Research On Multiple Imagery-oriented Modeling Design Of Liquor Bottle Based On Kansei Engineering

The increasingly fierce of market competition and people's emotional consumption psychology make development models of consumer-oriented product increasingly dominate.Therefore,understanding the emotional needs of consumers is the key to a successful product design.Facing the complex and diverse emotional needs of consumers,a single kansei image often cannot meet the consumers' psychological needs.How to design a product that can meet the needs of consumers' multiple kansei images is an important issue that designers need to study.In recent years,the increasingly younger consumers of liquor have made liquor companies turn their attention to younger groups.In addition,tapping the emotional preferences of younger consumers has become the key to occupy the market and enhanc product competitiveness.As an important part of liquor packaging,container modeling is one of the important factors that affect consumers' purchasing decisions.Kansei engineering uses a combination of qualitative and quantitative methods to establish the connection between perceptual needs and modeling elements,so that modeling design can accurately convey information that can satisfy consumers' preferences,so as to attract consumers and promote sales.Therefore,this article took the modeling design of liquor bottle as an example,and obtained a multiple imagery-oriented modeling design of liquor bottle for young people based on the theory of kansei engineering.The specific work is as follows:(1)Determine the consumer's kansei image of the liquor bottle.First,card sorting was used to obtain 8 image vocabularies as the initial emotional dimension.Second,the morphological analysis method was used to establish a modeling solution library of liquor bottle and obtain the representative design solutions.Then,it ranks the importance of kansei words based on the residual sum of squared differences of the kansei words obtained by procrustes analysis.The top three words were used as the target kansei image.(2)Establish a modeling-kansei evaluation system of liquor bottle.The BP neural network was used to construct the mapping relationship between the morphological elements of the liquor bottle modeling and the target kansei image.By training and verifying the BP neural network,a modeling-kansei evaluation system was established for each of the three target kansei words.(3)Multi-objective optimization of liquor bottle modeling design.In order to obtain the multi-imagery oriented modeling design of liquor bottle,the fitness value was determined by the modeling image evaluation system,and the Strength Pareto Evolutionary Algorithm 2 was used to obtain the set of three overall optimal solutions under the target kansei.(4)Acquired the best modeling design of liquor bottle.Considering the interactivity between the target kansei images and the grayness of the evaluation of the degree of interaction,the gray-fuzzy integral decision model combined by the gray correlation analysis and Choquet fuzzy integral was used to evaluate and compare the Pareto solutions to obtain the best modeling design of liquor bottle.