In many situations, processes are often represented by a function that involves a response variable and a number of predictive variables. In this work, we show how to treat data whose relation between the predictive and response variables is nonlinear and, therefore, cannot be adequately represented by a linear model. This kind of data are known as nonlinear profiles. Our aim is to show how to build nonlinear control limits and a baseline prototype using a set of observed in-control profiles (Phase I analysis). Using R, we show how to afford situations in which nonlinear profiles arise and how to plot easy-to-use nonlinear control charts. This new class of control charts can be incorporated to a Statistical Process Control (SPC) strategy in order to deal with complex systems using tools that are familiar to process owners, such as control charts. The tool is also suitable for the Control phase of a DMAIC Six Sigma cycle. The SixSigma R package makes use of regularizacion theory in order to smooth the profiles. In particular, a Support Vector Machine (SVM) approach is followed, with an unattended parameters setting option. The package also allows to represent smoothed and non-smoothed profiles, and to compute the so-called prototype and confidence bands, which are actually the counterparts of center line and control limits in classical control charts, to monitor new profiles (Phase II analysis). The methods have been described in the book "Quality Control with R", within Springer's Use R! Series.