Applying the mathematical and statistical modelling process of a system described by a set of both multivariate and time-indexed data, by computational statistic techniques (Machine Learning) in order to extract information and system behaviour patterns. In that way, predictive models are developed to describe its functioning and evolution. These predictions make reference to specifying the system status given a set of multivariate and/or temporary data. In order to develop this kind of models, they depend on the system to be studied and the data found in it. Different clustering techniques are to be applied (PCA, FA, Hierarchical Clustering, K-means, DBSCAN, OPTICS), classification and/or regression (Generalized Linear Models, Nearest Neighbors, Nave Bayes, SVM, RandomForest, XGBoost), and visualisation techniques (plotly, ggplot). Mainly, studies will be developed using Python languages (scikit-learn) and R (rpart, hclust, mclust, caret, RandomForest, nnet, gRain).