Machine Learning in IT Service Management

IT Service Management (ITSM) is a variety of activities directed towards maintenance of IT infrastructure. Hence, it is considered to be an important activity for any company, even to one not related to IT. Time of incidents’ resolution is the key performance indicator for ITSM. To reduce resolution time, authors propose infrastructure incident prediction model. Model is based on machine learning technologies. Application of machine learning models in ITSM allows significant improves in customer experience and handling issues more efficiently, decreasing service desk agents’ efforts and reducing service costs. This paper aims to propose predictive method of the incident resolution time estimation. Proposed model derives insights from incident data and predicts estimated time of resolution, allowing detection of incident resolution’s de-lays. Authors analyzed prediction accuracy, and derived results demonstrating that model can assist with prediction for a large set of incidents using dataset based on a real service desk incident data. Additionally, its practical use is applicable for service improvement process.