Improving Application Management Services through Ticket Data Clustering

Abstract

In the area of Application Management Services (AMS), good resource planning and effective cross-skill training are critical to success. Meeting these objectives would require systematic and repeatable approaches for determining the best way of forming resource pools and identifying who to train for what skills under a constrained budget. This chapter presents a methodology that aims to achieve above goals based on an optimal clustering of service request data (aka. ticket data). Specifically, tickets that require similar problem-solving skills are first clustered using a statistical clustering technique into groups, which are then used to assist resource pooling and cross-skill training plan generation. Preliminary results have shown that an average 40% of resource saving can be achieved in our simulation scenario, while maintaining the same Service-Level Agreements (SLA) with the customer. Encouraging feedback on the cross-skill training recommendation has also been received from several real AMS customers.