Knowledge importance ranking method for innovative design

Abstract

The invention relates to an innovative design-oriented knowledge importance ranking method. The method comprises the following steps of: 1, determining indexes required by knowledge importance ranking; 2, determining a proper fuzzy semantic quantization criterion, and calculating to obtain a fuzzy number weight average value; 3, determining experts participating in evaluation and weights of the experts; 4, the experts compare the indexes in pairs, and a matrix formed by triangular fuzzy numbers is used for representing the relative importance degree between the different evaluation indexes; 5, calculating the triangular fuzzy number matrix to obtain a weight value of each index, and performing normalization processing; 6, summarizing the evaluation suggestions of a plurality of experts, and determining the weight of each evaluation index; 7, giving a score for each item of knowledge to be evaluated by a scoring expert according to the matching degree, and obtaining a specific score of each index; and 8, summarizing and calculating evaluation results, and sorting innovation knowledge according to scores. The knowledge importance calculation accuracy is improved.

Description

Technical Field

[0001] The invention belongs to the field of computer data processing, and in particular relates to a knowledge importance ranking method for innovative design. Background Art

[0002] With the rapid development of generative artificial intelligence technology, there is a methodological gap that needs to be filled in terms of how to further effectively enhance the advantages of computers in data processing in innovative design and help designers form innovative solutions through effective management of innovative design source knowledge. With scientific methods, the most important source knowledge can be identified from the massive amount of available innovative knowledge, and designers can be provided with practical and more practical innovative knowledge, which can significantly improve the efficiency of product innovative design and has important practical significance.

[0003] At present, in the widespread practice of innovative design, in the process of selecting innovative knowledge, most experts in the field give subjective evaluations based on their professional knowledge and professional experience. Usually, the personal preferences of experts will affect the objectivity of the evaluation results to a certain extent. Especially when there are many evaluation indicators, the personal preferences of experts will be reflected in the differences in weights among various indicators, which will have an adverse effect on the objectivity of the indicators, and then cause significant deviations in the scientificity and objectivity of the indicator weight setting of the evaluation calculation. This deviation will be continuously amplified by the subsequent importance calculation process, which will have a harmful effect on the importance calculation results. In order to solve the current problem of excessive reliance on the personal preferences of scoring experts in the process of determining the weights of multi-dimensional value indicators in the process of innovative design, the deviation of knowledge importance calculation has been caused. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention proposes a knowledge importance ranking method for innovative design which can improve the accuracy of knowledge importance calculation.

[0005] The above-mentioned object of the present invention is achieved by the following technical solutions:

[0006] A knowledge importance ranking method for innovative design includes the following steps:

[0007] Step 1: Determine the indicators required for ranking knowledge importance;

[0008] Step 2: Determine the appropriate quantification criteria for fuzzy semantics and calculate the weighted average value of the fuzzy numbers;

[0009] Step 3: Identify the experts who will participate in the evaluation, mark their identities, and determine the weight of each expert;

[0010] Step 4: Experts compare each indicator pairwise and use a matrix composed of triangular fuzzy numbers to characterize the relative importance of different evaluation indicators;

[0011] Step 5: Calculate the triangular fuzzy number matrix to obtain the weight value of each indicator and perform normalization;

[0012] Step 6: Summarize the evaluation opinions of multiple experts and determine the weight of each evaluation indicator;

[0013] Step 7: The scoring experts give scores to each item of knowledge to be evaluated according to the degree of matching, and obtain specific scores for each indicator;

[0014] Step 8: Based on the weights of each evaluation indicator determined in step 6, the specific scores of each rating indicator determined in step 7, and the weights of each expert determined in step 3, the evaluation results are summarized and calculated, and the innovative knowledge is sorted according to the scores.

[0015] Moreover, in step 1, the specific contents of each indicator for the evaluation of the value of innovative knowledge in the innovative design are determined and used, and the total number of indicators is determined at the same time, and recorded as n;

[0016] Furthermore: In step 2, after calculating the parameter pair (a, b) in the fuzzy number weighted average Q, the weighted vector w of the operator is calculated to be (w1, w2, ..., w n ) T

[0017] w j =Q(j / n)-Q((j-1) / n),j∈N

[0018]

[0019] Moreover, in step 4, for one of the multivariate indicator problems, it is assumed that it contains n items, and the evaluation experts compare these n solutions pairwise under a certain standard to construct a triangular fuzzy complementary judgment matrix in, The triangular fuzzy number of each item There are three parts in it, which represent the lower boundary of the comparison of the importance of the two. median and upper boundary And satisfy the following relationship:

[0020]

[0021] Moreover, in step 4, 0.9 means that when comparing two by two, the former index i is extremely important compared with the latter index j; 0.8 means that when comparing two by two, the former index i is very important compared with the latter index j; 0.7 means that when comparing two by two, the former index i is significantly important compared with the latter index j; 0.6 means that when comparing two by two, the former index i is slightly important compared with the latter index j; 0.5 means that when comparing two by two, the former index i is equally important compared with the latter index j; 0.4 means that when comparing two by two, the latter index j is slightly important compared with the former index i; 0.3 means that when comparing two by two, the latter index j is significantly important compared with the former index i; 0.2 means that when comparing two by two, the latter index j is significantly important compared with the former index i; 0.1 means that when comparing two by two, the latter index j is extremely important compared with the former index i.

[0022] Moreover, in step 5, when calculating the matrix composed of triangular fuzzy numbers, the following five specific steps are specifically included:

[0023] 5.1 Calculate the matrix by the following formula All elements in row i of Expected value:

[0024]

[0025] 5.2 According to the size of the result calculated in step 5.1, the data Sort from largest to smallest and get

[0026] 5.3, based on the fuzzy number weighted average operator, the degree to which a certain indicator is stronger than others is obtained, denoted as

[0027] Among them, the fuzzy number weighted average operator is shown in the following formula:

[0028]

[0029] 5.4 Calculation The expected value of is as follows:

[0030]

[0031] 5.5, yes Normalize and get the final sorting vector ω=(ω1,ω2,…,ω n ) T ,in:

[0032]

[0033] Moreover, in step 6, the experts participating in the evaluation are set to be, and the comprehensive conclusion of the scores of k evaluation experts is calculated by the fuzzy number area ranking method, specifically:

[0034] 6.1. For each corresponding ω i The sorting vector values ​​are constructed as triangular fuzzy numbers and are expressed using B i =(d i ,g i ,f i ) indicates that They are respectively expressed as the most pessimistic evaluation, the most likely evaluation and the most optimistic evaluation;

[0035] 6.2: First determine k ranked fuzzy numbers (ω1,ω2,…,ω k ) is a closed interval [α,β] of real numbers, then the fuzziness of the triangular fuzzy number is Triangular fuzzy number ω i The area between the centroid and the origin in Its final ranking index Z i It is calculated by the following formula, where λ is taken as 0.5;

[0036]

[0037] Moreover, in step 7, according to the different importance of functions to users, the functions of the analysis object are divided into two categories: main functions and additional functions, among which the main functions are indispensable functions to realize the main needs of users, the functions that products must have, and the functions that directly interact with the products; the additional functions are functions added to the main functions to meet the diverse needs of users, and can be changed according to changes in user needs; according to the classification of functions, a functional feature hierarchy model of a system is established, in which the main functions and additional functions are located at the first level of the model; according to the functional characteristics in the innovative knowledge, the importance of the main functions and additional functions in the first level is ranked, and the scoring experts give scores to each piece of knowledge to be evaluated according to the degree of matching, and complete the evaluation of all knowledge.

[0038] Moreover, in step 8, the evaluation results of all evaluators are combined to calculate the average of multiple ratings. If there are k scoring experts participating in the evaluation, k groups of ranking vectors are obtained. When the number of k is large, the weight analysis of the scorers is introduced to calculate the final ranking vector:

[0039] Let the weight of the i-th rater be w i , while satisfying The final i-th weight result ω Fi It can be calculated by the following formula:

[0040]

[0041] The advantages and positive effects of the present invention are:

[0042] 1. The present invention proposes an innovative design process based on triangular fuzzy numbers to calculate and rank the importance of target knowledge, which makes up for the shortcomings of current methods, improves the accuracy of knowledge importance calculation, and fills the method gap in this application field.

[0043] 2. The present invention scientifically quantifies the subjective evaluations of multiple experts, thereby improving the objectivity of the evaluation results.

[0044] 3. The present invention forms a scientific method for evaluating multiple decision-making objectives.

[0045] 4. The present invention provides a reasonable and scientific evaluation method for conceptual design, which can better meet the needs of eliminating and compensating for uncertainties in innovative design. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of a knowledge importance ranking method for innovative design according to the present invention;

[0047] Figure 2 The present invention provides a knowledge importance ranking method for innovative design and various functional modules;

[0048] Figure 3 A characteristic analysis schematic diagram of an embodiment of a knowledge importance ranking method for innovative design according to the present invention;

[0049] Figure 4 This is a schematic diagram of a knowledge importance ranking method system for innovative design according to the present invention. DETAILED DESCRIPTION

[0050] The structure of the present invention is further described below with reference to the accompanying drawings and by way of examples. It should be noted that the present examples are descriptive rather than restrictive.

[0051] A knowledge importance ranking method for innovative design, see Figure 1-Figure 4 , the invention point is that it comprises the following steps:

[0052] Step 1: Determine the indicators required for ranking knowledge importance;

[0053] Step 2: Determine the appropriate quantification criteria for fuzzy semantics and calculate the weighted average value of the fuzzy numbers;

[0054] Step 3: Identify the experts who will participate in the evaluation, mark their identities, and determine the weight of each expert;

[0055] Step 4: Experts compare each indicator pairwise and use a matrix composed of triangular fuzzy numbers to characterize the relative importance of different evaluation indicators;

[0056] Step 5: Calculate the triangular fuzzy number matrix to obtain the weight value of each indicator and perform normalization;

[0057] Step 6: Summarize the evaluation opinions of multiple experts and determine the weight of each evaluation indicator;

[0058] Step 7: The scoring experts give scores to each item of knowledge to be evaluated according to the degree of matching, and obtain specific scores for each indicator;

[0059] Step 8: Based on the weights of each evaluation indicator determined in step 6, the specific scores of each rating indicator determined in step 7, and the weights of each expert determined in step 3, the evaluation results are summarized and calculated, and the innovative knowledge is sorted according to the scores.

[0060] In step 1, it is necessary to determine and use the specific content of each indicator for the evaluation of the value of innovative knowledge in the innovative design, and at the same time determine the total number of indicators, which is recorded as n;

[0061] In step 2, an appropriate fuzzy semantic quantization criterion is selected to obtain the parameter pair (a, b) in the fuzzy number weighted average Q. Then, the weighted vector of the operator is calculated using the formula w = (w1, w2, ..., w n ) T

[0062] w j =Q(j / n)-Q((j-1) / n),j∈N

[0063]

[0064] In step 3, multiple experts are selected to score the weight of each evaluation indicator, and the identity information of each scorer and the weight of each expert are recorded. The weight of each expert is mainly determined by the role of a single person, such as senior expert and general expert. If the expert group consists of three people, it is composed of one senior expert and two general experts. The weight of the senior expert is generally 0.4, and that of the general expert is 0.3.

[0065] In step 4, the experts compare the two options in pairs and use the matrix composed of triangular fuzzy numbers to characterize the relative importance of different evaluation indicators. For example, for a multivariate indicator problem, assuming that it contains n items, the evaluator compares the n options in pairs under a certain standard, and then constructs a triangular fuzzy complementary judgment matrix in, The triangular fuzzy number of each item There are three parts in it, which represent the lower boundary of the comparison of the importance of the two. median and upper boundary And satisfy the following relationship:

[0066]

[0067] In the method, it is further stipulated that: 0.9 means that when comparing two by two, the former index i is extremely important compared with the latter index j, 0.8 means that when comparing two by two, the former is very important compared with the latter, 0.7 means that when comparing two by two, the former is significantly important compared with the latter, 0.6 means that when comparing two by two, the former is slightly important compared with the latter, and 0.5 means that when comparing two by two, they are equally important; conversely, 0.4 means that when comparing two by two, the latter is slightly more important than the former, 0.3 means that when comparing two by two, the latter is significantly more important than the former, 0.2 means that when comparing two by two, the latter is significantly more important than the former, and 0.1 means that when comparing two by two, the latter is extremely important than the former.

[0068] In step 5, when calculating the matrix composed of triangular fuzzy numbers, the following five specific steps are specifically included:

[0069] 5.1. Calculate the matrix using the following formula All elements in row i of Expected value:

[0070]

[0071] 5.2. According to the size of the calculation results, the data Sort and get

[0072] 5.3. Based on fuzzy number weighted average operator:

[0073]

[0074] The degree to which a certain indicator is stronger than others can be recorded as It is also a triangular fuzzy number.

[0075] 5.4. Use the following formula to calculate Expected value:

[0076]

[0077] 5.5. Yes Normalize and get the final sorting vector ω=(ω1,ω2,…,ω n ) T ,in:

[0078]

[0079] In step 6, if there are k experts participating in the scoring, the comprehensive conclusion of the scores of the k experts can also be calculated by the fuzzy number area ranking method.

[0080] To sort k groups of sorted vectors using this method, we first need to sort each corresponding ω i The sorted vector values ​​are constructed as triangular fuzzy numbers using B i =(d i ,g i ,f i ) indicates that They are respectively represented as the most pessimistic evaluation, the most likely evaluation and the most optimistic evaluation.

[0081] First, determine k ranked fuzzy numbers (ω1,ω2,…,ω k ) is a closed interval [α,β] of real numbers, then the fuzziness of the triangular fuzzy number is Triangular fuzzy number ω i The area S(ω) formed between the center of mass and the origin i ), in Its final ranking index Z i It is calculated by the following formula, where λ is generally taken as 0.5.

[0082]

[0083] In step 7, if Figure 3 As shown in the figure, according to the different importance of functions to users, the functions of the analysis object are divided into two categories: main functions and additional functions. The main functions are the functions that are indispensable to realize the main needs of users, the functions that products must have, and the functions that directly interact with the products. The additional functions are functions added to the main functions to meet the diverse needs of users, and can be changed according to the changes in user needs. According to the classification of functions, a functional feature hierarchy model of a system can be established, in which the main functions and additional functions are located at the first level of the model. According to the functional characteristics in the innovative knowledge, the importance of the main functions and additional functions in the first level is ranked. The scorer needs to give a score to each knowledge to be evaluated according to the degree of matching to complete the evaluation of all knowledge.

[0084] In step 8, the evaluation results of all evaluators are combined to calculate the average of multiple ratings. If there are k experts participating in the evaluation, k groups of ranking vectors will be obtained in the end. When the number of k is large, the weight analysis of the evaluators can be introduced to calculate the final ranking vector. Let the weight of the i-th evaluator be w i Satisfy at the same time

[0085] The final i-th weight result ω Fi It can be calculated by the following formula:

[0086]

[0087] like Figure 4 As shown, the knowledge importance ranking method system for implementing the present invention for innovative design includes an operating system, an application program, an internal database, a processor, an internal memory and a display unit. The functional modules included in the application program are as follows: Figure 2 As shown: 201 a pre-set evaluation index system; 202 a triangular fuzzy number comparison matrix construction module; 203 a calculation module based on a weighted average operator of fuzzy numbers; 204 a weight calculation method module of a fuzzy number area ranking method; 205 a ranking algorithm module for scoring results of multiple raters.

[0088] When ranking the importance of innovative knowledge, the display unit provides a relevant information input interface, the input unit receives the input data, and the application program containing five functional modules processes the received data in turn, and saves the gradually processed intermediate data and analysis results into the internal database. When the calculation and sorting are completed, the output sorting results are saved in the internal memory and displayed on the display module.

[0089] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is limited to the above embodiment. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A knowledge importance ranking method for innovative design, characterized by: The following steps are involved: Step 1: Determine the indicators required for ranking knowledge importance; Step 2: Determine the appropriate quantification criteria for fuzzy semantics and calculate the weighted average value of the fuzzy number; Step 3: Identify the experts who will participate in the evaluation, mark their identities, and determine the weight of each expert; Step 4: Experts compare each indicator pairwise and use a matrix composed of triangular fuzzy numbers to characterize the relative importance of different evaluation indicators; Step 5: Calculate the triangular fuzzy number matrix to obtain the weight value of each indicator and perform normalization; Step 6: Summarize the evaluation opinions of multiple experts and determine the weight of each evaluation indicator; Step 7: The scoring experts give scores to each item of knowledge to be evaluated according to the degree of matching, and obtain specific scores for each indicator; Step 8: Based on the weights of each evaluation indicator determined in step 6, the specific scores of each rating indicator determined in step 7, and the weights of each expert determined in step 3, the evaluation results are summarized and calculated, and the innovative knowledge is sorted according to the scores.

2. The knowledge importance ranking method for innovative design according to claim 1, characterized in that: In step 1, the specific contents of various indicators for the evaluation of the value of innovative knowledge in the innovative design are determined and used, and the total number of indicators is determined at the same time, and recorded as n.

3. The knowledge importance ranking method for innovative design according to claim 1 is characterized by: In step 2, after calculating the parameter pair (a, b) in the fuzzy number weighted average Q, the weighted vector w of the operator is calculated to be (w1, w2, …, w n ) T w j =Q(j / n)-Q((j-1) / n),j∈N 4. The knowledge importance ranking method for innovative design according to claim 1 is characterized by: In step 4, for one of the multivariate indicator problems, it is assumed that it contains n items. The evaluation experts compare these n options in pairs under a certain standard and construct a triangular fuzzy complementary judgment matrix in, The triangular fuzzy number of each item There are three parts in it, which represent the lower boundary of the comparison of the importance of the two. median and upper boundary And satisfy the following relationship:

5. The knowledge importance ranking method for innovative design according to claim 3 is characterized in that: In step 4, n solutions are compared pairwise, and the ratios are as follows: 0.9 means that when comparing two items, the former indicator i is extremely important compared with the latter indicator j; 0.8 means that when comparing two items, the former indicator i is very important compared with the latter indicator j; 0.7 means that when compared pairwise, the former indicator i is significantly more important than the latter indicator j; 0.6 means that when compared pairwise, the former indicator i is slightly more important than the latter indicator j; 0.5 means that when compared pairwise, the former indicator i is equally important than the latter indicator j; 0.4 means that when compared pairwise, the latter indicator j is slightly more important than the former indicator i; 0.3 means that when compared pairwise, the latter indicator j is significantly more important than the former indicator i; 0.2 means that when compared pairwise, the latter indicator j is significantly more important than the former indicator i; 0.1 means that when compared pairwise, the latter indicator j is extremely important compared to the former indicator i.

6. The knowledge importance ranking method for innovative design according to claim 1 is characterized by: In step 5, when calculating the matrix composed of triangular fuzzy numbers, the following five specific steps are specifically included: 5.1 Calculate the matrix by the following formula All elements in row i of Expected value: 5.2 According to the size of the result calculated in step 5.1, the data Sort and get 5.3, based on the fuzzy number weighted average operator, the degree to which a certain indicator is stronger than others is obtained, denoted as Among them, the fuzzy number weighted average operator is shown in the following formula: 5.4 Calculation The expected value of is as follows: 5.5, yes Normalize and get the final sorting vector ω=(ω1,ω 2, …,ω n ) T ,in:

7. The knowledge importance ranking method for innovative design according to claim 1 is characterized in that: In step 6, the experts participating in the evaluation are set to be, and the comprehensive conclusion of the scores of k evaluation experts is calculated by the fuzzy number area ranking method, specifically: 6.

1. For each corresponding ω i The sorting vector values ​​are constructed as triangular fuzzy numbers and are expressed using B i =(d i ,g i ,f i ) indicates that They are respectively expressed as the most pessimistic evaluation, the most likely evaluation and the most optimistic evaluation; 6.2 Calculation of ranking index: First, determine k ranking fuzzy numbers (ω1,ω2,…,ω k ) is a closed interval [α,β] of real numbers, then the fuzziness of the triangular fuzzy number is Triangular fuzzy number ω i The area between the centroid and the origin in Its final ranking index Z i It is calculated by the following formula, where λ is taken as 0.5; 8. The knowledge importance ranking method for innovative design according to claim 1 is characterized by: In step 7, according to the different importance of functions to users, the functions of the analysis object are divided into two categories: main functions and additional functions. The main functions are the functions that are indispensable to realize the main needs of users, the functions that products must have, and the functions that directly interact with the products; the additional functions are functions added to the main functions to meet the diverse needs of users, and can be changed according to changes in user needs; according to the classification of functions, a functional feature hierarchy model of a system is established, in which the main functions and additional functions are located at the first level of the model; according to the functional characteristics in the innovative knowledge, the importance of the main functions and additional functions in the first level is ranked, and the scoring experts give scores to each piece of knowledge to be evaluated according to the degree of matching, and complete the evaluation of all knowledge.

9. The knowledge importance ranking method for innovative design according to claim 1 is characterized in that: In step 8, the evaluation results of all evaluators are combined to calculate the average of multiple ratings. If there are k scoring experts participating in the evaluation, k groups of ranking vectors are obtained. When the number of k is large, the weight analysis of the scorers is introduced to calculate the final ranking vector: Let the weight of the i-th rater be w i， Satisfy at the same time The final i-th weight result ω Fi It can be calculated by the following formula:

Images