Aplicación de redes neuronales artificiales en la estimación del nivel de servicio en un CALLCENTER.

Abstract

In the present project the construction of an artificial neural network is done to determine the degree or level of service of a Call Center. This network is elaborated with the purpose of demonstrating that through this type of computational model the quality of the service in the Call Center can be predicted, that is, to be able to determine the efficiency of attention of the operators in function of the demand of calls that they enter We start by giving a concise description of the traditional models of determination or prognosis of a variable, as well as detailing the composition and functioning of artificial neural networks, both in their topologies, learning algorithms, applications in different areas and their utility today. After this, the problem to be solved is defined by specifying the method to be used, the prediction models, exponential adjustment methods and different topologies of artificial neural networks in order to reach the desired solution. With these add-ons, a neural network is implemented using MINITAB as an interface and as a learning element SOLVER, with the purpose of seeing the estimation errors and the training stages of the associated models for this study based on comparative analyzes.