Application of RCA and AHP to Identify Functional Testing Failures in Loan Origination Systems by QA Engineers

Abstract

Recurring functional testing failures in DJOIN’s digital banking systems during 2023–2025 created operational risks, particularly in the Balance Validation Dashboard and financial transaction modules. This study aimed to identify the principal causes of recurring failures and prioritize appropriate improvement solutions by integrating Root Cause Analysis (RCA) and the Analytical Hierarchy Process (AHP). The analysis used 1,759 production bug tickets, comprising 1,724 tickets from the Coopmax system and 35 tickets from the KOCEK loan origination system. RCA was conducted using a fishbone diagram based on the 6M framework and 5 Whys analysis, supported by bug-tracking records, system logs, testing documentation, and stakeholder interviews. The analysis identified 12 root causes. The three highest-severity causes were hardcoded product-type assumptions in the Balance Validation Dashboard query, with a severity score of 4.67; the absence of a regression test suite after hotfix implementation, with a score of 4.33; and distributed system logs that complicated failure tracing, with a score of 4.33. AHP evaluation assigned the highest weight to solution effectiveness at 42.44%, followed by business impact at 30.65%, implementation cost at 13.07%, implementation time at 8.36%, and technical complexity at 5.48%. The comparison matrix achieved a consistency ratio of 0.024, indicating acceptable judgment consistency. Among 12 alternatives, automated testing tools ranked first with a priority score of 0.1872, followed by test-driven development at 0.1412 and CI/CD implementation at 0.1372. These findings provide an evidence-based prioritization of proposed corrective and preventive actions; however, their actual effectiveness requires post-implementation evaluation.