Optimisation of the production workflow is a complicated task, since it is nearly impossible for a person to predict bottlenecks and rising cycle times. Discrete event simulations are built with a purpose of monitoring and predicting such problems occurring. But all information from simulations should be analysed by a final user. In order to create a process for prediction and optimization two automated methods were developed in this project. The first method utilize statistical methods for monitoring separate stations. The second method is using machine learning methods for analysing the whole simulation at once to investigate…