A SYSTEMATIC REVIEW OF AI AND MACHINE LEARNING-DRIVEN IT SUPPORT SYSTEMS: ENHANCING EFFICIENCY AND AUTOMATION IN TECHNICAL SERVICE MANAGEMENT

Abstract

The rapid evolution of Artificial Intelligence (AI) and Machine Learning (ML) has brought significant advancements in IT support systems, transforming the efficiency, automation, and responsiveness of technical service management (TSM). Traditional IT support methods, which rely heavily on manual troubleshooting, rule-based ticketing systems, and reactive maintenance approaches, often suffer from delayed issue resolution, increased operational costs, and inefficiencies in service management. This systematic review, analyzing 563 peer-reviewed studies published before 2023, investigates the application of AI-driven solutions in automated troubleshooting, predictive maintenance, intelligent ticketing systems, and AI-powered virtual assistants. The findings indicate that AI-driven troubleshooting models reduce mean time to resolution (MTTR) by 50-60%, improving system uptime and minimizing service disruptions. Predictive maintenance models leveraging ML algorithms achieve up to 90% accuracy in failure detection, leading to a 40-50% reduction in unplanned downtime and optimizing IT infrastructure reliability. AI-based intelligent ticketing systems enhance classification accuracy by 50-60%, reducing misclassification errors by 30-40%, while sentiment-based prioritization improves critical incident response speed by 35%, ensuring faster resolution of high-priority issues. Additionally, AI-powered virtual assistants autonomously manage 50-60% of IT service requests, significantly decreasing first-level support workload by 40% and enabling IT personnel to focus on complex technical challenges. Despite these advancements, challenges persist, including algorithmic bias, model misclassification risks, and limitations in handling complex, non-standard IT issues, which impact the overall effectiveness of AI-driven IT support automation. A comparative analysis between AI and human-led IT support reveals that while AI-driven systems outperform human-led models in automation, scalability, and cost efficiency, human intervention remains critical for addressing high-complexity IT problems, strategic decision-making, and exception handling. This review highlights the transformative role of AI in IT service management, emphasizing its capabilities in optimizing IT workflows, improving service efficiency, and reducing operational burdens. However, the findings also reinforce the need for continuous improvements in AI fairness, adaptability, interpretability, and hybrid AI-human integration models to maximize the benefits of AI-driven IT support systems.