CN109741120B - Product deep personalized customization method for user requirements - Google Patents
Product deep personalized customization method for user requirements Download PDFInfo
- Publication number
- CN109741120B CN109741120B CN201810925717.5A CN201810925717A CN109741120B CN 109741120 B CN109741120 B CN 109741120B CN 201810925717 A CN201810925717 A CN 201810925717A CN 109741120 B CN109741120 B CN 109741120B
- Authority
- CN
- China
- Prior art keywords
- product
- user
- customization
- customized
- scheme
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a product deep personalized customization method based on user requirements, which is characterized by comprising the following steps: the method maps user requirements into customized requirements by combining with a user individual requirement expression model and quality function expansion, and performs product configuration solving by using instance retrieval and matching; the invention determines a design target set based on user requirements, obtains a customized scheme set by taking a product family as a carrier and combining configuration reasoning and variant design, and finally obtains an optimal scheme by using the goodness evaluation to form a high-efficiency and high-demand-fitting personalized customization process.
Description
Technical Field
The invention relates to the technical field of intelligent manufacturing and design, in particular to a product deep personalized customization method for user requirements.
Background
Personalized customization is a production and sales mode which combines information technology and network technology for product customization, and combines the advantages of customized production and small-batch production. In the process of implementing personalized customization, inaccuracy of user demand analysis, low user satisfaction and feasibility of the customization process become problems to be solved in the field. In the research of personalized customization of products, methods such as user demand acquisition, product innovation design and configuration design are used in each link of personalized customization. In order to obtain user requirements more comprehensively, researchers classify the user requirements, establish a user model based on an ontology, and establish a requirement level model by calculating the weight of each type of requirements to form a product requirement group with different functional attributes.
However, the prior art has the problems of insufficient user participation and low customization degree in the customization process.
Disclosure of Invention
In order to solve the defects of the prior art, the invention aims to provide a product deep personalized customization method for user requirements.
The invention is realized by the following steps:
the product deep personalized customization method based on the user requirements is characterized by comprising the following steps: the method maps user requirements into customized requirements by combining with a user individual requirement expression model and quality function expansion, and performs product configuration solving by using instance retrieval and matching; the user carries out real-time feedback and adjustment on the customization scheme through the visual interface, enhances the personalized customization degree of the product by combining with the variant design, and finally provides the optimal scheme for the user through goodness evaluation, thereby realizing the customization with high participation degree and high satisfaction degree of the user.
Preferably, the method comprises the following process steps:
step one, analyzing user requirements; obtaining a design target set through user requirement classification, user requirement weight acquisition and unified expression model;
step two, the personalized customization design process of the product: comprises a customized method set and a scheme product;
the customization method set is that a gene product required in the personalized customization process of the product is determined by establishing a product family module, the gene product comprises a universal part and a customized part obtained by deformation design, and then a customization scheme set is obtained by retrieval and matching according to an example and recombination;
considering two factors of the structure and the function of the product, respectively adopting interval intuitive fuzzy numbers and intuitive fuzzy numbers to carry out quantitative processing on the function and the structure of the product according to the characteristics of the function and the structure index of the product, and formulating an evaluation standard of the correlation between the structure and the function;
the scheme product is obtained by customizing a scheme goodness evaluation method for details and optimizing design;
and step four, finally forming order production through the feedback opinions of the user.
Preferably, the user requirement classification refers to the requirements of different users and the customization characteristics of specific products, and the user requirements are classified into the following four categories and defined:
(1) Defining functional requirements FR of a user, including the use requirements FR of the product u Need for model formation FR a Price demand FR p Scene requirement FR e (ii) a The functional requirement description set is expressed as: FR = (FR) u ,FR a ,FR p ,FR e )；
(2) Defining the technical requirement TN of the user, wherein the technical requirement TN of the product is included s And performance parameter TN p And the indexes are equal, and the technical requirement description set can be expressed as TN = (TN) s ,TN p )；
(3) Defining the structure requirement SN of the user, wherein the structure parameter SN of the product is included p ；
(4) Defining the composite requirements CN of users, wherein the requirements are various combinations of the three requirements FR, TN and SN;
the user demand weight acquisition combines a rough set theory and a primitive theory, and gives definition by adopting a primitive concept and an expression habit thereof: given a conditional attribute set G and a decision attribute set K, the quad I = (U, C, V, f) is the decision primitive, if C = GUK, and
i = (U, C, V, f) is called the decision table.
Decision primitive I = (U, C, V, f), wherein U = { U = 1 ,u 2 L u |U| Is the domain of discourse, and
c is a finite non-empty set of attributes, C = { C 1 ,c 2 L c |n| Is the set of all attributes;
is attribute C j Belongs to the value range of C; UXC → V is an information function, which is a single mapping, and makes the attribute C of any object in the domain U have a unique information value, namely
The rough set weight value acquisition basic steps are as follows:
(1) Determining an evaluation index attribute set G and a comprehensive evaluation value K, and calculating C j Degree of dependence of
Represents decision attributes K and C j Degree of dependency between them, see formula (1)
Wherein card (L) represents the cardinality of the collection;
(2) Find the index C j Weight of (2)
(3) The weights are normalized, see equation (3).
The weights obtained by the rough set method reflect the more rational expected values of the users;
the uniform expression model is characterized in that user requirements are converted through QFD, the user is assisted to provide specific description for deep personalized customization of products, and the functional principle, the technical route, the product structure and the characteristic expression of the products are in a corresponding and reversible relation with each other according to the product family model generation theory; specific product customization conditions can be influenced by customization requirements, part of various customization conditions are converted into design targets after QFD conversion, the design targets are mapped to a functional principle layer, a technical route layer, a product structure layer or a feature expression layer, and a configuration stage in the personalized customization process of the product is formed according to the mapping relation corresponding to different user requirement types; and the other part is converted into constraint conditions of a functional principle layer, a technical route layer, a product structure layer or a characteristic expression layer in a product configuration stage, and a uniform model is adopted to express the mapping relation.
Preferably, the product family is based on user demand analysis, the product family needs to accurately convert the user demand into the description related to the function principle and the geometric structure of the customized product, determine the accurate gene product, then select reasonable parts step by using a configuration design according to the specific parameter requirements of the customized product, search for a universal part meeting the user demand and a customized part to be modified, and finally constitute a new customized scheme through recombination, which is specific:
defining a product family model as PM, recording the product function principle as PF, recording the product structure as PS, and setting the product family model PM with the same user group requirements n Then the product family model can be expressed as PM n ＝(PF n ,PS n ) (ii) a In the product family design process, a technical route describes a technical solution of a product family function in the design process, the product family design is recorded as PD, the technical route is recorded as PT, and the product family design process can be expressed as PD = (PT, PM); wherein the product functional principle PF describes the functional architecture and functional class of the product family; the technical route PT describes a technical solution based on functional decomposition in a product family design system; the product structure PS describes the physical elements of the product family, and the three parts are closely related.
Preferably, the functional principle PF describes the functional architecture and class of the product family. In functional theory, a product family and its product variables can be represented by functional requirement FR) and functional requirement variable a. Where a is an example of a functional requirement, one functional requirement FR may correspond to multiple a; in the product custom design, the decomposition process of the FR is a process of decomposing from an abstract level to a thinner functional unit, and the functional characteristics of a product family are described according to the process;
the defined technical route PT describes a technical solution of a product family; the technical solutions can be expressed as design parameters DP and variables V, each technical solution can be expressed as a plurality of design parameters DP, and if the technical solutions meet the requirements of the DP, the technical route provides technical support for changing the customized solutions into real objects;
the product structure PS of the product family describes components and the relation between the components, and consists of general components and fixed components; defining a common part UP as a common part or element that each product variable in a product family has, a common part set can be represented as UP = (UP) 1 ,UP 2 ,L,UP n ) These shared elements may be structures or components common to the product that determine the basic nature and functional location of the product; defining a customized product MP is a basic element which makes one product different from other products and promotes the diversification of product families; in the product customization process, a customized part MP is a gene part GP which is suitable for a personalized customization scheme and cannot be found in a product family by a user; the customized part needs to be subjected to variant design on the gene components of the product family by a user or a designer so as to meet the personalized customization requirements of the user.
Preferably, the deformed design pointer carries out deformed design on similar products and parts obtained by example retrieval, and four deformed design rules of pasting, zooming, stripping and exchanging are carried out to form a product customized part MP meeting the requirements of a user and the product customized part MP is added into the example library.
Preferably, the search and the fear of matching of the examples mean that the design of the product family can be regarded as the configuration design and the modification based on the gene product, and the product family is modified on the basis of the existing similar examples according to the design requirement; the selection matching calculation process for similar examples is as follows:
1) Let P c ＝(p 1 ,p 2 ,L p n ) A set of values representing attributes of a user-customized product;
2) Let P j ＝(p j1 ,p j2 ,p j3 L p jn ) A set of values representing the presence of product attributes within the system;
3) Considering the maximum expected value of the customized product as a standard vector P j = (1,1,L 1), mixing P j ＝(p j1 ,p j2 ,p j3 L p jn ) Performing normalization processing by using a linear interpolation method;
4) Calculating P c ＝(p 1 ,p 2 ,L p n ) Weight w = (w) corresponding to each product attribute in the product 1 ,w 2 ,L w n ) Satisfy the following requirements
5) Setting a threshold lower limit lambda value, primarily selecting a product customization scheme by using a projection method, and selecting examples with similarity greater than lambda to form an alternative scheme set;
the specific example retrieval matching process is that a user determines initial conditions of a customized product, and the system retrieves an example base based on consistency expression of user requirements. Firstly, product matching is carried out, whether products meeting the requirements of a user exist or not is searched, the search result comprises the same products and similar examples, if the same products exist, the search result is directly communicated with the user to determine whether a target product is generated or not, and if the similar examples exist, the target product is obtained by searching gene parts and carrying out variant design; if the same or similar products do not exist, the product is decomposed to carry out the next step of component matching, the same is to search whether the same or similar components exist or not, the target component to be produced is obtained to be used as a product customized part MP, and finally, the customized scheme set is obtained by recombination. And obtaining an optimal scheme for the user to select through the goodness evaluation, obtaining a product scheme customized by the user if the user is satisfied, otherwise, carrying out co-operation with the user, resetting the initial conditions, and then carrying out retrieval and matching.
Preferably, the product structure reorganization is that after a general part UP required by customization is determined through retrieval and matching of an example, a customized part MP after modification design is combined, the method is suitable on the premise of meeting a technical route DP, the shawl customization is taken as an example, the user function requirement is set to be a shawl with a national style, if the same example is not found through the example retrieval and matching, the similar example is found to be used as a gene part GP and is modified, a modified example which is satisfied by a user is obtained to be used as the customized part MP of the whole product, then the parts are reorganized to form the customized product required by the user, and the deep customization process of the product is completed.
Preferably, the customized scheme goodness evaluation method sequentially comprises three processes of determining evaluation indexes and weights, primary convergence and goodness evaluation.
Preferably, the determining the evaluation index and the weight are: for the custom scheme set S = { B } expansion goodness evaluation, the custom scheme evaluation target feature set is represented as c = (c) 1 ,c 2 ,L,c m ) The corresponding magnitude is denoted v = (v) 1 ,v 2 ,L,v m ) Wherein
V(c i ) A magnitude domain that is an evaluation objective;
determining corresponding evaluation target feature set c = (c) by using rough set method 1 ,c 2 ,L,c m ) Weight coefficient of (a) = (a) 1 ，α 2 ，Lα m ) The index which must be satisfied is expressed by the index Λ, i.e. α r If Λ, then there is
Construction of evaluation index set H = { H = { (H) } 1 ,H 2 ,L,H m In which H is i ＝(c i ，v i ),i＝1，2，L，m；
The initial convergence is as follows: using the evaluation target alpha that has to be met r And (5) carrying out primary convergence on the customization scheme set S = { B }, eliminating unsatisfied customization schemes, and constructing the customization scheme set
Then a single solution in the customization solution set is marked as O pj Then product customization scheme set S 1 ,S 2 ,L,S n The number of schemes in (1) is recorded as n 1 ,n 2 ,L n n ；
The goodness evaluation is a customized solution O pj Is recorded as C (O) pj ) The following conditions exist in the goodness calculation:
1) And the goodness is calculated by using the comprehensive correlation function, and the method is suitable for multi-target evaluation of most product customization schemes.
2) Goodness C (O) pj ) Taking the minimum of the correlation function, i.e.
Each evaluation characteristic in the customized scheme evaluation process is required to meet the requirement and has no weight score;
3) Goodness C (O) pj ) Taking the maximum value of the correlation function, i.e.
The evaluation process of the customized scheme is expressed as long as one evaluation characteristic meets the requirement, and the evaluation process has no weight score;
aiming at the goodness calculation of the customized scheme, introducing variance and related concepts to calculate the stability of the overall goodness of the customized scheme set;
goodness of individual solutions in a customized solution set
For discrete random variables, the standard goodness of the maximum individual scheme in the customized scheme set is taken as E (Y), namely
The variance D (Y) = E { [ Y-E (Y) with respect to goodness Y] 2 }; note the book
The standard deviation of goodness is known from the concept of variance and the meaning of semantic evaluation of the product, D (Y) or
A customization scheme S can be described p Regarding the degree of deviation of the goodness of the evaluation index from the standard goodness E (Y), S in the customized recipe set S is reflected by (E (Y), σ (Y)) as a whole p And the goodness and stability of a certain evaluation index set.
By adopting the technical scheme, compared with the prior art, the invention mainly provides a product deep personalized customization method based on user requirements aiming at the problems of insufficient user participation and low customization degree in the customization process, establishes a user personalized requirement unified expression model, maps the user requirements by using quality function expansion (QFD) on the basis of fully conforming to the user requirements to obtain design requirements, maps the design requirements and the product family model by combining the existing product family, performs configuration solution through example retrieval and matching, recombines the configured results as the customized parts obtained by the customized parts and the variants through a visual interface to form a customized scheme set, and selects an optimal scheme by using goodness evaluation to provide the optimal scheme for the user. And (3) carrying out example verification by taking the deep personalized customization of national cape shoulder products as an example, and verifying the effectiveness of the method.
Drawings
FIG. 1 is a schematic diagram of analysis of a product deep personalized customization process based on user demand analysis in the invention 1;
FIG. 2 is a schematic diagram of a unified expression model of a user requirement hierarchy in the present invention;
FIG. 3 is a schematic diagram of a product family mapping according to the present invention;
FIG. 4 is a diagram of an exemplary search matching process of the present invention;
FIG. 5 is a schematic diagram of four design rules according to the present invention;
FIG. 6 is an exploded view of the morphological information features of the product of the present invention;
FIG. 7 is a flow chart of a product personalization customization mode in the present invention;
FIG. 8 is a schematic diagram of user requirement classification according to the present invention;
FIG. 9 is a schematic diagram of the customization goals and constraints of the present invention;
FIG. 10 is a library of example product family cases in accordance with the present invention;
FIG. 11 is the gene library of the modeling elements of national products in the present invention;
FIG. 12 is a schematic diagram of an example search flow in the present invention;
FIG. 13 is a schematic diagram of an example search in the present invention;
FIG. 14 is a flow chart illustrating the basic function setup of the customization system of the present invention;
FIG. 15 is a schematic diagram of a custom-made product scheme formation process in accordance with the present invention;
FIG. 16 (a) is a schematic diagram showing the selection of gene elements for product modeling in the present invention;
FIG. 16 (b) is a schematic diagram of a selection variation and a composition rule in the present invention;
FIG. 16 (c) is a preview customization scenario molding diagram in accordance with the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments; all other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention comprises the following steps:
design process analysis
The core idea of personalized customization of products is to reasonably and efficiently utilize and manage the existing product resources, and quickly recombine the existing product resources to form diversified products so as to meet the user requirements. The design process can be regarded as the solving process of a plurality of design problems, and the solving process of the problems is effectively organized and managed according to a certain design program to form a complete design situation set D = { D = { D = 1 ,D 2 ,L D n }。
From the analysis of the design process of the product, the design knowledge comprises user requirement knowledge, product design process knowledge and production and manufacturing knowledge, but the individual customization of the product is the product design of small batch and multiple varieties, and not only the innovation of the product but also the efficiency of the design, the feasibility of the designed product, the participation of the user and the like are considered in the design process. The personalized customization system needs to determine a design target set based on user requirements, obtain a customization scheme set by taking a product family as a carrier and combining configuration reasoning and variant design, and finally obtain an optimal scheme by using goodness evaluation to form a personalized customization process with high efficiency and high demand conformity; FIG. 1 is an analysis diagram of a product deep personalization process based on user demand analysis.
Unified expression model for user personalized demand information
1. User demand classification
Personalized product customization is a production model that aims to meet the user customization needs to the greatest extent [14], and since users play a critical role in the custom design and manufacturing process of products, the application introduces a user requirement layer concept in the product customization scheme generation process.
According to the method, the user requirements are divided into the following four types and defined according to the requirement types of different users and the customization characteristics of specific products:
(1) Defining functional requirements FR of a user, including the use requirements FR of the product u Need for model formation FR a Price demand FR p Scene requirement FR e . The functional requirement description set is expressed as: FR = (FR) u ,FR a ,FR p ,FR e )
(2) Defining the technical requirement TN of the user, including the technical index TN of the product s And performance parameter TN p And the indexes are equal, and the technical requirement description set can be expressed as TN = (TN) s ,TN p ).
(3) Defining the structure requirement SN of the user, wherein the structure parameter SN of the product is included p 。
(4) Defining the composite requirements CN of the users, wherein the requirements are various combinations of the three types of requirements FR, TN and SN.
2. User demand weight acquisition
The personalized requirements of the user are usually obtained by directly giving the weight to the user. However, it is difficult to objectively give a weight value when a user is used as an evaluation subject, and thus the user has ambiguity. The rough set method is used for solving the problem of obtaining the evaluation index weight in the group multi-attribute decision process, can effectively express the weight expectation of the user, and solves the problems of ambiguity and subjectivity interference in the user weight assignment process.
The method combines the rough set theory and the element theory, adopts the element concept and the expression habit thereof, and omits unnecessary definitions and explanations.
The definition given is: given a conditional attribute set G and a decision attribute set K, the quad I = (U, C, V, f) is the decision primitive, if C = GUK, and
i = (U, C, V, f) is called the decision table.
Decision primitive I = (U, C, V, f), where U = { U = 1 ,u 2 L u |U| Is domain of discourse, an
C is a finite set of non-null attributes, C = { C 1 ,c 2 L c |n| Is the set of all attributes;
is attribute C j Belongs to the value range of C; UXC → V is an information function, which is a single mapping, and makes the attribute C of any object in the domain U have a unique information value, namely
The rough set weight value acquisition basic steps are as follows:
(1) Determining an evaluation index attribute set G and a comprehensive evaluation value K, and calculating C j Degree of dependence of
Representing decision attributes K and C j See formula (1).
In which card (L) represents the cardinality of the set.
(2) Find the index C j See equation (2).
(3) The weight is normalized, see equation (3).
The weights obtained by the rough set method reflect more rational expected values of the user.
3. Unified expression model
The user requirements are converted through the QFD, and the specific description of the product deep personalized customization is provided for the user. The functional principle, the technical route, the product structure and the characteristic expression of the product are mutually corresponding and reversible relations according to the product family model generation theory.
The product customization conditions can be influenced by customization requirements, part of various customization conditions are converted into design targets after QFD conversion, the design targets are mapped to a functional principle layer, a technical route layer, a product structure layer or a feature expression layer, and a configuration stage in the personalized product customization process is formed according to the mapping relation corresponding to different user requirement types; the other part is converted into constraint conditions of a functional principle layer, a technical route layer, a product structure layer or a characteristic expression layer in a product configuration stage, a unified model is adopted for expressing the mapping relation, and fig. 2 is a unified expression model of a user requirement hierarchical structure.
Product family design
The product family design is a product aggregation sharing core general technology, is an effective method for developing diversified products by using limited resources, and based on user demand analysis, the product family needs to accurately convert the user demands into the description related to the function principle and the geometric structure of the customized products, determine accurate gene products, then select reasonable parts step by using configuration design according to the specific parameter requirements of the customized products, search general parts meeting the user demands and customized parts to be modified, and finally form a new customization scheme through recombination. Therefore, in the product family design, the determination of accurate gene products is used as the basic work of the product family design.
The gene product required by the product deep personalized customization design process is determined by establishing a product family model.
The product family model covers the functional principle and the structural information of a family of products meeting the same customization conditions, and is the expression of the design requirement knowledge of the product family. Defining a product family model as PM, recording the product function principle as PF, recording the product structure as PS, and setting the product family model PM with the same user group requirements n Then the product family model can be expressed as PM n ＝(PF n ,PS n ). In the product family design process, the technical route describes the technical solutions of the product family functions in the design process, the product family design is recorded as PD, and the technical route is recorded as PT, so that the product family design process can be expressed as PD = (PT, PM). Wherein the product function principle PF describes the functional architecture and functional class of the product family; the technical route PT describes a technical solution based on functional decomposition in a product family design system; the product structure PS describes the physical element composition of the product family. There is a close relationship between the three parts, and fig. 3 is a product family map.
The functional principle, the technical route and the product structure are mapped to each other to form the design process, the manufacturing process and the sale process of the product. In the whole process, functional requirements in the market are defined and classified to obtain technical routes and structural parameters of products, the physical structures of the products are realized through manufacturing, and the products are finally transmitted to users through sale.
1. Principle of function
Functional principle PF describes the functional architecture and class of a product family. Within the functional principle, a product family and its product variables can be represented by Functional Requirements (FR) and functional requirements variables (a). Where A is an example of a functional requirement, one functional requirement FR may correspond to multiple A's. In the product custom design, the decomposition process of the FR is a process of decomposing from an abstract level to a finer functional unit, and thus, the functional characteristics of the product family are described.
2. Technical route
The definition of the technical route PT describes the technical solution of the product family. It describes the technical application requirements involved in the customization process to ensure the feasibility of the customization scheme.
The technical solutions can be expressed as design parameters DP and variables V, each technical solution can be expressed as a number of design parameters DP, and the technical solutions are feasible to some extent if they meet the requirements of DP. The technical route provides technical support for changing the customized scheme into a real object.
3. Product structure
Defining a common part UP as a common part or element that each product variable in a product family has, a common part set can be represented as UP = (UP) 1 ,UP 2 ,L,UP n ) These shared elements may be structures or components common to the product that determine the basic nature and functional orientation of the product.
Defining the customized piece MP is a basic element for making one product different from other products, and promotes the diversification of product families. In the product customization process, the customization part MP is a gene part GP which the user cannot find suitable for the personalized customization scheme in the product family. The customized part needs to be subjected to variant design on the genetic components of the product family by a user or a designer so as to meet the personalized customization requirements of the user.
Custom product generation process
1. Retrieval and matching of instances
The product family design can be regarded as the configuration design and the modification based on the gene product, which is modified based on the similar examples according to the design requirements. The selection matching calculation process for similar examples is as follows:
1) Let P c ＝(p 1 ,p 2 ,L p n ) A set of values representing attributes of a user-customized product;
2) Let P j ＝(p j1 ,p j2 ,p j3 L p jn ) A set of values representing the presence of product attributes within the system;
3) Considering the maximum expected value of the customized product as a standard vector P j = (1,1,L 1), mixing P j ＝(p j1 ,p j2 ,p j3 L p jn ) Performing normalization processing by using a linear interpolation method;
4) Calculating P c ＝(p 1 ,p 2 ,L p n ) Weight w = (w) corresponding to each product attribute in the product 1 ,w 2 ,L w n ) Satisfy the following requirements
5) Setting a threshold lower limit lambda value, primarily selecting a product customization scheme by using a projection method, and selecting examples with similarity greater than lambda to form an alternative scheme set;
FIG. 4 is an example process of retrieving matches. The user determines initial conditions for customizing the product and the system retrieves the instance base based on a consistent expression of the user's needs. Firstly, product matching is carried out, whether products meeting the requirements of a user exist or not is searched, the search result comprises the same products and similar examples, if the same products exist, the search result is directly communicated with the user to determine whether a target product is generated or not, and if the similar examples exist, the target product is obtained by searching gene parts and carrying out variant design; if the same or similar products do not exist, the product is decomposed to carry out the next step of component matching, the same is to search whether the same or similar components exist or not, the target component to be produced is obtained to be used as a product customized part MP, and finally, the customized scheme set is obtained by recombination. And obtaining an optimal scheme for the user to select through the goodness evaluation, obtaining a product scheme customized by the user if the user is satisfied, and otherwise negotiating with the user, resetting initial conditions and then retrieving and matching.
2. Modified design
And performing variant design on similar products and parts obtained by example retrieval, forming a product customized part MP meeting the requirements of a user through four variant design rules of pasting, scaling, stripping and exchanging, and adding the product customized part MP into an example library. Fig. 5 is a diagram of four variant design rules.
Taking national cape custom as an example, on the basis of determining the function, gene elements needing custom can be obtained by searching an example library and performing variant design to obtain a custom piece MP so as to meet the requirement of personalized modeling under the custom requirement of a user.
3. Product structure reorganization based on user demand
The method is suitable on the premise of meeting a technical route DP, takes shawl customization as an example, sets user function requirements as a shawl with a national style, finds out the same example through example retrieval and matching, obtains a variant example which is satisfied by a user as the customized part MP of the whole product by searching for the similar example as a gene part GP and carrying out variant, then carries out recombination of parts to form the customized product required by the user, and finishes the deep customization process of the product.
The product innovation design based on the combination principle is applied to the recombination design of the product structure, and the basic information characteristics of the product are obtained through analysis. Such as the shape, function, color, etc. of the product. The method takes the customization of the national style shawl as an example, decomposes the modeling of the shawl, and obtains the modeling characteristic of the shawl as a frame pattern f e Main body pattern f m And a decorative pattern f i 。
Wherein the border pattern f e And can be formed according to different compositionsElement is decomposed into
A considerable amount of modeling information features can be obtained by such extrapolation, and FIG. 6 is an exploded view of product form information features.
The basic modeling feature array obtained by decomposition is as follows
Will basic model feature f e ,f m ,f i Permutation and combination are carried out to obtain a model feature matrix shown in the table 1.
TABLE 1 product modeling characteristics matrix
Then it shares
And (4) a combination mode. Wherein b, h and q are selected element numbers, and x, y and z are respectively f e 、f m 、f i Total number of elements in (1).
Customized scheme goodness evaluation method
1. Determining evaluation index and weight
The goodness evaluation is expanded against the custom solution set S = { B }. Expressing the custom scheme evaluation target feature set as c = (c) 1 ,c 2 ,L,c m ) The corresponding magnitude is denoted as v = (v) 1 ,v 2 ,L,v m ) Wherein
V(c i ) Is the magnitude domain of the evaluation objective.
Determining corresponding evaluation target feature set c = (c) by using rough set method 1 ,c 2 ,L,c m ) Weight coefficient of (a) = (a) 1 ，α 2 ，Lα m ) The index which must be satisfied is expressed by the index Λ, i.e. α r If Λ, then there is
Construction of evaluation index set H = { H = } 1 ,H 2 ,L,H m In which H is i ＝(c i ，v i ),i＝1，2，L，m。
2. First time convergence
Using the evaluation target alpha that has to be met r And (5) carrying out primary convergence on the customization scheme set S = { B }, eliminating unsatisfied customization schemes, and constructing a customization scheme set
Then a single solution in the custom solution set is denoted as O pj Then product customization scheme set S 1 ,S 2 ,L,S n The number of schemes in (1) is sequentially marked as n 1 ,n 2 ,L n n 。
3. Evaluation of goodness
Customization scheme O pj Is recorded as C (O) pj ) The goodness of the calculation is as follows:
1) And the goodness is calculated by using the comprehensive correlation function, and the method is suitable for multi-target evaluation of most product customization schemes.
2) Goodness C (O) pj ) Taking the minimum of the correlation function, i.e.
Showing that each evaluation characteristic in the custom scheme evaluation process needs to meet the requirement and has no weight score;
3) Goodness C (O) pj ) Taking the maximum value of the correlation function, i.e.
Shows that only one evaluation characteristic meets the requirement in the evaluation process of the customized scheme without the weight,
and aiming at the goodness calculation of the customized scheme, introducing variance and related concepts to calculate the stability of the overall goodness of the customized scheme set.
Goodness of individual solutions in a customized solution set
For discrete random variables, the standard optimal degree of the individual schemes in the customized scheme set with the highest optimal degree is taken as E (Y), that is
The variance D (Y) = E { [ Y-E (Y) with respect to goodness Y] 2 }。
Note the book
The standard deviation of goodness is known from the concept of variance and the meaning of semantic evaluation of the product, D (Y) or
A customization scheme S can be described p The degree of deviation of the goodness of the evaluation index from the standard goodness E (Y) as a whole. Reflecting S in the customization scheme set S by (E (Y), sigma (Y)) p Goodness and stability for a certain set of evaluation metrics.
In summary, this goodness evaluation is applied to goodness evaluation of a series of customized solution sets, starting with S = { B }, and using solution clustering to form a customized solution set S = { B } = { S }, where S = { B } is a measure of the goodness of the order 1 ,S 2 ,L,S n And in a custom scheme S p Sub-scheme of
For evaluating individuals, a relevance function is constructed, and goodness calculation and description of goodness standard deviation are applied to reflect S in the customized scheme set S p The comprehensive goodness and stability of a certain evaluation index set. Comprehensive convergence of the customized scheme set is favorably realized, and the selection goodness is betterAnd the good customization scheme is selected by the user so as to implement the customized design and realize the commercialization.
Product deep personalized customization instance verification based on product family
The embodiment takes national style shoulder personalized customization as an application object to verify the effectiveness of the method provided by the application, and the personalized customization of the product with high demand conformity is realized by acquiring and analyzing the user demand in the customization process, mapping the user demand to the customization scheme step by step, combining example retrieval matching and real-time variant design, and fig. 7 is a product personalized customization mode.
1. User demand analysis
In actual product personalized customization, in order to obtain historical data of a user, a user ID, registered account information and the like are usually used as basic bases for classifying the user, a product personalized customization system prototype interface (only adopting the prior art) is designed according to the basic data, the product personalized customization system prototype interface is used for obtaining and basic classifying user basic information, after the user inputs requirements and selects customization conditions, the system converts the user requirements into customization requirements through a unified expression model, and the prior art is used for setting a customization requirement input interface which is used as a direct mode for obtaining the user requirements and is a basic level interface for the user to input the requirements.
The requirements input by the user are analyzed, the individual requirements of the user can be met, the requirement of the user is assumed to be a national cape, and the requirements of the user are divided according to the customization characteristics of the product, namely, the functional requirements FR, the technical requirements TN and the structural requirements SN. Fig. 8 is a classification diagram of the user requirements under the requirement.
The functional requirement FR, the technical requirement TN and the structural requirement SN of the user are converted through a unified expression model to obtain a customized target and a customized constraint condition, the customized target and the constraint condition are customized, and a characteristic expression layer formed by a functional principle layer, a technical layer and a structural layer provides design guidance for the deep personalized customization of the national wind cape products.
2. Establishment of product family case base
The establishment of the national wind cape product family case library and the national wind product modeling element gene library is the premise for the national wind cape customization. Decomposing the structure of the national cape shoulder product to obtain the basic modeling characteristics of a frame pattern, a main body pattern and a modification pattern, and performing example retrieval and matching according to the basic modeling characteristics. Taking the requirement of national style shoulder customization as an example, the wax printing process product in the national style is taken as a product case base, the product case base is shown as figure 10, and the national product modeling element gene base is shown as figure 11.
3. Retrieval and matching of instances
The customized target and the constraint condition are obtained after the user requirements are analyzed, the customized target and the constraint condition are used as specific design requirements for instance retrieval, the user performs real-time feedback adjustment on retrieval and matching results, the instance retrieval and matching are performed according to the customized target and the constraint condition by taking the user requirements of 'national wind shawl' as an example, and an instance retrieval process is shown in FIG. 12.
Based on the example retrieval process and the example retrieval mode, an example retrieval and matching prototype system interface is designed, a user previews a retrieval result through a visual interface and selects and adjusts the retrieval result, and therefore user participation is improved, and the example retrieval interface is shown in the figure 13.
The customized elements of the product type are obtained by analyzing the product family under the requirement of the national wind shawl, and comprise size customization, function selection and form attribute customization, and the product form attribute configuration selection is provided for the user according to the characteristics of the product type, wherein the customized elements mainly comprise modeling customization, material customization, color customization, texture customization and decorative pattern customization, and the product appearance is pre-browsed through a customized system prototype interface, and fig. 14 is a customized system basic function setting diagram.
4. Modified design
Aiming at similar ethnic wind shoulder products and the constituent elements thereof obtained by example retrieval, a user selects and carries out real-time variant design on the products through a visual interface, and the products are customized according with the requirements of the user through four variant design rules of pasting, zooming, stripping and exchanging, and fig. 15 is a product customization scheme forming process through the four variant design rules.
Designing a prototype interface of a following customization scheme set forming process according to the variant design method, wherein in the step (a) of figure 16, a product modeling gene element selection interface is provided, a user recombines the selected product modeling gene element as a general part and a product customization part formed by variant design, a product customization scheme set is formed by selecting a composition rule, if the figure 16 (b) is a selection variant and composition rule interface, and finally the system recommends a scheme with higher goodness for the user to select through a visual interface, if the figure 16 (c) is a preview customization scheme interface, if the user is satisfied, the customization scheme is determined and an order is established, and if the user is not satisfied, the customization condition is returned and reset.
The effectiveness of the method provided by the application can be verified through the verification of the examples, and under the user requirement of 'national style shoulder customization', the method provided by the application can ensure the participation degree of the user and improve the personalized customization degree of the product, can better fit the personalized requirement of the user, and obtains higher user satisfaction.
Summary of the invention
According to the method, an expression model unified with user individual demand information is established, user demands are mapped into design demands by using Quality Function Development (QFD), then configuration solution is carried out by combining product family example reasoning and configuration design, a user carries out real-time variant design based on a technical route aiming at a configuration result with low satisfaction degree through a visual interface, a general part obtained by configuration and a customized part obtained by variant design are recombined to form a customized scheme set, an optimal scheme is selected by using goodness evaluation, the participation degree of the user is guaranteed in the whole customization process, and the customization depth and the customization scheme feasibility of products are improved.
The foregoing embodiments are described in order that those skilled in the art can readily understand and utilize the invention and it is readily apparent that various modifications can be made to the embodiments, and thus, the invention is not limited to the above embodiments, and those skilled in the art can make modifications and variations in the methods of the invention without departing from the scope of the invention.
Claims (8)
1. The product deep personalized customization method based on the user requirement is characterized in that: the method maps user requirements into customized requirements by combining with a user individual requirement expression model and quality function expansion, and performs product configuration solving by using instance retrieval and matching; the user feeds back and adjusts the customized scheme in real time through a visual interface, enhances the personalized customization degree of the product by combining with the variant design, and finally provides the optimal scheme for the user through goodness evaluation;
the method comprises the following specific process steps:
step one, analyzing user requirements; obtaining a design target set through user requirement classification, user requirement weight acquisition and a unified expression model in sequence;
step two, the personalized customization design process of the product: including custom method sets and recipe products;
the customization method set is that a product family module is established to determine gene products required in the personalized customization process of products, the gene products comprise common parts and customized parts obtained through deformation design, and then a customization scheme set is obtained through retrieval and matching and recombination according to examples;
considering two factors of the structure and the function of the product, respectively adopting interval intuitive fuzzy numbers and intuitive fuzzy numbers to carry out quantitative processing on the function and the structure of the product according to the characteristics of the function and the structure index of the product, and formulating an evaluation standard of the correlation between the structure and the function;
the scheme product is a scheme product obtained by customizing a scheme goodness evaluation method for detail and optimizing the design;
step four, finally forming order production through the feedback opinions of the user;
the user requirement classification refers to classifying the user requirements into the following four categories and giving definitions according to the requirement types of different users and the customization characteristics of specific products:
(1) Defining function requirements FR of a user, wherein the function requirements FR comprise use requirements FRu, modeling requirements FRa, price requirements FRp and scene requirements FRE of a product; the functional requirement description set is expressed as: FR = (FRu, FRa, FRp, FRe);
(2) Defining the technical requirement TN of a user, wherein the technical requirement TN comprises indexes such as a technical index TNS and a performance parameter TNp of a product, and a technical requirement description set can be expressed as follows: TN = (TNs, TNp);
(3) Defining a structure requirement SN of a user, wherein the structure requirement SN comprises a structure parameter SNp of a product;
(4) Defining the complex requirement CN of the user, wherein the requirement is a plurality of combinations of the three requirements FR, TN and SN;
the user demand weight acquisition combines a rough set theory and a primitive theory, and gives definition by adopting a primitive concept and an expression habit thereof: given a condition attribute set G and a decision attribute set K, a quadruplet I = (U, C, V, f) is a decision primitive, if C = G ≧ K and G ≦ K = e, I = (U, C, V, f) is called as a decision table;
decision primitive I = (U, C, V, f), where U = { U = 1 ，u 2 …u |U| Is the domain of discourse, and
c is a finite set of non-null attributes, C = { C 1 ，c 2 …c |n| Is the set of all attributes;
is attribute C j Belongs to the value range of C; f: UXC → V is an information function that is a single mapping that makes the attribute C of any object in the domain of discourse U have a unique information value, i.e.
The rough set weight value acquisition basic steps are as follows:
(1) Determining an evaluation index attribute set G andoverall evaluation value K, C j Degree of dependence of
Represents decision attributes K and C j Degree of dependency between them, see formula (1)
Where card (\8230;) represents the cardinality of the collection;
(2) Find the index C j Weight of (1), see formula (2)
(3) Weight normalization, see equation (3);
the weights obtained by the rough set method reflect the more rational expected values of the users;
the uniform expression model is characterized in that user requirements are converted through QFD, the user is assisted to provide specific description for deep personalized customization of products, and the functional principle, the technical route, the product structure and the characteristic expression of the products are in a corresponding and reversible relation with each other according to the product family model generation theory; specific product customization conditions can be influenced by customization requirements, part of various customization conditions are converted into design targets after QFD conversion, the design targets are mapped to a functional principle layer, a technical route layer, a product structure layer or a feature expression layer, and a configuration stage in the personalized product customization process is formed according to mapping relations corresponding to different user requirement types; and the other part is converted into constraint conditions of a functional principle layer, a technical route layer, a product structure layer or a characteristic expression layer in a product configuration stage, and a uniform model is adopted to express the mapping relation.
2. The method for deeply customizing a product based on user demands according to claim 1, wherein: the product family is based on user demand analysis, needs of the product family are accurately converted into descriptions related to the function principle and the geometric structure of the customized product, an accurate gene product is determined, then reasonable parts are selected step by means of configuration design according to specific parameter requirements of the customized product, a universal part meeting user demands and a customized part to be modified are searched, and finally a new customization scheme is formed through recombination, and the method is specific:
defining a product family model as PM, a product function principle as PF, a product structure as PS, and setting a product family model PMn required by the same user group, wherein the product family model can be expressed as PMn = (PFn, PSn); in the product family design process, a technical route describes a technical solution of a product family function in the design process, the product family design is recorded as PD, the technical route is recorded as PT, and the product family design process can be expressed as PD = (PT, PM); the functional principle PF of the product describes the functional framework and functional category of the product family; the technical route PT describes a technical solution based on functional decomposition in a product family design system; the product structure PS describes the physical elements of the product family, and the three parts are closely related.
3. The method for deep personalized customization of products based on user's needs as claimed in claim 1, wherein: defining a technical route PT to describe a technical solution of a product family; the technical solutions can be expressed as design parameters DP and variables V, each technical solution can be expressed as a plurality of design parameters DP, and if the technical solutions meet the requirements of the DP, the technical route provides technical support for changing the customized solution into a real object;
the product structure PS of the product family describes components and the relation between the components, and consists of general components and fixed components; defining a common part UP as a common part or element that each product variable in a product family has, a common part set can be represented as UP = (UP) 1 ，UP 2 ，…，UP n ) And these are togetherThe shared element can be a structure or a part common to the product, and determines the basic property and the functional positioning of the product; defining the customized piece MP is a basic element which makes one product different from other products, and promotes the diversification of product families; in the product customization process, the customization part MP is a gene part GP which is suitable for the personalized customization scheme and cannot be found in a product family by a user; the customized part needs to be subjected to variant design on the gene part of the product family by a user or a designer so as to meet the personalized customization requirement of the user.
4. The method for deeply customizing a product based on user requirements as claimed in claim 1, wherein: the variant design pointer carries out variant design on similar products and parts obtained by example retrieval, and a product customized part MP meeting the requirements of a user is formed by four variant design rules of pasting, zooming, stripping and exchanging and is added into an example library.
5. The method for deeply customizing a product based on user requirements as claimed in claim 1, wherein: the retrieval and matching of the examples means that the product family design can be regarded as the configuration design and the modification based on the gene products, and the product family design is modified on the basis of the existing similar examples according to the design requirements; the selection matching calculation process for similar examples is as follows:
1) Let P c ＝(p 1 ，p 2 ，…p n ) A set of values representing attributes of a user-customized product;
2) Let P j ＝(p j1 ，p j2 ，p j3 …p jn ) A set of values representing the presence of product attributes within the system;
3) Consider the maximum expected value of the user-customized product as the standard vector P = (1, \82301; 1), let P be j ＝(p j1 ，p j2 ，p j3 …p jn ) Performing normalization processing by using a linear interpolation method;
4) Calculating P c ＝(p 1 ，p 2 ，…p n ) The weight w = (w) corresponding to each product attribute in the product 1 ，w 2 ，…W n ) Satisfy w x ∈[0，1]，
5) Setting a threshold lower limit lambda value, primarily selecting a product customization scheme by using a projection method, and selecting examples with similarity greater than lambda to form an alternative scheme set;
the specific example retrieval and matching process is that a user determines initial conditions for customizing a product, the system retrieves an example library based on the consistency expression of user requirements, firstly, product matching is carried out, whether a product meeting the user requirements exists is searched, a search result comprises the same product and similar examples, if the same product exists, the system is directly communicated with the user to determine whether a target product is generated, and if the similar example exists, the target product is obtained by searching a gene part through variant design; if the same or similar products do not exist, the product is decomposed to carry out the next step of component matching, the same is that whether the same or similar components exist is searched, the target component to be produced is obtained to be used as a product customization piece MP, recombination is carried out to obtain a customization scheme set, the optimal scheme is obtained through goodness evaluation to be selected by a user, if the user is satisfied, the product scheme customized by the user is obtained, otherwise, negotiation is carried out with the user, initial conditions are reset, and then searching and matching are carried out.
6. The method for deeply customizing a product based on user requirements as claimed in claim 1, wherein: the product structure reorganization means that after a general part UP required by customization is determined through example retrieval and matching, a customized part MP after variant design is combined, the method is suitable on the premise of meeting a technical route PT, the user function requirement is set as a cape with a national style, if the same example is not found through example retrieval and matching, a variant example which is satisfied by a user is obtained as the customized part MP of the whole product by searching for the similar example as a gene part GP and carrying out variant, then the components are reorganized to form the customized product required by the user, and the deep customization process of the product is completed.
7. The method for deeply customizing a product based on user requirements as claimed in claim 1, wherein: the customized scheme goodness evaluation method sequentially comprises three processes of determining evaluation indexes and weights, primary convergence and goodness evaluation.
8. The method for deep personalized customization of products based on user's needs as claimed in claim 7, wherein: the determination of the evaluation index and the weight are as follows: for the custom scheme set S = { B } expansion goodness evaluation, the custom scheme evaluation target feature set is represented as c = (c) 1 ，c 2 ，…，c m ) The corresponding magnitude is denoted as v = (v) 1 ，v 2 ，…，v m ) Wherein
V(c i ) A magnitude domain that is an evaluation objective;
determining corresponding evaluation target feature set c = (c) by using rough set method 1 ，c 2 ，…，c m ) Weight coefficient of (a) = (a) 1 ，α 2 ，…α m ) The index which must be satisfied is expressed by the index Λ, i.e. α r If Λ, then
Construction of evaluation index set H = { H = { (H) } 1 ，H 2 ，…，H m In which H is i ＝(c i ，v i )，i＝1，2，…，m；
The initial convergence is as follows: using the evaluation object a that must be satisfied r ' = Lambda, performing primary convergence on the customization scheme set S = { B }, eliminating unsatisfied customization schemes, and constructing the customization scheme set
Then a single solution in the custom solution set is denoted as O pj Then product customization scheme set S 1 ，S 2 ，…，S n The number of schemes in (1) is sequentially marked as n 1 ，n 2 ，…n n ；
The goodness evaluation is a customized solution O pj Is recorded as c (O) p j) The goodness calculation has the following conditions:
1) Calculates the goodness by using the comprehensive correlation function, is suitable for multi-target evaluation of most product customization schemes,
2) Goodness C (O) pj ) Taking the minimum of the correlation function, i.e.
Each evaluation characteristic in the customized scheme evaluation process is required to meet the requirement and has no weight score;
3) Goodness C (O) pj ) Taking the maximum value of the correlation function, i.e.
The customized scheme evaluation process is expressed as long as one evaluation characteristic meets the requirement, and the customized scheme evaluation process has no weight score;
aiming at the goodness calculation of the customized scheme, introducing variance and related concepts to calculate the stability of the overall goodness of the customized scheme set;
goodness of individual solutions in a customized solution set
For discrete random variables, the standard goodness of the maximum individual scheme in the customized scheme set is taken as E (Y), namely
The variance D (Y) = E { [ Y-E (Y) with respect to goodness Y] 2 }；
Note the book
The standard deviation of goodness is known from the concept of variance and the meaning of semantic evaluation of the product, D (Y) or
A customization scheme S can be described p Regarding the degree of deviation of the goodness of the evaluation index from the standard goodness E (Y) as a whole, S in the customized recipe set S is reflected by (E (Y), σ (Y)) p Goodness and stability for a certain set of evaluation metrics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810925717.5A CN109741120B (en) | 2018-08-16 | 2018-08-16 | Product deep personalized customization method for user requirements |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810925717.5A CN109741120B (en) | 2018-08-16 | 2018-08-16 | Product deep personalized customization method for user requirements |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109741120A CN109741120A (en) | 2019-05-10 |
| CN109741120B true CN109741120B (en) | 2023-02-21 |
Family
ID=66354392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810925717.5A Active CN109741120B (en) | 2018-08-16 | 2018-08-16 | Product deep personalized customization method for user requirements |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109741120B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110517111B (en) * | 2019-08-15 | 2021-09-21 | 青岛科技大学 | Personalized customization method for product |
| CN111680359A (en) * | 2020-05-11 | 2020-09-18 | 武汉理工大学 | Rapid frame design method and device and computer storage medium |
| CN111815409B (en) * | 2020-07-06 | 2023-07-21 | 浙江大学 | Personalized design pattern matching method for 'Internet plus' environment mechanical products |
| JP7171094B2 (en) * | 2020-07-06 | 2022-11-15 | 浙江大学 | Personalized Design Pattern Matching Method for "Internet +" Environmental Machinery Products |
| CN112561667B (en) * | 2021-02-22 | 2021-06-04 | 浙江大胜达包装股份有限公司 | Fresh-keeping corrugated carton on-line customized production system and method thereof |
| CN116074120B (en) * | 2023-03-07 | 2023-05-30 | 北京思想天下教育科技有限公司 | Online collective opening system based on big data cloud platform |
| CN116720716B (en) * | 2023-08-09 | 2023-12-26 | 中笛(北京)文化发展有限公司 | Personalized customization method and system for ginger tea products based on big data |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6301516B1 (en) * | 1999-03-25 | 2001-10-09 | General Electric Company | Method for identifying critical to quality dependencies |
| CN105117562A (en) * | 2015-09-15 | 2015-12-02 | 河海大学常州校区 | Product green configuration method based on axiomatic design theory |
-
2018
- 2018-08-16 CN CN201810925717.5A patent/CN109741120B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6301516B1 (en) * | 1999-03-25 | 2001-10-09 | General Electric Company | Method for identifying critical to quality dependencies |
| CN105117562A (en) * | 2015-09-15 | 2015-12-02 | 河海大学常州校区 | Product green configuration method based on axiomatic design theory |
Non-Patent Citations (4)
| Title |
|---|
| 基于模块化产品族与客户需求转化的产品配置;覃燕红;《中国优秀硕士学位论文全文数据库 经济与管理科学辑》;20080515;正文第31-39页 * |
| 基于粗糙集的设计人员知识需求模型及知识推送研究;王克勤 等;《制造业自动化》;20140831;正文第20-21页 * |
| 基于语义关联度的产品可拓实例推理方法研究;吕健 等;《制造业自动化》;20150831;正文第137-139页 * |
| 面向大规模定制的产品族设计及评估研究;杨阳;《中国优秀硕士学位论文全文数据库 工程科技II辑》;20090315;正文第12-37页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109741120A (en) | 2019-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109741120B (en) | Product deep personalized customization method for user requirements | |
| Roszkowska | Multi-criteria decision making models by applying the TOPSIS method to crisp and interval data | |
| Lodwick et al. | Fuzzy optimization: Recent advances and applications | |
| CN107833117B (en) | Bayesian personalized sorting recommendation method considering tag information | |
| Kim et al. | An interactive procedure for multi-attribute group decision making with incomplete information | |
| CN109255586A (en) | A kind of online personalized recommendation method that E-Governance Oriented is handled affairs | |
| CN109615452A (en) | A kind of Products Show method based on matrix decomposition | |
| CN113742492B (en) | Insurance scheme generation method and device, electronic equipment and storage medium | |
| WO2023020257A1 (en) | Data prediction method and apparatus, and storage medium | |
| CN108280234A (en) | Data query method and device | |
| CN108319646A (en) | Vehicle source searching method and device based on user's history behavior | |
| CN109411093A (en) | A kind of intelligent medical treatment big data analysis processing method based on cloud computing | |
| CN110321492A (en) | A kind of item recommendation method and system based on community information | |
| CN111401988A (en) | Product configuration demand response system based on semantics and order generation method | |
| Ding et al. | Multiemotional product color design using gray theory and nondominated sorting genetic algorithm‐III | |
| Chen et al. | DPM-IEDA: dual probabilistic model assisted interactive estimation of distribution algorithm for personalized search | |
| CN114723535A (en) | Supply chain and knowledge graph-based item recommendation method, equipment and medium | |
| CN104317853B (en) | A kind of service cluster construction method based on Semantic Web | |
| Yan et al. | An integration of bidding-oriented product conceptualization and supply chain formation | |
| Sun et al. | Value-driven multiple criteria sorting with probabilistic linguistic information considering uncertain assignment examples | |
| CN108182288A (en) | A kind of recommendation method based on artificial immunity | |
| Liu et al. | Clothing similarity computation based on TLAC | |
| CN114997959A (en) | Electronic intelligent product marketing recommendation method | |
| CN110942082B (en) | Talent value evaluation method | |
| Sharif et al. | Link prediction based hybrid recommendation system using user-page preference graphs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |