The quality and service levels of a Service Desk are strongly correlated with the operators’ ability to respond pertinently to customer needs, by effectively leveraging the knowledge base already acquired on users’ behavioral profiles, as well as on interventions carried out over time.
The applications supporting Service Desk operators are fundamentally of two types: (i) CRM (Customer Relationship Management) systems, or even better, TTMS (Trouble Ticket Management Systems); (ii) document management and retrieval systems.
In both cases, the technologies currently available on the market present limitations: in fact, whether using commercial or open-source technologies, TTMS provide few basic mechanisms for user profiling, while document management systems provide full-text search mechanisms, which return documents inessential to the problem being solved or omit documents relevant to the application case in question.
The present Research and Development proposal aims at the design and implementation of an integrated system for the dynamic profiling and behavioral analysis of service desk users, as well as for the semantic processing of trouble tickets. This aims to leverage the know-how encapsulated within the organization’s knowledge base, in order to automatically suggest the best response to operators following a request for intervention and/or to activate an unsupervised process for identifying the reported problem, based on the interpretation of the email text and the automatic sending of emails to the user targeted at retrieving all useful information for opening a ticket.
The resulting platform, named iDESK, will allow for the overcoming of the limitations of traditional technologies and will enable the proposing company (Innovaway) to achieve: (i) a digital innovation in the delivery of service desk services, which will become more effective and efficient; (ii) an environment equipped with Business Analytics capabilities capable of processing enormous quantities of heterogeneous data (Big Data), especially textual data.
START DATE: March 2017
END DATE: February 2019
