Year: 2025
Type: article
Abstract: Testing large-scale systems like T2 RTGS, which integrates numerous interfaces for end-to-end payment flows, requires robust automation to reduce complexity and risks. This article explores the application of cognitive automation techniques, combining genetic algorithms and computer vision, to transform traditional quality assurance workflows in financial infrastructure testing. Genetic algorithms are utilized to optimize test case prioritization, focusing resources on high-risk integration points and enabling faster validation cycles. For monitoring SWIFT message queues in Opics and FX systems, computer vision techniques automate real-time anomaly detection, flagging discrepancies without manual oversight. Additionally, the article highlights the implementation of machine learning-enhanced reconciliation models that significantly reduce false positives in payment discrepancies by learning from historical resolution records. By presenting measurable results and demonstrating AI-centric testing strategies, this article offers a technical roadmap for QA professionals facing complex integration challenges in financial systems, showing how cognitive automation not only detects errors faster but also fosters greater collaboration through end-to-end integration testing.
Author: Aparna Thakur
Language: English
Subfield: Software
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4409669908