Development of an Integrated Data Analytics Model for Quality Failure Management in the Flexible Packaging Industry

Quality failures in the flexible packaging industry increase production costs, generate material waste, and negatively affect manufacturing sustainability. This study developed an integrated Data Analytics model for quality failure management by combining descriptive, diagnostic, predictive, and prescriptive analytics. A dataset containing 1242 real production records was analyzed using operational variables including material grammage, production speed, ambient humidity, paper type, and reel type. Six Machine Learning classification algorithms—Logistic Regression, K-NN, SVM, Naive Bayes, Decision Tree, and Random Forest—were evaluated using Accuracy, Precision, Recall, F1-Score, and AUC-ROC. Nonlinear models achieved superior defect-detection performance compared with linear approaches. Decision Tree obtained the best overall balance between predictive performance and interpretability, achieving a Recall of 0.927 and an F1-Score of 0.962, matching Random Forest on these metrics while offering greater transparency. Random Forest achieved the highest AUC-ROC value (0.995). Model robustness was confirmed through K-Fold cross-validation, a temporal holdout on previously unseen production data, and additional sensitivity checks under class imbalance. A prescriptive optimization model was subsequently integrated with the Decision Tree classifier to identify process configurations associated with lower defect probabilities and costs before production occurs, projecting a 28.3% reduction in expected quality-related costs. The proposed framework supports data-driven quality management and contributes to sustainable manufacturing through reduced material waste and more efficient operational decision-making.