Strategic concept formation of consumer goods based on knowledge acquisition from questionnaire data

Product concept formation, which occurs in the early stage of product development, is critical to the successful development of a new product or to the suitable improvement of a current product. We propose a method for computer aided strategic concept formation based on knowledge acquisition from questionnaire data. The product concept should be developed based on consumers needs that are usually embedded in consumer survey data, and moreover the foresight of a domain expert should be added to it using a domain strategy. To meet these requirements, our proposed method adopts the 3-phased interactive computing process, where (1) evolutionary algorithms such as simulated breeding (including genetic algorithms) and inductive learning techniques are used to extract one type of strategic knowledge, (2) a technique like a simple expert system is used for the other type of strategic knowledge and (3) a reinforcement learning technique is employed to converge thoughts using two types of strategic knowledge. It enables the user to generate a creative and well-grounded product concept based on marketing strategy, while also stimulating user's creativity. The system called BICSS was developed based on this method. The proposed method has been qualitatively validated by a case study.