The aim of the paper is to present an approach to optimal control of biotechnological processes, described by fuzzy sets which is based on genetic algorithms (GA). It allows more adequate, flexible and intelligent (human oriented) description of bioprocess dynamic and inclusion of this information in an effective and robust optimization procedure. Related algorithms and software are designed and tested with data from the real biotechnological process of -d-glucose-O2-oxidase-1 synthesis. The proposed approach and algorithms are applicable to other bioprocesses.