Introduction

With upcoming data science and neural network knowhow, a lot of application fields are emerging in the process industry. In order to use these new methods (e.g. for control algorithms), it is reasonable to gain experiences with small scale apparatus, but comparable processes already in the laboratory. The following brief overview shows that there are good examples for procedures and interoperability when adopting ML solutions. The main focus is set on an easy-to-follow procedure for the integration of machine-learning ML solutions following the example of Min et al.  , which is described in the following in more detail.