Our website is currently undergoing maintenance. Article records and full texts remain available. Submissions can be made through the Editorial Manager.

Daily Ceramic Product Design Method Based on Optimized BP Neural Network Model and Kano Model

Y.L. Zhang

Journal of Ceramic Science and Technology, Vol. 15, No. 1 (2024), pp. 7-20

DOI: 10.4416/JCST2023-00015

Download full article (PDF)

Abstract

The demand for and design of daily ceramic products design tools is one of the application With the continuous development of society, the demand for daily ceramic product design tools is increasing, among which Back Propagation neural network (BPNN) and Kano model are widely used in the design field. This paper aims to optimize these two models and explore the methods of daily ceramic modeling design. Firstly, this paper summarizes the data and technology of daily ceramic design. Secondly, the demand of ceramic products is reclassified by Kano model, and the BPNN and Kano model are defined and optimized. Finally, the feasibility and effectiveness of this method are verified by simulation experiments. The results show that: (1) Consumers are more interested in flower and plant patterns, accurate pattern printing and custom pattern selection. In addition, consumer sensitive demand items are divided into 9 high sensitive demand items and 6 low sensitive demand items. (2) The error value ranges from 0.02 to 0.06 without using the order of magnitude normalization, but the error value ranges from 0.01 to 0.04 after using the order of magnitude normalization. This means that the performance of the model has been significantly improved by using order of magnitude normalization. (3) The perceptual evaluation values of the first group of four representative samples of ceramic products are 5.23, 4.89, 3.76 and 3.75 respectively. The perceptual evaluation values of the second group of descriptive words are 2.98, 5.76, 5.04 and 3.16 respectively. The perceptual evaluation values of the third group of descriptive words are 3.25, 6.64, 5.83 and 5.05 respectively. Different ceramic products have different evaluation values. Overall, the evaluation value of the first group is higher, indicating that the first group is more popular. The research in this paper provides some reference value for improving the quality of daily ceramic products.

Keywords

BP neural network, Kano model, daily ceramic product, data mining, requirement, satisfaction

View this issue · Article & DOI index