Human-in-the-Loop Automation Workflow for SAR Review and Escalation Using Low-Code Platforms

Suspicious Activity Reports (SARs) constitute the principal regulatory mechanism through which financial institutions identify, investigate, and communicate potentially illicit financial behavior under Anti-Money Laundering (AML) compliance frameworks. Although advances in transaction monitoring, anomaly detection, and behavioral analytics have significantly improved alert-generation capabilities, SAR preparation remains an inherently investigative process requiring contextual interpretation, documentation of investigative reasoning, and supervisory validation [1]. Existing SAR preparation workflows often rely on manually coordinated investigative activities across fragmented enterprise systems, resulting in delays, inconsistent reporting practices, and increased exposure to regulatory scrutiny [2]. This paper proposes a human-in-the-loop (HITL) automation workflow for SAR review and escalation implemented through low-code orchestration platforms. The proposed approach integrates automated evidence aggregation, behavioral timeline construction, case routing, and escalation governance while preserving investigators as final decision authorities responsible for determining suspicion and filing outcomes [3]. By embedding structured automation within governed compliance workflows, the proposed architecture improves investigative transparency, reporting consistency, and auditability while maintaining alignment with regulatory expectations that SAR determinations remain subject to documented human judgment and supervisory oversight [4].