Method for analyzing complex interaction relationship of sustainable design factors of product

Abstract

The invention discloses a method for analyzing a complex interaction relationship of sustainable design factors of a product, which comprises the following steps of: obtaining a document related to a designed product from internet big data, and processing the document to obtain a high-score vocabulary; taking the high-frequency vocabularies as sustainable design practice factors in sustainable design practice, determining an optimal factor, and constructing a directed graph to determine a highest-level factor; the method comprises the following steps: constructing a LoRA model, collecting product images of similar products of a designed product through an approach, and inputting the product images into a DeepBooru model to obtain an image label of each product image; using the product images and the image labels as samples, constructing a training set, inputting the training set into the LoRA model for training, and storing the trained LoRA model; and then selecting factors from all the highest-level factors of the directed graph, generating a product design scheme by using a model, and determining an optimal product design scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicle design, and in particular to a method for analyzing complex interactive relationships of product sustainable design factors. Background Art

[0002] In a globalized economic environment, sustainable development has become an important topic of common concern to the international community, especially in response to the challenges of environmental degradation and limited resources. As environmental problems become increasingly severe, environmental challenges including global warming, deteriorating air quality and energy crisis have forced governments and companies to seek greener and more sustainable solutions. Among the many solutions, the promotion of new energy vehicles (NEVs) is considered to be an effective challenge to the traditional fossil fuel dependence model and a key technological innovation to achieve carbon neutrality in the transportation sector. Studies have shown that NEVs have significant advantages in reducing the carbon footprint of cities, improving energy efficiency and promoting the transformation of energy consumption structure. The popularization of NEVs is not only seen as a technological revolution, but also a key component in the overall plan of sustainable consumption and production. The design, production and commercialization of NEVs contain many challenges and opportunities, which are located at the intersection of technology, consumer behavior and environmental policies.

[0003] However, despite the increasing maturity of technology, there are still significant differences in the popularity of new energy vehicles and the widespread acceptance of consumers around the world. This difference is partly due to the regional characteristics of market policies, economic incentives and consumer behavior. The new energy vehicle market has experienced rapid development. Through a series of policy supports, the research and development and market promotion of new energy vehicles have been effectively promoted; these policies have not only promoted technological progress, but also accelerated the market transition from traditional vehicles to new energy vehicles. Despite this, the market acceptance of new energy vehicles still faces the challenge between consumer expectations and product design. Consumers' cognition, emotions and psychological expectations of new energy vehicles have a great impact on their purchasing decisions. In addition, the application of the concept of sustainable design in the automotive industry has not yet been fully accepted by the market, which requires designers to consider not only the feasibility of technology, but also the needs of consumers for aesthetics and comfort. New energy vehicles serve consumers. In an era oriented to consumers, how to better understand and analyze consumers' inner demands for sustainable design, and at the same time, from a forward-looking and practical perspective to correlate and analyze the deep relationship between factors, is still a challenge for sustainability-related managers and enterprises. Summary of the invention

[0004] The purpose of the present invention is to provide a method for analyzing the complex interactive relationship of product sustainable design factors, so as to provide a new solution for product design and thus improve design efficiency.

[0005] In order to achieve the above tasks, the present invention adopts the following technical solutions:

[0006] A method for analyzing the complex interactive relationships of product sustainable design factors, including:

[0007] Step 1: Obtain documents related to the designed product from the Internet big data, perform data mining on them, filter out words related to the product, and perform data processing on the words to obtain high-scoring words;

[0008] Step 2: using the high-frequency words as sustainable design practice factors in sustainable design practice, determining the optimal sustainable design practice factor by calculating the evaluation values ​​of all sustainable design practice factors; constructing a directed graph based on the optimal sustainable design practice factor, and determining the highest-level factor in the directed graph;

[0009] Step 3, build the LoRA model by collecting product images of similar products to the designed product, inputting these product images into the DeepBooru model, and obtaining the image label of each product image; using the product images and image labels as samples, construct a training set and input it into the LoRA model for training, and save the trained LoRA model;

[0010] Then, from all the highest-level factors of the directed graph, a high-level factor that a product designer is concerned about is selected, and all the optimal sustainable design practice factors associated with the selected high-level factor are selected from the directed graph, and the high-level factor and all the optimal sustainable design practice factors are used as input information to generate a product design plan through the LoRA model; by adjusting one or the optimal sustainable design practice factor in the input information, a new product design plan is generated using the LoRA model after each adjustment;

[0011] Step 4: Determine the optimal product design solution from all generated product design solutions.

[0012] Furthermore, the data processing of the vocabulary to obtain high-scoring vocabulary includes:

[0013] Step 11, calculate the word frequency TF(t,d):

[0014]

[0015] Among them, f t,d represents the frequency of product-related vocabulary t in document d; f t′,d represents the frequency of any word t′ in document d;

[0016] Step 12, calculate the inverse document frequency IDF(t,D):

[0017]

[0018] Where D is a document set, d′ represents a document, and |D| represents the total number of documents in the document set. For a word t, |{d′∈D:t∈d′}| provides the total number of documents in the document set D that contain the word t.

[0019] Step 13, calculate the term frequency-inverse document frequency TF-IDF value as follows:

[0020] TF-IDF=TF(t,d)×IDF(t,D)

[0021] The TF-IDF values ​​of all product-related words are calculated, and all TF-IDF values ​​are sorted, and the top m words with the highest TF-IDF values ​​are selected as high-scoring words related to the product.

[0022] Furthermore, the method of calculating the evaluation values ​​of all sustainable design practice factors to determine the optimal sustainable design practice factor includes:

[0023] First, construct the decision matrix D, where the elements d in the matrix ij represents the score of the i-th sustainable design practice factor on the j-th criterion; then the intuitionistic fuzzy number is introduced, and for each element d in the decision matrix ij Applying intuitionistic fuzzy number processing, we get the intuitionistic fuzzy decision matrix; including the membership degree u ij and non-membership v ij The certainty and hesitation π ij Calculation of π ij =1-u ij -v ij ; The membership degree represents the positive evaluation degree of the i-th factor under the j-th criterion, the non-membership degree represents the negative evaluation degree of the i-th factor under the j-th criterion, and the hesitation degree represents the uncertainty of the evaluation of factor i under criterion j;

[0024] Then the intuitionistic fuzzy decision matrix is ​​standardized, and the standardized intuitionistic fuzzy number is expressed as Then, the weight of the criteria is objectively determined by the entropy weight method, and then for each sustainable design practice factor, its comprehensive evaluation value S under all criteria is calculated. i , the calculation formula is:

[0025]

[0026] Where n is the number of criteria, m is the number of sustainable design practice factors, and w j represents the weight of the jth criterion, μ′ ij ,v′ ij,π′ ij Respectively represent the standardized membership, non-membership and hesitation; μ′ kj , v′ kj It represents the membership and non-membership of the kth factor under the jth criterion after standardization;

[0027] Based on the comprehensive evaluation value S i , all sustainable design practice factors are ranked, and the top n sustainable design practice factors with the highest comprehensive evaluation values ​​are selected as the optimal sustainable design practice factors.

[0028] Further, the criteria are adaptability to the market, relevance to the industry in which the product operates and complexity of sustainable practices.

[0029] Furthermore, the construction of a directed graph based on the optimal sustainable design practice factors and determining the highest level factors in the directed graph include:

[0030] Assign numbers to the best sustainable design practice factors obtained p (p=1,2,…,n), and then construct the adjacency matrix A according to the relationship between the sustainable design practice factors, where the element a in the adjacency matrix A is pq Represents the sustainable design practice factor β p and β q The logical relationship between them is β p Impact Beta q If a pq =1, it means β p With β q There exists a p To β q relationship; if a pq =0, it means β p With β q There is no p To β q Direct relationship;

[0031] According to the adjacency matrix A, a Boolean operation is performed to obtain the reachable matrix R M :

[0032] (A+I) k-1 ≠(A+I) k =(A+I) k+1 =R M

[0033] Where I is defined as the identity matrix of the same order as A, and k is the minimum power to reach the final stable state of the reachable matrix;

[0034] Next, from β pThe set of all optimal sustainable design practice factors that can be reached is recorded as the reachable set R(β p ), and all possible arrivals at β p The set of optimal sustainable design practice factors is represented as the antecedent set A(β p ), where the reachable matrix R is obtained M Finally, you can start from any sustainable design practice factor and see which other factors it can directly or indirectly affect;

[0035] R(β p ) and A(β p ) find the intersection R(β p )∩A(β p ), then R(β p )∩A(β p ) is the factor β p can reach, and can reach β p All factors are set; a directed graph is drawn based on the reachability matrix. The directed graph is used to represent the logical relationship between the optimal sustainable design practice factors and determine the highest level factors in the directed graph:

[0036] If the intersection of the reachable set of a certain optimal sustainable design practice factor and its predecessor set is equal to its reachable set itself, then the factor and all the optimal sustainable design practice factors in the corresponding reachable set are the highest-level factors; the same method is used to determine all the highest-level factors.

[0037] Furthermore, determining the optimal product design solution from all generated product design solutions includes:

[0038] Step 4.1, set multiple evaluation criteria for the product, and then the decision maker scores each generated product design scheme from different evaluation criteria, and uses the scores to construct the evaluation decision matrix E. The elements E in the decision matrix ij It represents the evaluation value of the decision maker on the i-th product design scheme with respect to the j-th evaluation criterion;

[0039] Step 4.2, normalize the decision matrix; after normalization, its elements are represented as R ij ; For the jth evaluation criterion, from the corresponding E ij The difference between the maximum and minimum values ​​is selected as the difference value sj of the jth evaluation criterion;

[0040] Step 4.3, evaluate the differences between each product design and other product design solutions:

[0041]

[0042] Among them, P j(i,i′) represents the difference between the i-th and i′-th product design solutions, R i′j represents the normalized value of the evaluation value of the i′th product design scheme with respect to the jth evaluation criterion;

[0043] Step 4.4, for the i-th and i′-th product design solutions, calculate the preference value under each evaluation criterion:

[0044]

[0045] Among them, π(i,i′) represents the preference value of the i-th product design scheme relative to the i′-th product design scheme, and w j represents the weight of the preset j-th evaluation criterion, and u represents the number of evaluation criteria;

[0046] Step 4.5, for each product design solution i, calculate its positive flow and negative flow:

[0047]

[0048] Among them, i′≠i, φ + (i) φ - (i) are the positive and negative flows of the product design schemes, respectively; v represents the number of product design schemes;

[0049] Step 4.6, for each product design i, calculate its net flow φ(i), which is the difference between positive flow and negative flow:

[0050] φ(i)=φ + (i)-φ - (i)

[0051] Among all product design schemes, the one with the largest net flow φ(i) value is selected as the optimal product design scheme under the high-level factors that product designers are concerned about.

[0052] A terminal device comprises a processor, a memory and a computer program stored in the memory; the characteristic is that when the processor executes the computer program, the method of analyzing the complex interactive relationship of product sustainable design factors is implemented.

[0053] A computer-readable storage medium stores a computer program; the computer program, when executed by a processor, implements the method for analyzing the complex interactive relationship of product sustainable design factors.

[0054] Compared with the prior art, the present invention has the following technical features:

[0055] The present invention provides the intertwined relationships and impact analysis results of factors in the process of sustainable design of product industry supported by big data and natural language processing technology, and based on this, provides a set of product design solutions, which comprehensively considers the sustainable design needs of consumers and sustainable design practices (SDP), thereby helping enterprise management and designers to better understand the complex causal influence relationship in the sustainable design process behavior, determine the optimal product design solution, and improve product design efficiency and user experience. It organically combines the knowledge of operations research, design, neural networks, intelligent science, and consumer behavior to a certain extent, giving full play to the greatest advantage of interdisciplinary cross-disciplinary in the field of sustainability. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic flow diagram of the method of the present invention;

[0057] Figure 2 is a schematic diagram of a directed graph in one embodiment of the present invention;

[0058] Figure 3 A schematic diagram of a product design solution generated in one embodiment of the present invention. DETAILED DESCRIPTION

[0059] See attached Figure 1 The present invention provides a method for analyzing the complex interactive relationship of product sustainable design factors, comprising the following steps:

[0060] Step 1: Obtain documents related to the designed product from the Internet big data, perform data mining on them, filter out words related to the product, and perform data processing on the words to obtain high-scoring words.

[0061] Step 11, calculate TF(t,d). Term frequency (TF) defines the probability of a word appearing in a document, while inverse document frequency (IDF) measures the specificity of a word.

[0062]

[0063] Among them, f t,d represents the frequency of product-related vocabulary t in document d; f t′,d represents the frequency of any word t′ in document d, |{f t′,D :t′∈d}| gives the total number of all words in document d.

[0064] Step 12, calculate the inverse document frequency IDF(t,D):

[0065]

[0066] Where D is a document set, d′ represents a document, and |D| represents the total number of documents in the document set; for a word t, |{d′∈D:t∈d′}| provides the total number of documents in the document set D that contain the word t.

[0067] Step 13, calculate the term frequency-inverse document frequency TF-IDF value as follows:

[0068] TF-IDF=TF(t,d)×IDF(t,D)

[0069] The TF-IDF values ​​of all product-related words are calculated, and all TF-IDF values ​​are sorted, and the top m words with the highest TF-IDF values ​​are selected as high-scoring words related to the product.

[0070] In one embodiment of the present invention, for example, when the product to be designed is a new energy vehicle, a document collection related to new energy vehicles is obtained from Internet data, and the vocabulary in the document is analyzed, and finally 18 high-scoring vocabulary are obtained, including interior, appearance, power, sound insulation, comfort, cost performance, etc.

[0071] Step 2: Use the high-frequency words as sustainable design practice factors in the sustainable design practice SDP, and determine the optimal sustainable design practice factor by calculating the evaluation values ​​of all sustainable design practice factors; construct a directed graph based on the optimal sustainable design practice factor, and determine the highest-level factor in the directed graph.

[0072] Step 2.1, determine the optimal sustainable design practice factors.

[0073] First, construct the decision matrix D, where the elements d in the matrix ij represents the score of the i-th sustainable design practice factor on the j-th criterion, where the criteria are the adaptability to the Chinese market, the relevance of the industry in which the product is located, and the complexity of sustainable practices; then the intuitionistic fuzzy number is introduced, and each element d in the decision matrix ij Applying intuitionistic fuzzy number processing, we get the intuitionistic fuzzy decision matrix; this involves the membership degree u ij and non-membership v ij The certainty and hesitation π ij Calculation of π ij =1-u ij -v ij ; The membership degree represents the positive evaluation degree of the i-th factor under the j-th criterion, the non-membership degree represents the negative evaluation degree of the i-th factor under the j-th criterion, and the hesitation degree represents the uncertainty of the evaluation of factor i under criterion j.

[0074] Then the intuitionistic fuzzy decision matrix is ​​standardized, and the standardized intuitionistic fuzzy number is expressed as Then, the weight of the criteria is objectively determined by the entropy weight method, and then for each sustainable design practice factor, its comprehensive evaluation value S under all criteria is calculated. i , the calculation formula is:

[0075]

[0076] Where n is the number of criteria, m is the number of sustainable design practice factors, and w j represents the weight of the jth criterion, μ′ ij ,v′ ij ,π′ ij Respectively represent the standardized membership, non-membership and hesitation; μ′ kj , v′ kj It represents the membership and non-membership of the kth factor under the jth criterion after standardization.

[0077] Based on the comprehensive evaluation value S i , all sustainable design practice factors are ranked, and the top n sustainable design practice factors with the highest comprehensive evaluation values ​​are selected as the optimal sustainable design practice factors.

[0078] Step 2.2, construct a directed graph and determine the highest level factors.

[0079] Assign numbers to the best sustainable design practice factors obtained p (p=1,2,…,n), and then construct the adjacency matrix A according to the relationship between the sustainable design practice factors, where the element a in the adjacency matrix A is pq Represents the sustainable design practice factor β p and β q The logical relationship between them is β p Impact Beta q If a pq =1, it means β p With β q There exists a p To β q relationship; if a pq =0, it means β p With β q There is no p To β q Direct relationship; for example, the two factors are "sound insulation" and "comfort". Since sound insulation affects comfort, the value of the logical association between the two is 1.

[0080] According to the adjacency matrix A, a Boolean operation is performed to obtain the reachable matrix R M :

[0081] (A+I)k-1 ≠(A+I) k =(A+1) k+1 =R M

[0082] Where I is defined as the identity matrix of the same order as A, and k is the minimum power to reach the final stable state of the reachable matrix (no more reachable relations).

[0083] Next, from β p The set of all optimal sustainable design practice factors that can be reached is recorded as the reachable set R(β p ), and all possible arrivals at β p The set of optimal sustainable design practice factors is represented as the antecedent set A(β p ), where the reachable matrix R is obtained M After that, we can start from any sustainable design practice factor to see which other factors this factor can directly or indirectly affect; the reachable set R(β p ) is used to describe this concept, that is, from a specific factor β p Starting from, the set of all factors that can be reached along the directed relationship chain.

[0084] R(β p ) and A(β p ) find the intersection R(β p )∩A(β p ), then R(β p )∩A(β p ) is the factor β p can reach, and can reach β p All factors set; if R(β p )∩A(β p )=R(β p ), then R(β p )The factors in the set are the highest level of all factors.

[0085] Based on the reachability matrix, a directed graph is drawn to represent the logical association between the optimal sustainable design practice factors and to determine the highest level factors in the directed graph:

[0086] If the intersection of the reachable set of a certain optimal sustainable design practice factor and its predecessor set is equal to its reachable set itself, then the factor and all the optimal sustainable design practice factors in the corresponding reachable set are the highest-level factors; the same method is used to determine all the highest-level factors.

[0087] Step 3, build a LoRA model, collect product images of similar products of the designed product from the Internet and other channels, input these product images into the DeepBooru model, and obtain the image label of each product image; use the product images and image labels as samples, build a training set and input it into the LoRA model for training, and save the trained LoRA model; wherein, the LoRA model is an existing model and will not be repeated here.

[0088] Then, from all the highest-level factors in the directed graph, a high-level factor that a product designer is concerned about is selected, and all optimal sustainable design practice factors associated with the selected high-level factor are selected from the directed graph. The high-level factor and all the optimal sustainable design practice factors are used as input information, and a product design plan (i.e., a product image) is generated through the LoRA model; by adjusting (including deleting, replacing, etc.) one or more optimal sustainable design practice factors in the input information, a new product design plan is generated using the LoRA model after each adjustment.

[0089] In one embodiment of the present invention, when designing a new energy vehicle, the product design scheme is as follows: Figure 3 As shown in (a1) to (c4).

[0090] Step 4: Determine the optimal product design solution from all generated product design solutions; the details are as follows:

[0091] Step 4.1, set multiple evaluation criteria for the product, and then the decision maker scores each product design solution generated in step 3 based on different evaluation criteria, and uses the scores to construct the evaluation decision matrix E. The elements E in the decision matrix ij It represents the evaluation value of the decision maker on the jth evaluation criterion for the i-th product design scheme; the evaluation criterion may be, for example, aesthetics, practicality, etc., and a specific score is set for each evaluation criterion.

[0092] Step 4.2, normalize the decision matrix; after normalization, its elements are represented as R ij ; For the jth evaluation criterion, from the corresponding E ij Select the difference between the maximum and minimum values ​​as the difference value s of the jth evaluation criterion j .

[0093] Step 4.3, evaluate the differences between each product design and other product design solutions:

[0094]

[0095] Among them, P j (i,i′) represents the difference between the i-th and i′-th product design solutions, R i′jIt represents the normalized value of the evaluation value of the i′th product design scheme with respect to the jth evaluation criterion.

[0096] Step 4.4, for the i-th and i′-th product design solutions, calculate the preference value under each evaluation criterion:

[0097]

[0098] Among them, π(i,i′) represents the preference value of the i-th product design scheme relative to the i′-th product design scheme, and w j represents the weight of the preset j-th evaluation criterion, and u represents the number of evaluation criteria.

[0099] Step 4.5, for each product design solution i, calculate its positive flow and negative flow:

[0100]

[0101] Among them, i′≠i, φ + (i) φ - (i) are the positive and negative flows of product design solutions, respectively; v represents the number of product design solutions.

[0102] Step 4.6, for each product design i, calculate its net flow φ(i), which is the difference between positive flow and negative flow:

[0103] φ(i)=φ + (i)-φ - (i)

[0104] Among all product design schemes, the product design scheme with the largest net flow φ(i) value is selected as the optimal product design scheme under the high-level factors that the product designer is concerned about. By replacing different high-level factors and repeating the above steps, the optimal product design scheme corresponding to different high-level factors of concern can be obtained.

[0105] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for analyzing the complex interactive relationship of product sustainable design factors, characterized in that: include: Step 1: Obtain documents related to the designed product from the Internet big data, perform data mining on them, filter out words related to the product, and perform data processing on the words to obtain high-scoring words; Step 2: using the high-frequency words as sustainable design practice factors in sustainable design practice, determining the optimal sustainable design practice factor by calculating the evaluation values ​​of all sustainable design practice factors; constructing a directed graph based on the optimal sustainable design practice factor, and determining the highest-level factor in the directed graph; Step 3, build the LoRA model by collecting product images of similar products to the designed product, inputting these product images into the DeepBooru model, and obtaining the image label of each product image; using the product images and image labels as samples, construct a training set and input it into the LoRA model for training, and save the trained LoRA model; Then, from all the highest-level factors of the directed graph, a high-level factor that a product designer is concerned about is selected, and all the optimal sustainable design practice factors associated with the selected high-level factor are selected from the directed graph, and the high-level factor and all the optimal sustainable design practice factors are used as input information to generate a product design plan through the LoRA model; by adjusting one or the optimal sustainable design practice factor in the input information, a new product design plan is generated using the LoRA model after each adjustment; Step 4: Determine the optimal product design solution from all generated product design solutions.

2. The method for analyzing complex interactive relationships of product sustainable design factors according to claim 1, characterized in that: The data processing of the vocabulary to obtain high-scoring vocabulary includes: Step 11, calculate the word frequency TF(t,d): Among them, f t,d represents the frequency of product-related vocabulary t in document d; f t′,d represents the frequency of any word t′ in document d; Step 12, calculate the inverse document frequency IDF(t,D): Where D is a document set, d′ represents a document, and |D| represents the total number of documents in the document set. For a word t, |{d′∈D:t∈d′}| provides the total number of documents in the document set D that contain the word t. Step 13, calculate the term frequency-inverse document frequency TF-IDF value as follows: TF-IDF=TF(t,d)×IDF(t,D) The TF-IDF values ​​of all product-related words are calculated, and all TF-IDF values ​​are sorted, and the top m words with the highest TF-IDF values ​​are selected as high-scoring words related to the product.

3. The method for analyzing complex interactive relationships of product sustainable design factors according to claim 1, characterized in that: The above method calculates the evaluation values ​​of all sustainable design practice factors to determine the optimal sustainable design practice factor, including: First, construct the decision matrix D, where the elements d in the matrix ij represents the score of the i-th sustainable design practice factor on the j-th criterion; then the intuitionistic fuzzy number is introduced, and for each element d in the decision matrix ij Applying intuitionistic fuzzy number processing, we get the intuitionistic fuzzy decision matrix; including the membership degree u ij and non-membership v ij The certainty and hesitation π ij Calculation of π ij =1-u ij -v ij ; The membership degree represents the positive evaluation degree of the i-th factor under the j-th criterion, the non-membership degree represents the negative evaluation degree of the i-th factor under the j-th criterion, and the hesitation degree represents the uncertainty of the evaluation of factor i under criterion j; Then the intuitionistic fuzzy decision matrix is ​​standardized, and the standardized intuitionistic fuzzy number is expressed as Then, the weight of the criteria is objectively determined by the entropy weight method, and then for each sustainable design practice factor, its comprehensive evaluation value S under all criteria is calculated. i , the calculation formula is: Where n is the number of criteria, m is the number of sustainable design practice factors, and w j represents the weight of the jth criterion, μ′ ij ,v′ ij ,π′ ij Respectively represent the standardized membership, non-membership and hesitation; μ′ kj , v′ kj It represents the membership and non-membership of the kth factor under the jth criterion after standardization; Based on the comprehensive evaluation value S i , all sustainable design practice factors are ranked, and the top n sustainable design practice factors with the highest comprehensive evaluation values ​​are selected as the optimal sustainable design practice factors.

4. The method for analyzing complex interactive relationships of product sustainable design factors according to claim 3, characterized in that: The criteria mentioned are adaptability to the market, relevance to the industry in which the product operates and complexity of sustainable practices.

5. The method for analyzing complex interactive relationships of product sustainable design factors according to claim 1, characterized in that: The directed graph is constructed based on the optimal sustainable design practice factors, and the highest level factors in the directed graph are determined, including: Assign numbers to the best sustainable design practice factors obtained p (p = 1, 2, ..., n), and then construct an adjacency matrix A based on the relationship between the sustainable design practice factors, where the element a in the adjacency matrix A is pq Represents the sustainable design practice factor β p and β q The logical relationship between them is β p Impact Beta q If a pq =1, it means β p With β q There exists a p To β q relationship; if a pq =0, it means β p With β q There is no p To β q Direct relationship; According to the adjacency matrix A, a Boolean operation is performed to obtain the reachable matrix R M : (A+I) k-1 ≠(A+I) k =(A+I) k+1 =R M where is defined as the identity matrix of the same order as A, and k is the minimum power to reach the final stable state of the reachable matrix; Next, from β p The set of all optimal sustainable design practice factors that can be reached is recorded as the reachable set R(β p ), and all possible arrivals at β p The set of optimal sustainable design practice factors is represented as the antecedent set A(β p ), where the reachable matrix R is obtained M Finally, you can start from any sustainable design practice factor and see which other factors it can directly or indirectly affect; R(β p ) and A(β p ) find the intersection R(β p )∩A(β p ), then R(β p )∩A(β p ) is the factor β p can reach, and can reach β p All factors are set; a directed graph is drawn based on the reachability matrix. The directed graph is used to represent the logical relationship between the optimal sustainable design practice factors and determine the highest level factors in the directed graph: If the intersection of the reachable set of a certain optimal sustainable design practice factor and its predecessor set is equal to its reachable set itself, then the factor and all the optimal sustainable design practice factors in the corresponding reachable set are the highest-level factors; the same method is used to determine all the highest-level factors.

6. The method for analyzing complex interactive relationships of product sustainable design factors according to claim 1, characterized in that: The step of determining the optimal product design solution from all generated product design solutions includes: Step 4.1, set multiple evaluation criteria for the product, and then the decision maker scores each generated product design scheme from different evaluation criteria, and uses the scores to construct the evaluation decision matrix E. The elements E in the decision matrix ij It represents the evaluation value of the decision maker on the i-th product design scheme with respect to the j-th evaluation criterion; Step 4.2, normalize the decision matrix; after normalization, its elements are represented as R ij ; For the jth evaluation criterion, from the corresponding E ij Select the difference between the maximum and minimum values ​​as the difference value s of the jth evaluation criterion j ; Step 4.3, evaluate the differences between each product design and other product design solutions: Among them, P j (i,i′) represents the difference between the i-th and i′-th product design solutions, R i′j represents the normalized value of the evaluation value of the i′th product design scheme with respect to the jth evaluation criterion; Step 4.4, for the i-th and i′-th product design solutions, calculate the preference value under each evaluation criterion: Among them, π(i,i′) represents the preference value of the i-th product design scheme relative to the i′-th product design scheme, and w j represents the weight of the preset j-th evaluation criterion, and u represents the number of evaluation criteria; Step 4.5, for each product design solution i, calculate its positive flow and negative flow: Among them, i′≠i, φ + (i) φ - (i) are the positive and negative flows of the product design schemes, respectively; v represents the number of product design schemes; Step 4.6, for each product design i, calculate its net flow φ(i), which is the difference between positive flow and negative flow: φ(i)=φ + (i)-φ - (i) Among all product design schemes, the one with the largest net flow φ(i) value is selected as the optimal product design scheme under the high-level factors that product designers are concerned about.

7. A terminal device comprising a processor, a memory and a computer program stored in the memory; characterized in that: When the processor executes the computer program, the method for analyzing the complex interactive relationship of product sustainable design factors according to any one of claims 1 to 6 is implemented.

8. A computer-readable storage medium, wherein a computer program is stored in the medium; characterized in that: When the computer program is executed by a processor, the method for analyzing the complex interactive relationship of product sustainable design factors according to any one of claims 1 to 6 is implemented.

Images