A study on parametric design of ceramic liquor bottles by combining grey relational analysis and particle swarm optimization support vector regression.

Nowadays, with the vigorous development of cultural and creative industries, the design of traditional cultural products often depends on the designer's personal artistic accomplishment and aesthetic judgment. However, the cognitive difference between designers and consumers will lead to the fact that cultural and creative products cannot be recognized by consumers when they are introduced to the market. In order to seek the design method with intelligence assistance and better design cultural products that can be in line with users' emotional intentions, a parametric design method of cultural and creative products based on Kansei engineering, integrating grey relational analysis (GRA) and particle swarm optimization-support vector regression (PSO-SVR) is proposed in this paper. It aims at accurately quantifying users' emotional needs and mapping them into product design parameters. Firstly, web crawler and expert interview are used to collect users' emotional needs and product samples; and the morphological deconstruction of products is carried out as well. Secondly, GRA is used to calculate the optimal weight of users' emotional needs and extract the key emotional needs. Finally, PSO-SVR is used to construct the mapping model of key users' emotional needs and morphological characteristics so as to determine the optimal design parameters of ceramic liquor bottles. The result shows that the most critical emotional needs are "tough", "fashionable" and "exquisite". In addition, the evaluation results of the optimized design scheme have a high degree of fit with the prediction results of the mapping model, which verifies the feasibility of the proposed method. Compared with traditional cultural and creative products, this research can better balance the subjective aesthetics of designers and the objective needs of consumers, which can help enterprises build a communication bridge between users and designers, accurately predict the emotional needs of users, reduce the risk of product development failure and improve the market competitiveness of enterprises.