This thesis work, developed in collaboration with AvioAero - A GE Aviation Business, has the goal to analyze the behavior of an engine in windmill conditions. The windmilling phenomenon is among the sub-idle conditions of an aircraft engine. Windmilling is in far off-design conditions, because of this, the aero-thermal processes within the components are extremely complicated and neither fully understood nor simple to predict. The objective of this thesis project is the prediction of the core speed curve after engine shutdown earlier with a mathematical model. The mathematical model is based on a simple learning algorithm and consists of two different regressions.