AI-Augmented 8D Problem Solving:A Design Science Framework for Intelligent Quality Management

This study explains and evaluates an AI-powered improvement to the 8D (disciplines) Problem-Solving methodology using a Design Science Research (DSR) approach. It addresses three distinct roadblocks faced in the conventional 8D approach : cognitive overload during root cause analysis, irregular problem definition and inadequate knowledge reutilization across cases. The solution offered in this paper is an integrated approach between large language models (LLM) and retrieval-augmented generation (RAG), using which a modular AI-8D artifact is designed. This artifact was developed using design sprints and evaluated using a multi-method strategy. Key evaluation metrics included root cause identification accuracy, time-to-corrective-action, report quality, cognitive load, and perceived explainability. The evaluation revealed that this AI-powered framework improves analytical accuracy, reduces problem solving lead time and improves documentation quality, while maintaining transparency and practitioner oversight. The study contributes a validated sociotechnical artifact and six transferable design principles for integrating AI into structured quality management workflows. It also extends DSR methodology by demonstrating the application of echeloned design to complex, multi-stage systems. Limitations related to domain specificity and evaluation constraints are discussed, with directions for future cross-industry validation.