EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An Explainable Hybrid Artificial Intelligence Framework for Predictive Analytics and Intelligent Decision-Making in Complex Information Systems

Srikanth Girijala, Amit Rathod, Ashish Shrivastav, Dr. Deepjyoti Kalita, Dr. Mukesh Kumar Chandrakar, Dr. Rakhi Dua

Computer and Decision Making An International Journal · 03.11.2026

DOI: 10.1922/comdem.v3i6s.18907 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer and Decision Making An International Journal, online erschienen am 03.11.2026. Verfasst von Srikanth Girijala und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The growing complexity of modern information systems has increased the need for intelligent predictive frameworks that can identify operational failures while providing transparent, actionable decision support. Preprocessing, handling class imbalance, baseline machine-learning algorithms, a stacking-based hybrid ensemble, SHAP explainability, and a risk-oriented decision-support layer were used in a quantitative computational design. The predictive models used were Logistic Regression, Decision Tree, Support Vector Machine, Random Forest, XGBoost and Artificial Neural Network, and the hybrid framework involved the combination of predictions from the Random Forest, XGBoost and ANN models. Random Forest outpaced all the other models in terms of accuracy, precision, specificity, F1-score and Matthews correlation coefficient, whereas the hybrid stacking model had the highest ROC-AUC (0.9762) and PR-AUC (0.8056) with 89.71% recall. In the hybrid model, 61 machine failures were correctly identified, and 7 were missed. The most important factors determining the risk of failure were identified by the SHAP analysis as: torque, wear of the tool, rotational speed and air temperature. Overall, the proposed framework showed that a framework that integrates predictive performance, interpretability and risk-based decision support can improve intelligent maintenance systems that are transparently useful and operationally effective. The framework also provided a translation of predicted failure probabilities into maintenance-priority categories, aiding the monitoring, preventive action, and human-centred operational decision-making in industrial settings.

Zitieren

APA

Girijala, S., Rathod, A., Shrivastav, A., Kalita, D. D., Chandrakar, D. M. K., & Dua, D. R. (2026). An Explainable Hybrid Artificial Intelligence Framework for Predictive Analytics and Intelligent Decision-Making in Complex Information Systems. Computer and Decision Making An International Journal. https://doi.org/10.1922/comdem.v3i6s.18907

Vancouver

Girijala S, Rathod A, Shrivastav A, Kalita DD, Chandrakar DMK, Dua DR. An Explainable Hybrid Artificial Intelligence Framework for Predictive Analytics and Intelligent Decision-Making in Complex Information Systems. Computer and Decision Making An International Journal. 2026. doi:10.1922/comdem.v3i6s.18907

BibTeX
@article{Girijala2026,
  author = {Girijala, Srikanth and Rathod, Amit and Shrivastav, Ashish and Kalita, Dr. Deepjyoti and Chandrakar, Dr. Mukesh Kumar and Dua, Dr. Rakhi},
  title = {An Explainable Hybrid Artificial Intelligence Framework for Predictive Analytics and Intelligent Decision-Making in Complex Information Systems},
  journal = {Computer and Decision Making An International Journal},
  year = {2026},
  doi = {10.1922/comdem.v3i6s.18907}
}
Kostenlos starten →