The accuracy of a predictive system is critical for predictive maintenance and to support the right decisions at the right times. Statistical models, such as ARIMA and SARIMA, are unable to describe the stochastic nature of the data. Neural networks, such as long short-term memory (LSTM) and the gated recurrent unit (GRU), are good predictors for univariate and multivariate data. The present paper describes a case study where the performances of long short-term memory and gated recurrent units are compared, based on different hyperparameters. In general, gated recurrent units exhibit better performance, based on a case study on pulp paper presses. The final result demonstrates that, to maximize the equipment availability, gated recurrent units, as demonstrated in the paper, are the best options.

Maintenance professionals and other technical staff regularly need to learn to identify new parts in car engines and other equipment. The present work proposes a model of a task assistant based on a deep learning neural network. A YOLOv5 network is used for recognizing some of the constituent parts of an automobile. A dataset of car engine images was created and eight car parts were marked in the images. Then, the neural network was trained to detect each part. The results show that YOLOv5s is able to successfully detect the parts in real time video streams, with high accuracy, thus being useful as an aid to train professionals learning to deal with new equipment using augmented reality. The architecture of an object recognition system using augmented reality glasses is also designed.

This paper presents a case study and a model to predict maintenance interventions based on condition monitoring of diesel engine oil in urban buses by accompanying the evolution of its degradation. Many times, under normal functioning conditions, the properties of the lubricants, based on the intervals that manufacturers recommend for its change, are within normal and safety conditions. Then, if the lubricants’ oil condition is adequately accompanied, until reaching the degradation limits, the intervals of oil replacement can be enlarged, meaning that the buses’ availability increases, as well as their corresponding production time. Based on this assumption, a mathematical model to follow and to manage the oil condition is presented, in order to predict the next intervention with the maximum time between them, which means the maximum availability.

The use of clean and renewable energy sources is increasingly important, for economic and environmental reasons. Wind plays a key role among renewable energy sources. Hence, the location, monitoring and maintenance of wind turbines are areas that have received more and more attention in recent years. The paper presents a survey of datasets of wind resources, wind farm installed capacity and wind farm operation, which contain generous amounts of data. Those datasets are important tools, freely available for analysis of wind resources and study of the performance of wind turbines. A short analysis of one of the datasets is also presented, identifying different operational regions, and the ones more likely to aggregate failures. Principal Component Analysis (PCA) is used to study wind turbines’ behavior.