Automatic Derivation of Service Candidates from Business Process Model Repositories

Abstract. Although several approaches for service identification have been defined in research and practice, there is a notable lack of automatic analysis techniques. In this paper we take the integrated approach by Kohlborn et al. as a starting point, and combine different analysis tech-niques in an novel way. Our contribution is an automated approach for the identification and detailing of service candidates. Its output is meant to provide a transparent basis for making decisions about which services to implement with which priority. The approach has been implemented and evaluated for an industry collection of process models.