SAP S/4HANA-Based Continuous Auditing and Fraud Detection in Integrated Accounting Systems

• Fazle Hakeem Ghory

  Author

This paper presents a meta-model for real-time fraud detection within enterprise accounting systems, specifically targeting SAP S/4HANA. Using Design Science Research, we derive six design principles from literature analysis and semi-structured interviews with SAP practitioners. The proposed framework integrates data acquisition, real-time analytics (combining rule-based checks, unsupervised anomaly detection, and supervised classification), process mining, and a decision support layer. The framework is evaluated using anonymized transactional data from a manufacturing company (12 months, ~1.2 million transactions). Compared to a baseline rule-only system, our framework achieves an average F1-score of 91.2% (vs. 74.5% for baseline) with a false positive rate of 4.8%. Processing latency averages 1.8 seconds per transaction with 15% CPU overhead. A simulated collusion scenario demonstrates early detection (after the third suspicious transaction) compared to baseline (alert only after payment). This research contributes a validated, real-time auditing solution tailored for SAP S/4HANA environments.

• PDF

Copyright (c) 2026 International Journal of Adaptive Management and Business Intelligence

This work is licensed under a Creative Commons Attribution 4.0 International License.