The manufacturing industry is faced with strong competition making the companies’ knowledge resources and their systematic management a critical success factor. Yet, existing concepts for the management of process knowledge in manufacturing are characterized by major shortcomings. Particularly, they are either exclusively based on structured knowledge, e. g., formal rules, or on unstructured knowledge, such as documents, and they focus on isolated aspects of manufacturing processes. To address these issues, we present the Manufacturing Knowledge Repository, a holistic repository that consolidates structured and unstructured process knowledge to facilitate knowledge management and process optimization in manufacturing. First, we define requirements, especially the types of knowledge to be handled, e. g., data mining models and text documents. Next, we develop a conceptual repository data model associating knowledge items and process components such as machines and process steps. Furthermore, we discuss implementation issues including storage architecture variants and present both an evaluation of the data model and a proof of concept based on a prototypical implementation.