Simulation-Driven Predictive AI for Printer Repair Services: A Proactive Ticket Resolution Approach

This study presents a predictive, artificial intelligence–driven ticket resolution framework aimed at improving customer service efficiency within the printer maintenance domain, a critical technical sup-port area for both organizational and individual users. Owing to the limited availability of reliable, publicly accessible real-world maintenance data, the proposed system is developed and evaluated using a simulation-based approach. A synthetic dataset comprising 150 maintenance cases and 10 relevant attributes was generated to realistically model common printer service scenarios. Twelve machine learning classification algorithms were implemented and systematically evaluated using two distinct random states (20 and 42) to ensure robustness and reproducibility. Model performance was assessed using accuracy and F-score metrics. The experimental results demonstrate that the Light Gradient Boosting Machine (LightGBM) classifier outperformed the other models, achieving an accuracy of 0.70 and an F-score of 0.7033. These findings confirm the feasibility and effectiveness of simulation-driven predictive modeling for proactive ticket resolution in contexts where real-world data are unavailable or incomplete. The study underscores the potential of artificial intelligence to transition customer support systems from traditional reactive mechanisms toward proactive, intelligent service management solutions in operational maintenance environments.