In the context of Industry 4.0, managing large amounts of data is essential to ensure informed decision-making in intelligent production environments. It enables, for example, predictive maintenance, which is essential for anticipating and identifying causes of failures in machines and equipment, optimizing processes, and promoting proactive management of human, financial, and material resources. However, generating accurate information for decision-making requires adopting suitable data preprocessing and analysis techniques. This study explores the identification of machine failures based on synthetic industrial data. Initially, we applied the feature selection techniques Principal Component Analysis (PCA), Minimum Redundancy Maximum Relevance (mRMR), Neighborhood Component Analysis (NCA), and Denoising Autoencoder (DAE) to the collected data and compared their results. In the sequence, a comparison among three widely known machine learning classifiers, namely Random Forest (RF), Support Vector Machine (SVM), and Multilayer Perceptron neural network (MLP), was conducted, with and without considering feature selection. The results showed that PCA and RF were superior to the other techniques, allowing the classification of failures with rates of 0.98, 0.97, and 0.98 for the accuracy, precision, and recall metrics, respectively. Thus, this work contributes by solving an industrial problem and detailing techniques to identify the most relevant variables and machine learning algorithms for predicting machine failures that negatively impact production planning. The findings provided by this study can assist industries in giving preference to employing sensors and collecting data that can contribute more effectively to machine failure predictions.

A paradigm of Industry 4.0 is to promote interconnectivity between virtual and physical components (such as sensors, equipment, and other assets), both with people, with each other, and with the Internet. For this purpose, ordinary machines must be transformed into sentient and conscious devices capable of learning independently to improve their overall performance and maintenance in interaction with their surroundings. Existing processes will become smart enough to significantly reduce their failures and quickly adapt for a more flexible and productive scenario. Real-time monitoring and traceability of product states and positions are some of the most significant needs of Industry 4.0. Therefore, information and communication technologies are essential for realizing these systems.

Tradicionalmente, a manufatura se baseava em uma economia linear, extraindo, fabricando, utilizando e descartando recursos, levando o planeta em direção à exaustão dos recursos. A Indústria 4.0 introduz tecnologias habilitadoras que podem promover estratégias circulares, enfrentando o desafio de reduzir o uso de recursos não renováveis. Este estudo modela um framework conceitual aplicando tecnologias digitais a uma economia circular, revisando sistematicamente a literatura específica utilizando a metodologia PRISMA. Os dados foram coletados por meio de estudos de caso envolvendo quatro empresas dos setores automotivo, eletrônico e de calçados. Os resultados confirmam a Internet das Coisas (IoT) e o Big Data como tecnologias cruciais para aplicações da economia circular. A integração bem-sucedida dessas tecnologias na Indústria 4.0 está prestes a melhorar a sustentabilidade industrial em vários setores de manufatura.