Systematizing Data Preparation in Smart Manufacturing via Axiomatic Design: A Toolkit Integrating GUI and Agentic AI

Data serves as the primary driving force behind smart manufacturing, yet the transition from raw data to analysis-ready data remains a critical operational bottleneck. Often dismissed as janitorial work, data preparation consumes a disproportionate amount of analytical effort due to a reliance on fragmented, ad-hoc, and project-specific solutions. To overcome these barriers, this study proposes a systematic, requirements-driven approach grounded in the principles of axiomatic design. By cross-examining diverse manufacturing experiments, recurring operational needs were identified and mapped to a set of modular functional requirements. This theoretical foundation was materialized into a dual-approach toolkit that integrates two complementary interaction models: an interactive graphical user interface and an agentic artificial intelligence system. While the former ensures reproducibility and ground truth validation through granular manual control, the latter leverages a large language model to orchestrate complex tool sequences via natural language. The applicability of this unified architecture is validated through a comprehensive micro-drilling case study, demonstrating the seamless execution of several data preparation tasks. The findings of this study highlight the synergistic relationship between user experience and agent experience. By offering a robust, human-in-the-loop pathway, the developed system transforms data preparation from a peripheral burden into a rigorous scientific discipline. Consequently, this approach democratizes access to advanced data workflows for diverse manufacturing environments. Thus, this study contributes to the advancement of fundamental data practices within the smart manufacturing domain.