Applying Agile and AI methodologies to incident management in mobile networks

Abstract

Effective incident management is important for maintaining service continuity and user satisfaction in the ever-changing telecommunications industry of today. Conventional methods frequently depend on strict, sequential workflows that impede responsiveness and delay resolution. To overcome these constraints, this paper suggests an improved incident management framework that incorporates AI and Agile approaches. We simulate both a traditional resolution model and an enhanced Agile and AI scenario using a critical fiber cut that impacts more than 15,000 ONUs as a case study. The improved model uses AI to automatically classify incidents, correlate data, and assign tasks based on roles, while Agile sprint planning allows for parallel execution and real-time progress monitoring. Time-boxed sprints are used to organise tasks, improving coordination and cutting down on escalation delays. In comparison to the conventional method, simulation results indicate a 20% reduction in overall expenses and a 1h 45m reduction in resolution time. The results imply that a more proactive, open, and effective response to network incidents is made possible by fusing AI automation with Agile principles. By showing how Agile and AI can work together to optimise incident workflows, increase resilience, and improve service quality in critical infrastructure environments, this study adds to the expanding body of research on digital transformation in telecom operations.