A Method for Calculating the Importance of Product Quality Characteristics Based on Logical Similarity under Multi-Granularity Fuzzy Uncertainty

Abstract

This invention belongs to the field of product structure design and discloses a method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty, including the following steps: Step 1, obtaining the importance of customer needs under multi-granularity fuzzy language evaluation, specifically including: Step 1.1, calculating the fine-grained customer importance reflecting logical similarity; Step 1.2, calculating the importance of customer needs based on grey relational analysis; Step 2, obtaining the importance of quality characteristics oriented towards customer needs, specifically including: Step 2.1, obtaining the relationship matrix between customer needs and quality characteristics; Step 2.2, determining the degree of influence of quality characteristics on customer needs; Step 2.3, calculating the importance of quality characteristics under the condition that customer needs are met.

Description

Technical Field

[0001] This invention belongs to the field of product structure design, specifically a method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty. Background Technology

[0002] Integrating customer needs into every stage of the structural design process is the foundation and prerequisite for ensuring the rationality of complex product structural designs. Product quality characteristics are a multifaceted reflection of the perception and requirements for product quality. For example, for a hydraulic press, weight, power, energy consumption, and price are all quality characteristics. The main purpose of conducting customer needs analysis in the early stages of product design is to map these needs into product quality characteristics that can guide design and development, serving as driving factors for solving the product's functional structure. Simply controlling every product quality characteristic within acceptable limits clearly cannot achieve optimal product structural design. It is necessary to prioritize and classify product quality characteristics based on customer needs data, accurately extract key quality characteristics that have a significant impact on product quality, and clarify the product's quality and functional requirements. Only then can the subsequent solution for the product's functional structure be guided.

[0003] In the early stages of actual product design, customers have limited understanding of product functions and quality, and product information is often vague and imprecise, exhibiting a significant degree of uncertainty. Customers tend to express their product needs qualitatively using vague language. For example, for electric motors, customer needs might include high power, high efficiency, low price, and light weight. The importance of product quality characteristics can be derived from the importance of customer needs and the correlation between customer needs and quality characteristics. A comprehensive and objective determination of customer need importance is necessary to accurately obtain the importance of quality characteristics and perform precise quality characteristic extraction. A common approach is to invite several customers to provide assessments of the importance of their needs, and then aggregate all customer assessment data based on these assessments to calculate the final customer need importance. Early methods assigned the same importance to a particular customer across all customer needs, essentially assuming that each customer had the same level of importance for all needs. However, because customer groups may have different knowledge structures regarding various customer needs, different customers may have varying levels of understanding of a particular need, and the reliability of data from different customers may also differ across needs, fine-grained customer importance should be calculated separately for each customer need.

[0004] Meanwhile, most customers are non-professionals, lacking accurate self-awareness and unable to clearly assess their understanding of each customer need. However, for the target product, there are usually several similar products that all customers are familiar with. Customers can easily provide vague evaluations of their satisfaction with these similar products based on their own understanding, but the granularity of these evaluations may vary. In other words, customers often use multi-granularity vague language to express their satisfaction with similar products. For example, customers with extensive product knowledge tend to use finer-grained vague language to express their satisfaction with similar products, while those with less product knowledge tend to use coarser-grained vague language. In practice, different customers should be allowed to use a granularity of vague language suitable for their own circumstances to express the information they want to provide. In fact, although the multi-granularity vague language evaluations given by different customers regarding similar products differ at the granularity level, they share a potential similarity at the logical order level. The corresponding dominant granular structure serves as the carrier for conveying this logical similarity. Customers with high logical similarity to other customers are more likely to reflect the true thoughts of the customer group, their data is more reliable, and they should have higher importance. Conversely, customers with low logical similarity to other customers are less likely to reflect the true thoughts of the customer group, their data is less reliable, and they should have lower importance. If we can start from this type of multi-granularity fuzzy language evaluation, calculate the fine-grained customer importance under each customer need, and then aggregate the customer need data, we can improve the accuracy of calculating the importance of customer needs and quality characteristics. This will enable precise extraction of product quality characteristics, making the product's quality and functional requirements clearer, and laying the foundation for subsequent product functional structure solving. Summary of the Invention

[0005] To address the aforementioned technical problems in the existing technology, this invention provides a method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty, comprising the following steps:

[0006] Step 1: Obtain the importance of customer needs under multi-granularity fuzzy language evaluation, specifically including:

[0007] Step 1.1: Calculate the fine-grained customer importance score, which reflects logical similarity;

[0008] Step 1.2 Calculate the importance of customer needs in grey relational analysis;

[0009] Step 2: Obtain the importance of quality characteristics related to customer needs, specifically including:

[0010] Step 2.1: Obtain the relationship matrix between customer needs and quality characteristics;

[0011] Step 2.2 Determine the extent to which quality characteristics affect customer needs;

[0012] Step 2.3: Calculate the importance of quality characteristics under the condition of meeting customer requirements.

[0013] Furthermore, in step 1.1, assume customer E k (k=1, 2, ..., m) in customer requirement item c j The dominant granular structure induced by multi-granularity fuzzy language evaluation is Calculate the pairwise customer requirements for the target customer item c. j Let SM be the logical similarity matrix of the dominant granule structure. j :

[0014] (2-5)

[0015] In customer requirement item c j Customer E k The importance of fine-grainedness is:

[0016] (2-6)

[0017] in To normalize the coefficients, Standardize to [0, 1] and satisfy This means that the sum of the importance of all fine-grained customers under a single customer need is 1.

[0018] Furthermore, in step 1.2, the grey relational coefficient of the j-th customer requirement under the evaluation of the k-th customer is:

[0019] (2-12)

[0020] in It is the distinguishing factor, and This article assumes It is 0.5, and we have:

[0021] (2-13)

[0022] (2-14)

[0023] Find the customer requirement item c. j Customer E k Importance is By combining grey relational analysis of customer needs, the final importance of customer needs is calculated. for:

[0024] (2-15)

[0025] in,

[0026] (2-16).

[0027] Furthermore, in step 2.1, these h experts are invited to assess the degree of interaction between product quality characteristics and customer needs, and the opinions of all experts in the expert team on product quality characteristics EC are summarized. j Based on the interval evaluation information, the expert team provided the product quality characteristics EC. j The extent of the impact on various customer needs.

[0028] Furthermore, in step 2.2, the product quality characteristic EC can be obtained. j Customer needs C i The degree of influence is set as (j=1, 2, …, s; i=1, 2, …, n), then we have:

[0029] (2-32)

[0030] in ZJ, an expert in the QFD expert team k Importance, k=1, 2, …, h;

[0031] Let product quality characteristics EC j The overall impact on all customer needs is If j = 1, 2, …, s, then we have:

[0032] (2-33)

[0033] in For customer needs C i Importance.

[0034] Furthermore, in step 2.3, let the quality characteristic EC j The overall impact on all customer needs The simplified interval number is:

[0035] (2-35)

[0036] in,

[0037] (2-36)

[0038] (2-37)

[0039] Let the interval roughness number be... The interval roughness ratio is:

[0040] (2-38)

[0041] Quality characteristics EC can be obtained j The overall impact on all customer needs The simplified real number is: (2-39)

[0042] Product quality characteristics EC based on QFD analysis j The importance of standardized real numbers is:

[0043] (2-40)

[0044] Where j = 1, 2, …, s, and s is the total number of quality characteristics.

[0045] This invention maps customer needs to product quality characteristics, which helps to clarify the product's quality and functional requirements. Extracting key quality characteristics that have a significant impact on product quality from numerous product quality characteristics is the starting point for ensuring and improving product quality, and is conducive to improving product performance. Starting from customers' multi-granularity fuzzy language evaluation of similar products, this invention analyzes the logical similarity between customers and proposes a precise extraction technology for product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty. This improves the accuracy and convenience of customer need data aggregation, and the rationality and effectiveness of the proposed technology are verified through engineering examples. Attached Figure Description

[0046] Figure 1 This is a schematic diagram illustrating the relationship between customer needs and product quality characteristics. Detailed Implementation

[0047] The invention will now be further described with reference to the accompanying drawings.

[0048] The present invention provides a method for calculating product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty. First, it analyzes customers' multi-granularity fuzzy language evaluations of similar products, introduces a dominant granular structure to mine the logical similarity between customers, calculates the fine-grained customer importance under various customer needs, and combines grey relational analysis to consider the mutual influence between customer needs, accurately aggregating customer need data. Based on this, it uses IRN to perform product QFD to obtain the importance of quality characteristics oriented towards customer needs. Specifically, it includes the following steps:

[0049] Step 1: Obtain the importance of customer needs under multi-granularity fuzzy language evaluation, including the following steps:

[0050] Step 1.1: Calculate the fine-grained customer importance score, which reflects logical similarity;

[0051] Let m be the number of customers invited (E1, E2, ..., E...). m For a set of similar products A = {a1, a2, …, a} N The products in} are designed to meet various customer needs (C1, C2, ..., C n A satisfaction evaluation is conducted, where a1, a2, ..., a N This represents similar products within the same category, where N is the total number of similar products. Consider the common fuzzy logic of 3-granularity, 5-granularity, 7-granularity, and 9-granularity, as shown in Table 2.1. Customer E k The satisfaction ratings for similar products are shown in Table 2.2, where k=1, 2, …, m. Representing Client E k In response to customer needs C j Given similar products a i Satisfaction evaluation.

[0052] Table 2.1 Granularity of Fuzzy and Uncertain Language

[0053]

[0054] Table 2.2 Customer E k Satisfaction evaluation of similar products

[0055]

[0056] Definition 1: Given a set A of similar products, and considering various customer needs, each customer uses ordinal numbers to express their preferences for these similar products. Customer ordinal preferences are determined by a ranking function O(a). i )=o i The form of expression, where o i Indicates similar products of the same type. i sequence.

[0057] (1) o i < o j This indicates that, in response to customer needs for the target item, the customer has a preference for similar products (a). i The satisfaction rate is higher than that of similar products. j Satisfaction level;

[0058] (2) o i = o j This indicates that, regarding a specific customer need, the customer has a preference for similar products (a). i Satisfaction is equal to its satisfaction with similar products of the same type. j Satisfaction level;

[0059] (3) o i >o j This indicates that, regarding a specific customer need, the customer has a preference for similar products (a). i The satisfaction rate was lower than that of similar products. j Satisfaction level.

[0060] Based on the multi-granularity fuzzy language evaluation provided by customers, we can obtain the advantageous granular structure of a set A of similar products for different customer needs:

[0061] (2-1)

[0062] in It refers to similar products of the same type under the specified customer requirements. i The superior particle structure, given the specific customer needs, results in customer satisfaction that is higher than or equal to that of similar products. i It consists of all similar products of the same kind.

[0063] Because of differing knowledge backgrounds, clients may use various granular fuzzy languages ​​to evaluate expressions related to set A = {a1, a2, ..., a...} N The degree of preference for the products listed. Let the preference value be... Based on Obtain similar products a from category A i The dominant particle structure.

[0064] Definition 2: Given a similar product a i , For similar products a i Regarding the order under the evaluation function E The dominant particle, the set of similar products A, under the evaluation function E, with respect to order The dominant particle structure is:

[0065] (2-2)

[0066] Definition 3: Let the preference value be E(a) i )=v i ∈V, by performing a fuzzy dominance comparison on the elements in V, v can be calculated. i Better than v j The probability of p ij = p(v i , v j ), where p ij ∈[0, 1]. Therefore, let Indicates similar products of the same type. i Compared to similar products, a jLet P be the dominance degree, and P be the dominance relation matrix on the set of similar products A. Then we have:

[0067] (2-3)

[0068] Definition 4: If ,but For similar products Fuzzy dominant particles, It is a fuzzy set. The fuzzy dominance granular structure on the set of similar products A is denoted as . .

[0069] Definition 2.5: Let P1 and P2 be sets of similar products of the same class. The two preorder relations above, and the corresponding dominant particle structures are respectively and If P1 and P2 are not both global relations, then and The similarity is:

[0070] (2-4)

[0071] If P1 and P2 are a global relation, then it is stipulated that... .

[0072] Assume customer E k (k=1, 2, ..., m) in customer requirement item c j The dominant granular structure induced by multi-granularity fuzzy language evaluation is Calculate the pairwise customer requirements for the target customer item c. j Let SM be the logical similarity matrix of the dominant granule structure. j .

[0073] (2-5)

[0074] Customers with high logical similarity to other customers can reflect the true thoughts of the customer group to a greater extent and should be given higher importance; conversely, customers with low logical similarity to other customers should be given lower importance. Therefore, in customer need item c... j Customer E k The importance of fine-grainedness is:

[0075] (2-6)

[0076] in For the normalization coefficient, it can be... Standardize to [0, 1] and satisfy This means that the sum of the importance of all fine-grained customers under a single customer need is 1.

[0077] Step 1.2 Calculate the importance of customer needs in grey relational analysis;

[0078] After obtaining the fine-grained customer importance under different customer needs, it is necessary to calculate the customer need importance. Customer need importance is a crucial basis for mapping customer needs to product quality characteristics. Let there be m customers (E1, E2, ..., E...). m For n customer needs (C1, C2, ..., C...), n Importance assessment is then performed. If fuzzy language is used to map to fuzzy numbers for quantitative calculation, it's necessary to standardize the granularity of the fuzzy language and determine the membership degree of the fuzzy numbers, which is very difficult for non-professional clients. Therefore, clients choose to use the interval model, which is the most flexible and convenient way to express uncertainty, to represent their perception of the importance of their needs. The interval number is set in the range [0, 8], with larger values ​​indicating greater importance. The client group's assessment matrix of client need importance is shown in Table 2.3. Meanwhile, another common problem with most current methods for determining the importance of client needs is the lack of necessary consideration and corresponding processing of the mutual influence between client needs, which can affect the aggregation effect of client need information to some extent. Therefore, grey relational analysis is used to take into account the mutual influence between client needs.

[0079] Table 2.3 Customer Groups' Evaluation of the Importance of Customer Needs

[0080]

[0081] Where a ij * = [a ij - ,a ij + ], where a ij - With a ij + These are the interval numbers a ij * The lower and upper bounds of a, and a ij - ≤a ij + , when a ij - = a ij + When the interval number becomes a real number, it can be considered a special case of the interval number. This requires the use of arithmetic operations on interval numbers and the conversion to real numbers. Here, we assume that the interval numbers are independent and unrelated, because customer satisfaction ratings for the product are independent of each other.

[0082] Let [x1, y1] and [x2, y2] be two interval numbers. The arithmetic operations on interval numbers are shown in formulas (2-7)-(2-10). Let G be the real value corresponding to the interval number G*=[x, y]. The realization operation is shown in formula (2-11).

[0083] (2-7)

[0084] (2-8)

[0085] (2-9)

[0086] (2-10)

[0087] (2-11)

[0088] Where s is the prudence coefficient. Let s = 0.5.

[0089] Convert all the customer need importance intervals given by customers in Table 2.3 into real numbers, and let a be an integer. ij * The corresponding real number is a ij Then, grey relational analysis is performed to calculate the grey relational coefficients for each customer requirement. The characteristic sequence of customer requirements is represented as U0={u o1 , u o2 , …, u om}, where u oi =max{a i1 , a i2 , …, a in (i=1, 2, …, m, where m is the total number of customers). The sequence of actions for the k-th customer demand is denoted as U. K ={u k1 , u k2 , …, u km}, where u ki =a ki (k=1, 2, …, n, where n is the total number of customer demand types). Then, the grey relational coefficient of the j-th customer demand under the evaluation of the k-th customer is:

[0090] (2-12)

[0091] in It is the distinguishing factor, and ,set up It is 0.5, and we have:

[0092] (2-13)

[0093] (2-14)

[0094] Find the customer requirement item c. j Customer E k Importance is By combining grey relational analysis of customer needs, the final importance of customer needs can be calculated. for:

[0095] (2-15)

[0096] in,

[0097] (2-16).

[0098] Step 2: Obtain the importance of quality characteristics that are relevant to customer needs;

[0099] After determining the importance of customer needs in step 1, the importance of customer needs oriented towards specific customer needs is obtained based on the relationship matrix between customer needs and product quality characteristics in the House of Quality (HOQ). For example... Figure 1 As shown, the elements in the relationship matrix represent the degree of mutual influence between customer needs and product quality characteristics; the left side of the relationship matrix represents the customer needs C that are of greatest concern to the customer. i and its importance Where i = 1, 2, …, n, and n is the total number of customer requirements; the top of the relation matrix represents the product quality characteristics EC. j The values ​​below the relation matrix represent the importance values ​​of the product quality characteristics that need to be solved. , where j=1, 2, …, s, and s is the total number of product quality characteristics.

[0100] Step 2.1: Obtain the relationship matrix between customer needs and quality characteristics;

[0101] The experts in the product QFD expert team are referred to as ZJ. t (t=1, 2, …, h), where h represents the total number of experts in the QFD team. Because the expert team has an accurate understanding of its own knowledge structure, the importance vector of the expert team can be predicted in advance and treated as a known quantity. Furthermore, since expert members all possess certain professional capabilities, a unified granularity of expert fuzzy language can be achieved. Assume that the importance vector of the expert team in QFD-guided evaluation is... These h experts were invited to evaluate the degree of interaction between product quality characteristics and customer needs, mapping "not important," "relatively unimportant," "moderately important," "relatively important," "important," and "very important" to 1, 2, 3, 4, 5, 6, and 7 respectively. A higher score indicates a closer relationship between the quality characteristic and the customer need, and a greater degree of influence of the quality characteristic on the customer need. Since expert evaluations are difficult to pinpoint precisely as single-point real numbers, they are also expressed as interval numbers. Let there be n customer needs and s quality characteristics, then we can obtain the expert evaluation scores. t The given interval relationship matrix between customer needs and quality characteristics is shown in Table 2.4.

[0102] Table 2.4 Expert ZJ t The given matrix shows the relationship between customer needs and quality characteristics.

[0103]

[0104] in , where n represents the total number of customer requirements and s represents the total number of quality characteristics.

[0105] Summarize the opinions of all experts in the expert team regarding the product quality characteristics EC j Based on the interval evaluation information, the expert team provided the product quality characteristics EC. j The degree of impact on various customer needs is shown in Table 2.5, j=1, 2, …, s.

[0106] Table 2.5 Quality Characteristics EC Given by the Expert Team j The extent of the impact on customer demand

[0107]

[0108] Step 2.2 Determine the extent to which quality characteristics affect customer needs;

[0109] Let product quality characteristics EC j Customer needs C i The degree of influence is Let j = 1, 2, …, s, i = 1,2, …, n. To facilitate explanation of the calculation process, the product quality characteristics EC provided by the expert team are summarized below. j Customer needs C i The degree of influence is as follows: , ... ... Substitute the interval number Write them as , ... ... IRN is a novel data format that combines interval numbers and rough sets, retaining the expressive flexibility of interval numbers and the computational simplicity of rough sets. It is gaining increasing attention in uncertain data scenarios. This paper describes how expert-provided interval evaluation data can be converted into IRN for calculating the importance of product quality characteristics.

[0110] Assume the quality characteristics EC given by the expert team j Customer needs C i The lower and upper boundary sets for evaluating the degree of influence are respectively and Then we have:

[0111] (2-17)

[0112] (2-18)

[0113] lower boundary The lower approximation limit can be expressed as:

[0114] (2-19)

[0115] lower boundary The upper approximation limit can be expressed as:

[0116] (2-20)

[0117] Similarly, the upper boundary can be obtained. The lower approximation limit can be expressed as:

[0118] (2-21)

[0119] upper boundary The upper approximation limit can be expressed as:

[0120] (2-22)

[0121] Then the lower boundary can be obtained. The lower bound of the rough number can be expressed as:

[0122] (2-23)

[0123] lower boundary The upper limit of the roughness number can be expressed as:

[0124] (2-24)

[0125] in and They are the lower approximation limits. The number of elements contained in the approximate bound set.

[0126] Similarly, the upper boundary can be obtained. The lower bound of the rough number can be expressed as:

[0127] (2-25)

[0128] lower boundary The upper limit of the roughness number can be expressed as:

[0129] (2-26)

[0130] in and Approximate limits and The number of elements in the array.

[0131] Set the boundary The rough boundary is:

[0132] (2-27)

[0133] upper boundary The rough boundary is:

[0134] (2-28)

[0135] lower boundary The interval is approximated as:

[0136] (2-29)

[0137] upper boundary The interval is approximated as:

[0138] (2-30)

[0139] Let expert ZJ k The given quality characteristics EC j Customer needs C i The degree of influence is the interval roughness number (j=1, 2, …, s; i=1, 2, …, n), then we have:

[0140] (2-31)

[0141] Where k = 1, 2, …, h, h is the total number of experts in the expert team; j = 1, 2, …, s, s is the total number of quality characteristics; i = 1, 2, …, n, n is the total number of customer requirements.

[0142] Based on the combined opinions of all experts in the expert team, the product quality characteristic EC can be obtained. j Customer needs C i The degree of influence is set as (j=1, 2, …, s; i=1, 2, …, n), then we have:

[0143] (2-32)

[0144] in ZJ, an expert in the QFD expert team k Importance, k=1, 2, …, h.

[0145] Let product quality characteristics EC j The overall impact on all customer needs is If j = 1, 2, …, s, then we have:

[0146] (2-33)

[0147] in For customer needs C i Importance.

[0148] Step 2.3: Calculate the importance of quality characteristics under the condition of meeting customer requirements;

[0149] Following formula (2-31), the quality characteristic EC j The overall impact on all customer needs Represented as:

[0150] (2-34)

[0151] Let the mass characteristic EC j The overall impact on all customer needs The simplified interval number is:

[0152] (2-35)

[0153] in,

[0154] (2-36)

[0155] (2-37)

[0156] Let the interval roughness number be... The interval roughness ratio is:

[0157] (2-38)

[0158] Quality characteristics EC can be obtained j The overall impact on all customer needs The simplified real number is: (2-39)

[0159] Product quality characteristics EC based on QFD analysis j The importance of standardized real numbers is:

[0160] (2-40)

[0161] Where j = 1, 2, …, s, and s is the total number of quality characteristics.

Claims

1. A method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty, comprising the following steps: Step 1: Obtain the importance of customer needs under multi-granularity fuzzy language evaluation, specifically including: Step 1.1: Calculate the fine-grained customer importance score, which reflects logical similarity; Step 1.2 Calculate the importance of customer needs in grey relational analysis; Step 2: Obtain the importance of quality characteristics related to customer needs, specifically including: Step 2.1: Obtain the relationship matrix between customer needs and quality characteristics; Step 2.2 Determine the extent to which quality characteristics affect customer needs; Step 2.3: Calculate the importance of quality characteristics under the condition of meeting customer requirements.

2. The method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty as described in claim 1, characterized in that: In step 1.1, assume customer E k (k=1, 2, ..., m) in customer requirement item c j The dominant granular structure induced by multi-granularity fuzzy language evaluation is Calculate the pairwise customer requirements for the target customer item c. j Let SM be the logical similarity matrix of the dominant granule structure. j : (2-5) In customer requirement item c j Customer E k The importance of fine-grainedness is: (2-6) in To normalize the coefficients, Standardize to [0, 1] and satisfy This means that the sum of the importance of all fine-grained customers under a single customer need is 1.

3. The method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty as described in claim 1, characterized in that: In step 1.2, the grey relational coefficient of the j-th customer requirement under the evaluation of the k-th customer is: (2-12) in It is the distinguishing factor, and This article assumes It is 0.5, and we have: (2-13) (2-14) Find the customer requirement item c. j Customer E k Importance is By combining grey relational analysis of customer needs, the final importance of customer needs is calculated. for: (2-15) in, (2-16)。 4. The method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty as described in claim 1, characterized in that: In step 2.1, these h experts are invited to assess the degree of interaction between product quality characteristics and customer needs, and the opinions of all experts in the expert team on product quality characteristics EC are summarized. j Based on the interval evaluation information, the expert team provided the product quality characteristics EC. j The extent of the impact on various customer needs.

5. The method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty as described in claim 4, characterized in that: In step 2.2, the product quality characteristic EC can be obtained. j Customer needs C i The degree of influence is set as (j=1, 2, …, s; i=1, 2, …, n), then we have: (2-32) in ZJ, an expert in the QFD expert team k Importance, k=1, 2, …, h; Let product quality characteristics EC j The overall impact on all customer needs is If j = 1, 2, …, s, then we have: (2-33) in For customer needs C i Importance.

6. The method for calculating the importance of product quality characteristics based on logical similarity under multi-granularity fuzzy uncertainty as described in claim 5, characterized in that: In step 2.3, let the mass characteristic EC be... j The overall impact on all customer needs The simplified interval number is: (2-35) in, (2-36) (2-37) Let the interval roughness number be... The interval roughness ratio is: (2-38) Quality characteristics EC can be obtained j The overall impact on all customer needs The simplified real number is: (2-39) Product quality characteristics EC based on QFD analysis j The importance of standardized real numbers is: (2-40) Where j = 1, 2, …, s, and s is the total number of quality characteristics.

Images