Purpose The study aims to identify trends, assess research status by country and SCOR standard processes, and provide machine learning-based classification methods in the research area.
Methods Qualitative analysis is conducted on a total 493 literature data . And nine machine learning models to classify the SCM process are performed Results Artificial intelligence in Supply Chain research has been consistent since 1998, with a noteworthy concentration from 2017 to 2023. High-performing models (XGBoost, LightGBM, CatBoost) achieve F1 values exceeding 70% and AUC values over 80%, effectively classifying literature data into SCOR's 6 processes.
Conclusion In Supply Chain research, focus is on 'Enable' for decision support and 'Plan' for predictive optimization,. This addresses global AI-driven efficiency demands for strengthening domestic supply chains. Korea's 1% contribution requires policy-driven activation. Employing ML-based classification enhances future analysis, allowing ongoing research and easy referencing for national and corporate endeavors.