Incident Ticket Analytics for IT Application Management Services

An important IT service outsourcing business is to resolve incidents related to IT infrastructures our clients contract our company to support. Incidents are recorded as structured and unstructured data in tickets, which contain various characteristics about the incidents including timestamps, description and resolution Analyzing such incident tickets becomes a critical task in managing the operations of the service in order to keep the operations within the agreed upon service level agreement. Ticket analytics is essential to identify anomalies and trends, as well as detect unusual patterns in the operations; such analysis is hard to do manually especially for large accounts with complex organization and scopes. This paper focuses on ticket analytics and some key statistical techniques applied in the analyses. Finally, we use real-data examples to demonstrate these techniques and discuss major challenges of ticket analyses.