ArgCBR Call Centre


Nowadays, a good customer support can be the key to increase the satisfaction of customers and hence, the profits of the company that provides it. Commonly, companies offer customer support to their customers by means of a call centre. Call centres are systems which control every process implicated in the provision of technological and customer support services to private or public organisations. In a call centre, there are a number of technicians whose role is to provide the customers with technical assistance for their problems (also known as tickets). The call centre technicians have computers provided with a helpdesk software and phone terminals connected to a telephone switchboard that balances the tickets among them.

The experience and skills of technicians are decisive to obtain a quick and accurate response to the customers’ needs. Therefore, this experience must be appropriately stored and managed for the call centre software to improve the performance of the whole system. However, in many companies the technicians that work for a specific project sign secrecy clauses and are not able to share certain knowledge with other technicians working in different projects by default. Furthermore, when a group of technicians work together to solve a complex problem conflicts of opinion can arise.

Multi-Agent Systems (MAS) are suitable to provide a framework to perform collaborative processes in distributed environments. These systems allow to have the data distributed among agents and to preserve the knowledge privacy. In this paper, we present a call centre application that allows technicians to take the maximum profit of their knowledge by arguing and reaching agreements about the best solution to provide to the customer. The application is based on a case-based argumentation infrastructure for agent societies that we have designed.


This work is supported by the Spanish government grants [CONSOLIDER-INGENIO 2010 CSD2007-00022, TIN2011- 27652-C03-01, and TIN2012-36586-C03-01] and by the GVA project [PROMETEO 2008/051].