"The analysis of surgical activities became a popular field of research in recent years and various methods have been published to detect surgical phases from manifold data sources in operating rooms. Goal of this research is to develop a method for extracting realtime information of surgical activities. In this work we use fine-grained data of surgical devices and operating room equipment which is produced permanently during a surgery. This low-level data help describing the current surgical phases and reflect real-time status of the endoscope, insufflator, electrosurgical devices and light sources. This is the basis for the development of a structured process to extract surgical phase recognition models. The developed artifact is constructed by adapting the method engineering methodology to find a best practice for utilizing this fine-grained data for automatic intra-surgical activity detection. Further we show how to integrate expert knowledge and transfer the modeled information into an automated and scalable information system for surgical phase recognition. We evaluated our method with 15 data sets of laparoscopic surgeries and obtained an accuracy rate of about 83% with this approach."