CN-121998442-A - Crowd-driven product form design multi-attribute decision-making method in cloud environment

CN121998442ACN 121998442 ACN121998442 ACN 121998442ACN-121998442-A

Abstract

A multi-attribute decision method for group intelligence driven product form design under cloud environment includes such steps as quantitatively calculating qualitative product attributes by using Bithagoras hesive fuzzy set, creating multi-attribute evaluation system of product form design scheme, creating ash correlation coefficient decision matrix based on expert knowledge, calculating the total score of basic alternative scheme by multi-criterion compromised solution sorting method and improved close value method, obtaining user weight by calculating similarity between user evaluation matrices, calculating product form scheme score based on user preference by approaching ideal solution sorting method, optimizing product form design scheme by aggregating expert evaluation value and user preference value, and applying case to explain the effectiveness and feasibility of said method.

Inventors

CHEN JIAN
HE ZHAOXUAN
WANG WEIWEI
WANG YI
LI ZHIHAN

Assignees

陕西科技大学

Dates

Publication Date: 20260508
Application Date: 20241106

Claims (6)

1. A multi-attribute decision method for group intelligent driving product form design in cloud environment is characterized by comprising the following steps: Step 1, establishing a product form design scheme multi-attribute evaluation system based on a Pichia hesitans fuzzy set; Step 2, providing a product form design decision multi-attribute index calculation method according to the product form design scheme multi-attribute evaluation system established in the step 1, and carrying out quantitative calculation on qualitative product attributes; Step 3, constructing a product form scheme decision model based on expert knowledge, constructing a product form design multi-attribute decision technology framework based on the expert knowledge, and calculating to obtain a scheme evaluation value based on the expert knowledge according to expert evaluation; Step 4, a product form scheme decision model based on user preference is built, a product form design multi-attribute decision technology framework based on the user preference is built, and a scheme evaluation value based on the user preference is calculated according to user evaluation; And 5, carrying out final decision by fusing the scheme evaluation value based on expert knowledge and the scheme evaluation value based on user preference to obtain a final scheme evaluation value.
2. The method for multi-attribute decision-making of group-intelligence driven product morphology design in cloud environment according to claim 1, wherein the step 1 is to build a multi-attribute evaluation system of product morphology design based on the pythagoras hesitation fuzzy set, and the specific method is as follows: based on the characteristics of product form design decision, simultaneously combining with the characteristics of the hesitation and the fuzziness of an evaluator, and evaluating by using a Pythagorean hesitation fuzziness set, objectively describing the uncertainty and the fuzziness between the attributes in the product form design scheme, wherein the mathematical description is as follows: let X be a field, M be a Pichia-Las hesitant modulo set on X: M={<x,Γ M (x),ψ M (x)>|x∈X} (1) In the formula (1), X is a set of all product form design alternatives, M is a Bidago hesitation ambiguity set on X and is used for representing ambiguity and uncertainty of a certain product form design scheme X in a multi-attribute decision process, Γ M (X) and ψ M (X) are non-empty finite sets on [0,1], Γ M (X) represents membership degree of X to M and reflects degree that a decision maker considers that the product form design scheme X meets or is superior to an evaluation standard on a certain attribute, and ψ M (X) represents non-membership degree that X belongs to M and reflects degree that a decision maker considers that the product form design scheme X does not meet or is inferior to the evaluation standard on a certain attribute, and Mu M ∈Γ M (x),v M ∈ψ M (x), mu M ,v M E [0,1],
3. The method for computing the multi-attribute index of the product form design driven by crowd-sourcing according to claim 1, wherein the domain expert and the user together use the Bithagoras hesitation fuzzy number alpha= < Γ α ,Ψ α > in the cloud environment to give the evaluation of each attribute of each alternative scheme, and the final comprehensive score is computed according to the Bithagoras hesitation fuzzy weighted geometric average operator and the evaluation information obtained by aggregating each attribute, and the mathematical description is as follows: Assuming that m alternative scheme sets are P= { P 1 ,P 2 ,P 3 ,...,P m }, n attributes are expressed as F= { F 1 ,F 2 ,F 3 ,...,F n }, U expert sets D= [ D 1 ,D 2 ,D 3 ,...,D u ], q user group sets U= [ U 1 ,U 2 ,U 3 ,...,U q ], and an expert D u and a user U q evaluate and assign satisfaction and dissatisfaction of an attribute F j corresponding to a scheme P i in the form of Picorn hesitation blur number, and are recorded as alpha ij ＝<Γ ij ,ψ ij >, and a Picorn hesitation blur number decision expert individual evaluation matrix is constructed And a user individual evaluation matrix Wherein the method comprises the steps of And The values of the jth attribute in the scheme P i respectively representing the jth expert and the jth user are normalized based on the Pythagorean hesitation fuzzy set to obtain a normalized individual decision matrix And Integrating the scheme attributes by combining the Picgo hesitation fuzzy integration operator, and calculating to obtain the score value of each attribute of the scheme by the formula (2); The composite score of each attribute in the alternative is: In the formula (2), S α is a scoring function of alpha, which represents the comprehensive score of a certain attribute in a certain alternative scheme, and I gamma α I and I gamma α I respectively represent the number of elements in gamma α ,ψ α , and are derived from scores of designers and users on the attribute; In the cloud environment, α 1 and α 2 are evaluation information of two different decision individuals, and the euclidean distance between α 1 and α 2 is: In the formula (3), l (Γ) and l (ψ) represent the number of elements in the membership set and non-membership set, Representing the number of membership in alpha 1 , Representing the number of membership in alpha 2 , Representing the number of non-membership degrees in alpha 1 , Representing the number of non-membership degrees in alpha 2 , l (pi) representing the number of elements in the hesitation set, And The hesitation degrees of alpha 1 and alpha 2 are respectively expressed, and the calculation formula is as follows: In the formula (4), pi α represents the hesitation of a decision maker in the decision making process of the product form design scheme.
4. The method for crowd-sourced driven product morphology design multi-attribute decision making in a cloud environment of claim 1, wherein the method comprises the steps of, The product morphology scheme decision model based on expert knowledge in the step 3 comprises the following specific steps: The method comprises the steps of combining a gray correlation analysis method, a multi-criterion compromises sorting method and an improved close value method to make decisions, fully considering the correlation degree between evaluation indexes of an alternative scheme and intuitively expressing the closeness degree between the alternative scheme and an ideal scheme, realizing multi-angle evaluation, giving out the evaluation language of an expert based on a Bithago hesitation fuzzy set, converting the evaluation language into a specific numerical value to construct a preliminary decision matrix, introducing the gray correlation analysis method to construct a gray correlation coefficient decision matrix, calculating the correlation weight of attributes, quantifying the difference and the correlation between different alternative schemes, finally combining the multi-criterion compromises sorting method and the improved close value method to calculate the scheme evaluation value based on expert knowledge, and specifically solving the following steps: 1) Constructing a scheme evaluation matrix based on expert knowledge Firstly, determining an alternative scheme set P and an evaluation attribute set F, evaluating by an expert according to the actual performance of each scheme on each attribute and based on the Pythagorean hesitation fuzzy number, constructing an m multiplied by n preliminary decision matrix C= (C ij ) m×n , and secondly, normalizing the matrix to be In the formulas (5) - (6), C is an expert individual evaluation matrix, and C mn represents the score of the expert individual on the performance of the scheme m on the attribute n; evaluating a matrix for the standardized expert individuals; evaluating information for the standardized expert individuals; 2) Calculating an overall score for each attribute of a schema Integrating an expert evaluation matrix C' through a Picornian hesis fuzzy weighted geometric average operator, and obtaining a comprehensive score S α of each attribute of the scheme through calculation of the formula (2); 3) Establishing a grey correlation coefficient decision matrix and determining attribute weights For each attribute in the decision matrix, calculating the gray correlation coefficient between the attribute and the ideal sequence C 0 by using a gray correlation analysis method, forming a new decision matrix by using all the calculated gray correlation coefficients, namely a gray correlation coefficient decision matrix xi, and calculating the correlation degree between each attribute to obtain the importance of each attribute on the whole scheme and the influence degree between the attributes. In the formula (7), the amino acid sequence of the compound, And Positive and negative ideal solutions respectively representing the jth attribute of the ith scheme; In equation (8), ζ ij represents the gray correlation coefficient between the jth attribute of the ith scheme and the ideal solution, C 0 (j) is the jth attribute value in the ideal scheme, x (i, j) is the value of the ith scheme in the matrix with respect to the jth attribute, ρ is the resolution coefficient, taken 0.5, min and maxmax represent the minimum and maximum operations taken through all i, j, In the formula (9), xi is a grey correlation coefficient decision matrix, xi mn represents the grey correlation coefficient between the nth attribute of the mth scheme and the ideal solution, The gray correlation r of the attribute is: in the formula (10), n is the number of attributes, and the j-th attribute weight w j is: 4) Obtaining expert knowledge-based solution evaluation values Firstly, the multi-criterion compromise solution ordering method is utilized to determine the group utility value S i , the individual regretta value R i and the benefit ratio value Q i of each scheme, then the improved close value method is utilized to expand the dissimilarity between schemes, the degree of the superiority and inferiority of the schemes is more accurately evaluated, more comprehensive and balanced decision can be made through the combination of the multi-criterion compromise solution ordering method and the improved close value method, ① Calculating a population utility value S i , an individual regrind value R i , and a benefit ratio value Q i In the formula (12), v is a decision coefficient, 0.5 is taken, the closeness degree between the benefit ratio value Q i and an ideal scheme is measured, an optimal scheme is determined, and the smaller the benefit ratio value Q i is, the better the scheme is; ② Calculate the value of the close OV i In formulas (13) - (15): Is a virtual worst ideal solution; Respectively is And OV i is the close value of the product morphology scheme, and the smaller the close value OV i , the better the scheme performance; ③ Aggregate evaluation value In the formula (16), the amino acid sequence of the compound, And The normalized values of the benefit ratio value Q i and the close value OV i are respectively obtained, lambda is the distribution coefficient, and lambda=0.5 is taken.
5. The method for crowd-sourced driven product morphology design multi-attribute decision making in a cloud environment of claim 1, wherein the method comprises the steps of, The step 4 of constructing a product form scheme decision model based on user preference comprises the following specific steps: Giving an evaluation language of a user based on a Bithagoras hesitation fuzzy set in a product morphology scheme decision model based on user preferences, converting the evaluation language into a specific numerical value, constructing an evaluation matrix based on the user preferences by adopting the Bithagoras hesitation fuzzy set, calculating the similarity between the user evaluation decision matrices through Hamming distances, obtaining weights of different users, enabling the weight determination to be more scientific, and finally introducing a method for calculating the closeness degree between an alternative scheme and an ideal scheme by an approximate ideal solution ordering method to obtain a preference information value; 1) Constructing a user preference based solution evaluation matrix The user U q assigns the attribute F j corresponding to the scheme P i as R ij in the form of Pythagorean hesitation fuzziness according to the preference degree of each attribute of the scheme, and constructs a user individual evaluation matrix R= (R ij ) m×n , and the standardized matrix is R mn represents the score of the user individual on the performance of scheme m on attribute n; evaluating the matrix for the standardized individual users; and evaluating information for the standardized individual users. 2) Determining user weights Because the demands and the importance of different users are different, evaluation information conflict can occur, so that the satisfaction degree of the users cannot be truly reflected, and the similarity of the hesitant and fuzzy Pythagorean matrix is adopted to calculate the user weight. The similarity SIM between the evaluation matrixes of each user is calculated through the Hamming distance to obtain the support degree Sup of the user preference about other users, the larger the similarity is, the larger the support degree is, the more convincing the evaluation information endowed by the corresponding user is, namely, the user weight gamma is endowed, and therefore, the user can be endowed with larger weight, and the calculation formula is as follows: In equations (19) - (23), SIM (q, l) represents the similarity between the user U q and U l evaluation matrices, As the fuzzy number And Also known as Hamming distance, w j is an attribute weight, SIM is a similarity matrix for all users, sup q represents the support of user U q , y q represents the weight of user U q ; 3) Calculating user preference values for alternatives In order to comprehensively consider a plurality of attributes, a solution ordering approach approaching ideal is introduced to consider the scheme preference information value, and decision is made according to the closeness degree of an alternative scheme and an ideal scheme, so that the attribute value of a scheme index can be well considered; ① Information is gathered on the standardized decision matrix and the user weights, and the Pythagorean hesitation fuzzy weighting score of each attribute of the alternative scheme is obtained; ② Mapping the distances between each scheme and positive and negative ideal schemes according to the Pythagorean hesitant fuzzy distance; wherein d + (P i ) and d - (P i ) are the distances from scheme P i to the positive and negative ideal schemes, respectively; And Respectively representing satisfaction and dissatisfaction of the attribute j in the positive ideal solution; And Respectively representing satisfaction degree and dissatisfaction degree of the attribute j in the negative ideal solution, wherein w j is the weight of the attribute; ③ The merits of each scheme are calculated by the ratio of the distance between each scheme and the positive and negative ideal schemes: where τ i is the final user preference based solution score, with larger values representing the optimal solution.
6. The method for multi-attribute decision making of a crowd-driven product form design in a cloud environment according to claim 1, wherein the crowd-sourced solution optimization method in step 5 comprises the following specific steps: In order to enable the decision result to reflect the objective reality of the expert group and the preference degree of different user groups, final decision is carried out by adopting the dispersion minimization and fusion of the expert and the evaluation information of the users, and the alternative schemes are ordered according to the comprehensive score to obtain a final preferred scheme: in the formula, the alternatives are sequenced according to the comprehensive score min eta to obtain a final decision result, and the higher the eta value is, the most balanced the expert of the alternatives and the user satisfaction is, namely the optimal scheme is obtained.

Description

Crowd-driven product form design multi-attribute decision-making method in cloud environment Technical Field The invention belongs to the technical field of multi-attribute decision making of product form design, and particularly relates to a multi-attribute decision making method of product form design driven by crowd in a cloud environment. Background Literature "product innovation design multi-attribute decision evaluation considering customer demand preference" computer integrated manufacturing system, 2015, vol21 (02), p417-426 discloses product innovation design multi-attribute decision evaluation considering customer demand preference. According to the method, based on analysis of customer demand preference, an evaluation index weight determining method integrating a Kano model and a rough set theory is provided, so that the rationality of index weight determination is improved, and customer demand preference information is effectively reflected in an evaluation process. In consideration of the ambiguity and uncertainty of decision information in scheme evaluation, a multi-attribute decision gray correlation analysis model of the product innovation design scheme based on coarse number decision information processing is constructed, and the optimization of the product innovation design scheme is realized by integrating the index weight and calculating coarse number interval difference coefficients of each scheme. The method comprehensively considers the user preference and the ambiguity of the user decision information, but does not consider the user weight of different backgrounds and the hesitation of the user decision information. Meanwhile, the method only considers one role of the user when the decision member selects, has defects in the aspect of global decision, and cannot make more balanced selection. Disclosure of Invention In order to solve the problems that a decision main body is single and user participation is low in a product form scheme decision process in a cloud environment, the invention aims to provide a multi-attribute decision method for product form design driven by crowd in the cloud environment; the method is based on analysis of a product form design crowd-sourcing service mode in a cloud environment and a product form decision-making problem in the cloud environment, a product form multi-attribute decision-making method integrating expert knowledge and user preference is adopted, a product form design scheme with the most balanced satisfaction degree of an expert and a user is optimized, qualitative product attributes are quantitatively calculated by means of a Pythagorean hesis fuzzy set, and a multi-attribute evaluation system of the product form design scheme is established. And secondly, establishing an expert knowledge-based gray correlation coefficient decision matrix in expert decision, and calculating the overall score of the base alternative scheme by using a multi-criterion compromises sorting method and an improved close value method. And obtaining user weights by calculating the similarity among the user evaluation matrixes in the aspect of user decision, and calculating the product form scheme scores based on user preference by adopting an approximate ideal solution ordering method. And finally, the scale of decision members and the sources of decision information are widened, the objectivity and the accuracy of a decision result are improved, and the method has important significance for the direction of product morphological design and the floor property of a lifting scheme. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: A multi-attribute decision method for group intelligent driven product form design in cloud environment comprises the following steps: Step 1, establishing a product form design scheme multi-attribute evaluation system based on a Pichia hesitans fuzzy set; Step 2, providing a product form design decision multi-attribute index calculation method according to the product form design scheme multi-attribute evaluation system established in the step 1, and carrying out quantitative calculation on qualitative product attributes; Step 3, constructing a product form scheme decision model based on expert knowledge, constructing a product form design multi-attribute decision technology framework based on the expert knowledge, and calculating to obtain a scheme evaluation value based on the expert knowledge according to expert evaluation; Step 4, a product form scheme decision model based on user preference is built, a product form design multi-attribute decision technology framework based on the user preference is built, and a scheme evaluation value based on the user preference is calculated according to user evaluation; And 5, carrying out final decision by fusing the scheme evaluation value based on expert knowledge and the scheme evaluation value based on user preference to obtain a final scheme e