an important parameter that may limit the feasibility of the utilization in real scenarios for some applications. For this reason, in this work, a Guided Local Search (GLS) metaheuristic available in open-source OR-Tools will be tested to solve the Augerat instances of Capacitated Vehicle Routing Problems (CVRP). The stop criterion used here is the execution time, going from 1 s (standard) to 10 s, with a last run of 360 s. The numerical results demonstrate that increasing the execution time returns significant improvement in distance optimization. However, the optimization found considering high execution times can be expensive in terms of time, and not feasible for situations demanding faster algorithms, such as in Dynamic Vehicle Routing Problems (DVRP). Nonetheless, the GLS has proven to be a versatile algorithm for use where distance optimization is the main priority (high execution times) and in cases where faster algorithms are required (low execution times).

Abstract

Abstract
According to the World Health Organization forecast, over 55 million people worldwide have dementia, and about 10 million new cases are detected yearly. Early diagnosis is essential for patients to plan for the future and deal with the disease. Machine Learning algorithms allow us to solve the problems associated with early disease detection. This work attempts to identify the current relevance of the application of machine learning in dementia prediction in the scientific world and suggests open fields for future research. The literature review was conducted by combining bibliometric and content analysis of articles originating in a period of 20 years in the Scopus database. Twenty-seven thousand five hundred twenty papers were identified firstly, of which a limited number focused on machine learning in dementia diagnosis. After the exclusion process, 202 were selected, and 25 were chosen for analysis. The recent increasing interest in the past five years in the theme of machine learning in dementia shows that it is a relevant field for research with still open questions. The methods used to identify dementia or what features are used to identify or predict this disease are explored in this study. The literature review revealed that most studies used magnetic resonance imaging (MRI) and its types as the main feature, accompanied by demographic data such as age, gender, and the mini-mental state examination score (MMSE). Data are usually acquired from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Classification of Alzheimer's disease is more prevalent than prediction of Mild Cognitive Impairment (MCI) or their combination. The authors preferred machine learning algorithms such as SVM, Ensemble methods, and CNN because of their excellent performance and results in previous studies. However, most use not one machine-learning technique but a combination of techniques. Despite achieving good results in the studies considered, there are new concepts for future investigation declared by the authors and suggestions for improvements by employing promising methods with potentially significant results.