Optimising Enterprise SaaS Architecture: A Framework for Mitigating Role-Based Licensing Inflation in Multi-Pillar Oracle Cloud Deployments
Journal of Intelligent Decision Making and Information Science · 15.09.2026 · 3 (19s) · 757-780
DOI: 10.17762/jidmis.v3.6214 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Intelligent Decision Making and Information Science, online erschienen am 15.09.2026. Verfasst von Anand Rao Pydi. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Role-based licensing inflation is a practical governance challenge in multi-pillar Oracle Fusion Cloud deployments because broad job, duty, and abstract roles can expose users to privileges that influence commercial subscription measurement even when not routinely used. This study proposes a decision-driven SaaS License Governance Framework that links Oracle-style user-role-privilege drill-through evidence with calibrated machine-learning risk scores, business-access retention constraints, and cost-governance actions. The empirical layer uses 32,769 historical access decisions with nine categorical role/resource attributes from the Amazon Employee Access dataset; “denial” denotes a historically rejected access request rather than a licence cancellation. One-hot logistic regression, LightGBM, and CatBoost were compared, and the selected CatBoost model was calibrated using isotonic regression, explained with SHAP, and evaluated through bootstrap uncertainty and cold-start tests. On the untouched test set, the model achieved ROC-AUC 0.890, denial-class PR-AUC 0.507, Brier score 0.0366, and expected calibration error 0.0109. A governance threshold of 0.7414 retained 99.77% of historically approved access while flagging 1.22% of test cases, 82.5% of which were historical denials. The integrated commercial-exposure scenario reduced normalised persistent exposure by 14.03%, with sensitivity bounds of 9.05–15.72%. A worked 420-user Oracle-style example further shows how organisation-specific contractual unit costs can convert approved entitlement changes into annualised cost-avoidance metrics while clearly separating illustrative savings from actual Oracle billing. Cold-role degradation confirms that novel roles require human review. The framework therefore positions Oracle Fusion licence optimisation as an auditable, human-governed decision process rather than indiscriminate deprovisioning.
Zitieren
APA
Pydi, A. R. (2026). Optimising Enterprise SaaS Architecture: A Framework for Mitigating Role-Based Licensing Inflation in Multi-Pillar Oracle Cloud Deployments. Journal of Intelligent Decision Making and Information Science, 3(19s), 757-780. https://doi.org/10.17762/jidmis.v3.6214
Vancouver
Pydi AR. Optimising Enterprise SaaS Architecture: A Framework for Mitigating Role-Based Licensing Inflation in Multi-Pillar Oracle Cloud Deployments. Journal of Intelligent Decision Making and Information Science. 2026;3(19s):757-780. doi:10.17762/jidmis.v3.6214
BibTeX
@article{Pydi2026,
author = {Pydi, Anand Rao},
title = {Optimising Enterprise SaaS Architecture: A Framework for Mitigating Role-Based Licensing Inflation in Multi-Pillar Oracle Cloud Deployments},
journal = {Journal of Intelligent Decision Making and Information Science},
year = {2026},
volume = {3},
number = {19s},
pages = {757--780},
doi = {10.17762/jidmis.v3.6214}
}
EUVIMED