The German manufacturing industry is pressured by market changes that demand a broad and individualized product portfolio. Therefore, manufacturing has to become increasingly flexible and workers need to be trained to manage increasingly complex production environments. At the same time, societal developments affect the composition of the workforce. Demographic change, for example, yields a workforce with perceptive and cognitive limitations. However, existing training systems are static and do not adapt to the characteristics of seniors or inexperienced operators. This dissertation develops adaptive virtual training systems for industrial maintenance and changeover procedures. The adaptations are described by a classification scheme that formalizes the adaptive aspects of virtual training systems, such as interaction and visualization techniques, and how they need to adapt to the characteristics of the users. Two experiments confirm the benefits of the adaptations. The first experiment indicates that a simplification of the virtual environment can improve training for senior trainees. The second experiment showed that less detailed instructions improve training for experienced trainees. Expert evaluations and a cost-benefit analysis address the organizational perspective of virtual training systems and indicate that these systems can be adopted successfully.weiterlesen