Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Optimization of customer satisfaction using Bayesian networks

: Kempf, Michael

Fernandez, Jeffrey E. (Ed.) ; Society for Industrial and Systems Engineering -SISE-:
2nd Annual World Conference of the Society for Industrial and Systems Engineering 2013. Proceedings : Las Vegas, Nevada, USA, November 5-7, 2013
Binghamton/NY, 2013
ISBN: 9781938496011
Society for Industrial and Systems Engineering (SISE Annual World Conference) <2, 2013, Las Vegas/Nev.>
Conference Paper
Fraunhofer IPA ()
Bayesian network; Kundenzufriedenheit

Customer satisfaction is a measure of the degree to which a product or service meets the customer's expectations. Within this paper we will focus on measuring and optimizing customer satisfaction concerning administrative processes. Known techniques measuring the satisfaction degree are not capable to identify the criteria that have the most impact on the overall satisfaction. Furthermore, traditional statistical methods can't tell how changes in the overall satisfaction will affect the importance of the relevant criteria. To overcome these deficiencies we will introduce a novel probabilistic approach to optimize customer satisfaction, which is based on Bayesian Belief Networks - a sound probabilistic formalism for uncertain reasoning. These networks will be used to build up a model for the assessment, monitoring, simulation and optimization of customer satisfaction for administrative processes. The model will be built up hierarchically with the overall satisfaction modeled as root node, classes of criteria as nodes on the next level in the hierarchy and finally the satisfaction criteria themselves on the third level. After the definition of the qualitative model, the network has to be quantified. This will be done via prior knowledge on one hand and by learning from available data - namely the questionnaires - on the other hand.