Product Configuration with Bayesian Network

Abstract

For the satisfaction of individual customer requirements, products with many options are offered in mass customization. However, in the area of ecommerce, the large number of possible product configurations can overwhelm the customer, as he or she is not supported by a human sales expert. To minimize the customer’s overload, this paper examines the combination of a knowledge-based product configurator with an upstream probabilistic recommender system to provide a quick, individual and dynamic initial orientation for the customer. The application of the approach is demonstrated using an example from engineering design.