Secure Framework Enhances ERP Data for Financial Control Testing

Financial control testing increasingly relies on access to representative enterprise resource planning (ERP) data within quality assurance environments. However, directly copying production data introduces significant risks due to the presence of personal, supplier, banking, and commercially sensitive information. This research presents Secure ERP Quality Provisioning for Financial Control Testing (SEQ-FCT), a comprehensive, governed data-provisioning framework designed to address these challenges. SEQ-FCT integrates several key components: deterministic masking to protect sensitive information, synthetic scenario expansion to generate diverse test cases, referential tokenization to maintain data relationships, policy-based release approval for oversight, and automated validation for accuracy. The framework is evaluated using a single synthetic dataset comprising 186,000 finance-process records from six subsidiaries over a three-year period, including various transaction types, entity relationships, monetary values, and fraud/control-failure indicators. Because the dataset is synthetic, the results demonstrate internal consistency rather than direct production validation. Compared to various baselines like static masking and conditional tabular generative synthesis, SEQ-FCT achieved strong performance: 0.932 reconciliation F1, 0.887 fraud-trigger recall, and 0.914 control-failure F1, with an estimated leakage-risk score of 0.018. The analysis suggests that preserving financial process behavior is more reliable when masking, synthetic data generation, and governance checks are evaluated as a unified release pipeline, rather than as independent utilities.