XAI-IM: Explainable AI Framework for Injection Molding Quality Prediction in Smart Manufacturing

Injection molding is mainly used in manufacturing processes where maintaining consistent product quality remains a difficult challenge. Although machine learning (ML) techniques have improved quality assessment, their black-box nature limits transparency and reduces confidence in industrial decision-making. To address this limitation, this paper proposes XAI-IM, a domain-specific explainable artificial intelligence (XAI) approach for injection molding. The proposed approach integrates machine learning, explainability, root cause analysis, decision support, and continuous learning into a unified architecture. A conceptual use case demonstrates its capabilities to support transparent process optimization and intelligent manufacturing decisions. Future work will validate the proposed approach using real industrial datasets and evaluate its applicability in Industry 4.0-enabled smart manufacturing environments.