The subproject "Process Model for the Creation of Agile Learning Modules Using Data Mining" was carried out at the Chair of Machine Tool Development and Adaptive Control of Technische Universität Dresden within the collaborative research project "From a Conventional Production Plant to a Resilient Competence Factory through Industry 4.0 (Werk 4.0)". The objective of the project was the development, prototypical implementation, and scientific evaluation of a data-driven process model for the continuous creation, updating, and provision of qualification-relevant learning and information content in industrial production environments. The project was motivated by the increasing dynamics of modern manufacturing systems. Changes in products, production processes, technical documentation, software configurations, and organizational structures continuously generate new qualification requirements that are difficult to address using traditional training and education approaches alone. Against this background, the project investigated the concept of data-driven inline qualification, in which qualification needs are derived directly from changes occurring within the production and information environment. A comprehensive and modular process model was developed that combines research data management, data analytics, information extraction, change detection, and role- and context-dependent knowledge provision. The approach is based on the systematic acquisition and processing of qualification-relevant information from heterogeneous industrial data sources, including technical documentation, work and assembly instructions, process information, and multimedia content. A major focus of the project was the development of a quality-assured and continuously updateable information base as well as methods for the automated analysis of evolving information assets. To this end, concepts and methods for metadata modeling, data organization, semantic information processing, change analysis, and the reproducible handling of versioned information states were developed and investigated. The developed methods and software components were integrated into a modular demonstrator and experimentation environment and evaluated in several iterative evaluation phases together with the industrial project partners. The results demonstrate the technical feasibility of data-driven inline qualification and show that automated approaches for identifying, processing, and updating qualification-relevant information can significantly contribute to the adaptability, resilience, and learning capability of modern production systems. In addition to the technical outcomes, the project established sustainable foundations for future research, technology transfer, and reuse. These include a research data infrastructure, open and modular architectural concepts, scientifically grounded methods for processing evolving information assets, and the dissemination of results through peer-reviewed publications, open-access articles, and follow-up research activities. The project therefore provides an original scientific contribution at the intersection of research data management, knowledge management, Industry 4.0, and data-driven qualification. The developed concepts establish a robust foundation for future research and development activities in the field of adaptive, knowledge-based, and human-centered qualification systems for digital production environments.