data-driven discrete-event simulation of open banking api access granting

Open banking platforms must manage volatile api request streams alongside complex, multi-step approval workflows. this study investigates operational bottlenecks and workforce optimization within a multi-stage request review process, with direct implications for waiting time, turnaround time, and overall system responsiveness. a dynamic representation of the workflow is developed using data-driven discrete-event simulation in the anylogic environment. unlike purely theoretical models, the proposed framework incorporates real operational records: inter-arrival times are derived from historical logs and embedded into the simulation to reproduce realistic traffic fluctuations. each incoming request is modeled as an agent characterized by specific attributes and routed through a queueing network consisting of preliminary assessment, specialist evaluation, and technical validation stages. experimental results demonstrate that integrating actual event sequences significantly enhances model fidelity, particularly in capturing peak-load dynamics and stress conditions. the model enables scenario-based analysis of staffing configurations and resource allocation strategies. ultimately, the study delivers a decision-support tool that helps managers balance employee utilization with service quality while improving process stability and operational performance in banking ecosystems.