| With the rapid development of social economy and the significant improvement of living consumption level,customers’ highly personalized needs are increasingly prominent,and more and more enterprises invest more resources to focus on personalized customization.Different market environments and different development stages reflect different market competition focuses.How to quickly respond to customers’ personalized needs and shorten the product design cycle is the key to win the market competition.The key technology of rapid design of mass customization is modular and parametric product family design,and the configuration module and its provided function range are pre-set in advance to meet a certain level of customization.Although this kind of method is efficient,but under the new environment of increasing degree of customization,How to quickly realize personalized customized design has become a research hotspot for the personalized needs beyond the preset range.Therefore,on the basis of previous research on rapid customization design,this paper combined with case-based reasoning technology and carried out corresponding improvement research according to the characteristics of personalized customization industry,so as to improve the efficiency of rapid customization design on the basis of making full use of enterprise history database:First of all,in order to better understand users and conform to the key elements of personalized customization industry,a user portrait model construction method based on the dimension of personalized customization domain is proposed.Combining the characteristics of data technology,this paper makes full use of enterprise historical transaction data,marketing and sales data and user demand data.The modeling processes such as data collection and processing,tag system design and hierarchical structure are analyzed respectively,and the user portrait model is constructed based on the characteristics of the personalized customization domain.At the same time,the user portrait mapping is constructed into a multi-level user portrait library of modules and components by using the modular parameterized product family structure under the personalized customization technology,which provides a model basis for the subsequent fusion of user portrait clustering to improve the case retrieval efficiency.Then,based on modular and parameterized combination of CBR reasoning product configuration model for personalized customization needs to solve the problem of case retrieval efficiency and put forward based on the constructed model of a single field of personalized user portrait using improved K-means clustering technology to obtain similar user community,building group portrait and with the corresponding case library mapping,thus delimiting the molecular case base,reducing the case base and improving the efficiency of CBR retrieval.This method has good robustness and is more suitable for the case clustering process in the case reasoning method oriented to personalized customization.Based on user portrait user similarity calculation,considering the instance of personalized features and user preferences similarity calculation model for personalized characteristic and clusters of Angle measurement,solving a case of the similarity is more reasonable,more in line with the actual customer needs and technical personnel to meet the personalized custom orders for case change reference case;On this basis,an example optimal retrieval strategy and hierarchical structure are proposed,and a customized product configuration model is constructed by module parameterization combined with improved CBR.Finally,the model is applied to the case of customized elevator configuration,and the effectiveness of the model is verified by comparison. |
