EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-SA 4.0

Artificial Intelligence Based Risk Prediction and Control Framework for Engineering Project Management Environments

Dipti Jaiswal

Soochna Shastra - The Mega Journal of Information Technology · 28.07.2026

DOI: 10.59875/1108ssjit1002 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Soochna Shastra - The Mega Journal of Information Technology, online erschienen am 28.07.2026. Verfasst von Dipti Jaiswal. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper proposes a framework called Artificial Intelligence Based Risk Prediction and Control Framework that integrates risk prediction and control in order to heighten dependability in decisions, precision in prediction, and stability operationally in the area of the management of engineering projects. This framework contains a three-phase adaptive architecture that transforms heterogeneous project information into actionable intelligence and control signals. In the first phase, data normalization and the entropy-based feature weighting method standardize project cost, schedule, and performance data, and result in a unified composite risk data. The second phase consists of adaptive temporal forecasting, which identifies the nonlinear dependencies of the project as well as the evolving patterns of the project through dynamic weighting, feedback revision, and unpredicted predictions. The third phase is when the intelligent control optimization is achieved. This is achieved when the proportional-integral-derivative actions are integrated with reinforcement learning, which helps to minimize variations and stabilize project outcomes. The framework achieved a prediction accuracy of 96.4%, 94.7% precision, 93.8% recall, 94.1% F1 score, 0.93 reliability index and 92.8% risk mitigation effectiveness. It also achieved a computing time of 92.5 ms, 95.6% model robustness, 0.89 sensitivity index, 0.092 Mean Absolute Error, 0.124 Root Mean Square Error, and 95.2% decision confidence. The framework, in terms of risk control, the prediction ability, time response, and prediction error was better than Support Vector Regression, Deep Belief Networks, and Random Forest. The results indicate that adaptive forecasting and closed-loop control integration significantly enhances proactive risk governance in engineering project environments.

Zitieren

APA

Jaiswal, D. (2026). Artificial Intelligence Based Risk Prediction and Control Framework for Engineering Project Management Environments. Soochna Shastra - The Mega Journal of Information Technology. https://doi.org/10.59875/1108ssjit1002

Vancouver

Jaiswal D. Artificial Intelligence Based Risk Prediction and Control Framework for Engineering Project Management Environments. Soochna Shastra - The Mega Journal of Information Technology. 2026. doi:10.59875/1108ssjit1002

BibTeX
@article{Jaiswal2026,
  author = {Jaiswal, Dipti},
  title = {Artificial Intelligence Based Risk Prediction and Control Framework for Engineering Project Management Environments},
  journal = {Soochna Shastra - The Mega Journal of Information Technology},
  year = {2026},
  doi = {10.59875/1108ssjit1002}
}
Kostenlos starten →