Simulation-Based Evaluation of AI-Assisted ITSM Systems: A Comparative Study of Service Performance and Operational Efficiency

Abstract

Information Technology Service Management (ITSM) systems play an important role in managing operational incidents, service requests, and technical support activities within modern organizations. Traditional ITSM environments often rely on manual workflows, resulting in delayed incident resolution and reduced operational efficiency. This research presents a simulation-based framework for evaluating AI-assisted ITSM operational performance through comparative analysis and interactive modelling. The proposed framework uses operational incident datasets to compute baseline Mean Time to Resolution (MTTR) and simulate AI-assisted optimization using parameter-based reduction logic and controlled operational variability. The system was implemented using Python analytical libraries and an interactive Streamlit application that supports dataset selection, parameter adjustment, comparative evaluation, and graphical visualization. Experimental results demonstrated that AI-assisted optimization can reduce incident resolution time and improve operational efficiency under different simulation conditions. The research contributes a practical comparative evaluation framework capable of analysing baseline versus AI-assisted operational behaviour within ITSM environments through interactive simulation and visualization techniques.