CN111832905B - Method for identifying interaction association relation between related service demands of products - Google Patents

Method for identifying interaction association relation between related service demands of products Download PDF

Info

Publication number
CN111832905B
CN111832905B CN202010568648.4A CN202010568648A CN111832905B CN 111832905 B CN111832905 B CN 111832905B CN 202010568648 A CN202010568648 A CN 202010568648A CN 111832905 B CN111832905 B CN 111832905B
Authority
CN
China
Prior art keywords
influence
related service
degree
product
value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010568648.4A
Other languages
Chinese (zh)
Other versions
CN111832905A (en
Inventor
黄琳
宋文燕
王丽亚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Jiaotong University
Original Assignee
Shanghai Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Jiaotong University filed Critical Shanghai Jiaotong University
Priority to CN202010568648.4A priority Critical patent/CN111832905B/en
Publication of CN111832905A publication Critical patent/CN111832905A/en
Application granted granted Critical
Publication of CN111832905B publication Critical patent/CN111832905B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Computational Mathematics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Mathematical Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Algebra (AREA)
  • Development Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Educational Administration (AREA)
  • Software Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Computing Systems (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A method for identifying interaction association relation among related service demands of products comprises the following steps: step 1: establishing a group rough cloud model of the influence degree among the related service demands of the products; and 2, step: determining a numerical value of the degree of influence between the product-related service demands; and 3, step 3: ordering the product related service requirements based on a DEMATEL method. The beneficial effects of the invention are: allowing each evaluator to freely give their judgment and evaluation using the form of interval language value; the advantages of the DEMATEL method, the cloud model theory and the rough set theory are combined; the cloud model theory is used for processing the uncertain performance of personal evaluation, so that the use defect of the fuzzy set theory is effectively avoided; the subjective value and the objective value of the influence degree are comprehensively considered, and the influence degree between the related service requirements of the product can be more comprehensively revealed.

Description

Method for identifying interaction association relation between related service demands of products
Technical Field
The invention relates to a method for designing and analyzing product related services, identifying and calculating interactive association relation among requirements.
Background
Objective facts 1: the related services offered around the product are becoming more and more important.
Since the 20 th century 90 s, with the gradual blurring of the boundaries between tangible products and intangible services, enterprises begin to provide more comprehensive and personalized product-related services centered on customers, even to realize service transformation. Practice results also show that related services provided around the product can help effectively improve customer satisfaction, improve customer loyalty, enhance market competitiveness of the product, achieve business expansion of enterprises, and finally bring sustainable profit growth.
Objective facts 2: product-related service requirements are diverse, but enterprises cannot meet all the requirements, and the requirements need to be prioritized.
In the current user-centric era, customer demand is a fundamental input to all designs. In practice, the demand of customers for product extension service is often diversified and large, but due to practical constraints on enterprise resources, funds and the like, enterprises often cannot meet all the product extension service demands. Therefore, it is important to reasonably analyze the importance of the service requirements related to each product for developing a successful product-related service.
Objective facts 3: there is an inherent inter-relationship between these requirements.
Research has shown that: product extension service requirements are often interrelated. In other words, the fulfillment of certain requirements may affect the fulfillment of other requirements. Taking elevator product customers as an example, the product extension service requirements of the elevator product customers comprise:
(1) monitoring the running state in real time;
(2) the maintenance is reliable and timely;
(3) training elevator operation at regular intervals;
(4) the safe and efficient operation of the elevator is ensured.
If designers do not consider interdependencies between product-related service requirements, they may find that requirement (4) is of the greatest interest to the customer. In practice, the requirement (4) is affected by the requirements (1), (2) and (3), namely: if the first three requirements are well met, then (4) is naturally also achieved. Thus, in this case, the design priority of the demands (1), (2), and (3) should be higher than the demand (4).
Objective facts 4: there is a need to determine and identify the degree of impact between these needs, and provide better and more accurate product-related services with limited resources.
The interaction influence relationship of the related service requirements of the products is reasonably evaluated, analyzed and managed, the key related service requirements of the products are identified, designers can be helped to more reasonably configure design resources, better related service of the products is provided for customers, and the method has very important academic research significance and social practice value.
Objective facts 5: such related researches are few, subjective judgment on the influence degree needs to be given artificially in the analysis process, and the existing method has functional defects in the analysis of the subjective opinions.
Currently, there is little research on product-related service requirement interaction. In addition, a lot of fuzzy, subjective and uncertain personal perception and subjective judgment are often involved in the process of evaluating the demand, which all result in inaccurate final analysis results. The existing method has certain functional defects when the subjective viewpoints are analyzed.
The DEMATEL method is known as a visualization method that can reveal the causal relationship between elements of a complex system, as compared with methods such as AHP, ANP, IPA, and DEA. Thus, the DEMATEL method is also suitable for processing and analyzing complex interactions between product-related service requirements. While many scholars have used fuzzy set theory (fuzzy set theory) or rough set theory (rough set theory) in conjunction with the DEMATEL method for dealing with ambiguity, subjectivity and uncertainty of opinions, those methods still have many limitations:
(1) there is no simultaneous handling of personal evaluation uncertainty (which refers to the ambiguity and uncertainty of a person in thinking and expression preference for a certain evaluation item) and interpersonal evaluation uncertainty (which refers to the uncertainty of different judgments made for the same evaluation item due to differences in knowledge accumulation, working experience, etc. of different persons), or there is no inconsistency between measurement group decision makers, or there is no reflection in the final evaluation result.
(2) Using fuzzy set theory requires designers to set a large amount of prior information in advance to describe the degree of ambiguity, such as preset fuzzy membership functions, data distributions, fuzzy rules, etc. For designers, the requirement of professional mathematical knowledge is high, the calculation workload is large, and once the setting is wrong, the validity and the accuracy of the final analysis result are directly influenced.
(3) For fuzzy set theory, Zadeh proposed two fuzzy set forms, Type-1 and Type-2 in 1965 and 1975, respectively, and the membership degree of the two fuzzy sets is either an accurate value or an interval value. However, for Type-1 fuzzy sets: it is in itself quite contradictory to use an exact value to measure the ambiguity; for the Type-2 fuzzy set, the value of the fuzzy set considers the fuzziness of the membership degree, and ignores the uncertainty of the interval. Thus, both fuzzy sets have drawbacks and do not handle the ambiguity and uncertainty in personal evaluation well.
(4) Research has shown that people generally prefer to use the form of linguistic values to give a decision, and it is often not easy to express perception with an accurate number.
Disclosure of Invention
In order to overcome the defects of the existing method, the invention provides a brand-new system method for evaluating, analyzing and identifying the interaction incidence relation among the related service requirements of the product, namely a rough cloud model DEMATEL method, which can reveal the internal relation among the requirements, help designers to find out key requirements, reasonably utilize the design resources of a company and provide better related service of the product for customers.
The method comprises the following steps:
step 1: group rough cloud model for establishing influence degree between related service demands of products
Step 11: interval language value evaluation matrix for obtaining product related service demand mutual influence degree
Assuming that n kinds of existing product related service requirements exist, K industry related experts are invited to form an expert group, the influence relation degrees between different product related service requirements are compared pairwise according to the evaluation language value scale shown in the table 1, and a judgment value is given in a range mode. The resulting evaluation matrix is represented as follows:
Figure BDA0002548508400000031
wherein: y iskAn interval language value evaluation matrix (K ═ 1,2.. K) given by expert K is represented;
Figure BDA0002548508400000032
represents the evaluation of the influence degree of the expert k on the demand j for the product-related service demand i (i ═ 1,2,.., n; j ═ 1,2.., n). Furthermore, according to the DEMATEL method, the values on the diagonal of the evaluation matrix are all 0, i.e.: (y)ii＝0；i＝1,2,...,n)。
Table 1: language value table for evaluating influence degree between related service demands of products
Figure BDA0002548508400000033
Step 12: converting each interval language evaluation value into an interval cloud model (interval cloud model)
Definition 1(Li et al, 2019): let S be { S ═ Sα| α ═ 0,1, …, t, t ∈ N } is a discrete set of terms, U [ X ] }min,Xmax]Representing valid discourse domain and set in advance by expert group according to actual situation. Thus, for each language value, a corresponding base cloud model C can be obtainedα＝(Exα,Enα,Heα) The numerical features of each cloud model are calculated by Golden Section (Golden Section).
With language item set S ═ S0Without influence, s1Low degree of influence, s2Moderate in effect, s3High degree of influence, s4Very high-influence } for example, where t is 4 and U is [0,1 ═ c]And He2Each 0.01 is set in advance by an expert group, and the calculation steps of each basic cloud model are as follows:
Figure BDA0002548508400000034
Figure BDA0002548508400000041
Figure BDA0002548508400000042
the 5 calculated basic cloud models are shown in table 2:
table 2: basic cloud model corresponding to evaluation language value
Figure BDA0002548508400000043
Definition 2: suppose [ s ]α,sβ]Is an interval language value evaluation composed of elements in a language item set, then according to definition 1, two basic cloud models can be obtained respectively: cα＝(Exα,Enα,Heα) And Cβ＝(Exβ,Enβ,Heβ). Thus, s can be further obtainedα,sβ]Interval cloud model of
Figure BDA0002548508400000044
Figure BDA0002548508400000045
Thus, from the correlation formulas in definitions 1 and 2, the section cloud model of each section language evaluation value can be calculated:
Figure BDA0002548508400000046
step 13: converting each interval cloud model into corresponding rough cloud model
Various personal uncertainty (e.g., randomness and ambiguity of the inner heart) can be handled by using the cloud model. Furthermore, in order to simultaneously cope with evaluation uncertainty (inter uncertainty) between persons due to knowledge accumulation, work experience, and the like, the rough set theory is used in combination with the cloud model.
Definition 3(Liu et al, 2017): suppose that in the same effective discourse domain U, there are 2 interval cloud models
Figure BDA0002548508400000047
And
Figure BDA0002548508400000048
according to the 3En rule (3En principal), the two interval cloud models can be compared in size, and the specific comparison rule is as follows:
(1) if R isab>0, then
Figure BDA0002548508400000049
(2) If R isab0 and En1<En2Then, then
Figure BDA0002548508400000051
(3) If R isab0 and En1＝En2And He1<He2Then, then
Figure BDA0002548508400000052
(2) If R isab0 and En1＝En2And He1＝He2Then, then
Figure BDA0002548508400000053
Wherein:
Figure BDA0002548508400000054
a＝Ex 1 -3En1，
Figure BDA0002548508400000055
b＝Ex 2 -3En2，
Figure BDA0002548508400000056
definition 4(Li et al, 2019): assuming that K experts participate in the evaluation, each evaluation item (influence degree of the demand i on the demand j) has K interval language evaluation values [ s ]α,sβ]i(i ═ 1,2, …, K), then according toDefinitions 1 and 2 can be converted into K corresponding interval cloud models, and then an interval cloud model evaluation set is formed
Figure BDA0002548508400000057
And combining the rough set theory and the definition 3, for each evaluation item (influence degree of the demand i on the demand j), each cloud model in the interval cloud model set can be further converted into a rough cloud model, and the specific calculation steps are as follows:
(1) computing each interval cloud model
Figure BDA0002548508400000058
And upper and lower approximation sets:
Figure BDA0002548508400000059
Figure BDA00025485084000000510
(2) computing each interval cloud model
Figure BDA00025485084000000511
And upper and lower approximation limits:
Figure BDA00025485084000000512
Figure BDA00025485084000000513
(3) computing per-interval cloud model
Figure BDA00025485084000000514
The rough cloud model of (2):
Figure BDA00025485084000000515
thus, according to the correlation calculation formula in definition 4, each interval cloud model can be further converted into a rough cloud model form:
Figure BDA0002548508400000061
step 14: rough cloud model summarizing different experts
For each evaluation item (influence degree of the demand i on the demand j), the rough cloud model evaluation values of all experts are aggregated here using an arithmetic mean method. The specific calculation formula is as follows:
Figure BDA0002548508400000062
wherein:
Figure BDA0002548508400000063
a group rough cloud model representing how much demand i affects demand j.
Step 15: evaluation matrix of group-generating rough cloud model
Integrating the above calculation results to obtain an n × n rough cloud model evaluation matrix Y, which can be abbreviated as Y ═ Yij](i＝1,2,...,n；j＝1,2,...,n)：
Figure BDA0002548508400000064
Step 2: numerical determination of the degree of influence between product-related service demands
Step 21: calculating subjective value of degree of influence
For each i ≠ j: subjective value of the extent of influence of demand i on demand j
Figure BDA0002548508400000065
Comprises the following steps:
Figure BDA0002548508400000066
step 22: calculating objective values of degree of influence
Definition 5(Wang et al, 2015): suppose that in the valid universe U, there are 2 arbitrary interval cloud models
Figure BDA0002548508400000067
And
Figure BDA0002548508400000068
the distance between them can be given by the following formula:
Figure BDA0002548508400000069
Figure BDA0002548508400000071
thus, based on the concept of statistical variance and the correlation formula in definition 5, for each i ≠ j, the objective value of the degree of influence of a requirement i on a requirement j
Figure BDA0002548508400000072
Comprises the following steps:
Figure BDA0002548508400000073
Figure BDA0002548508400000074
Figure BDA0002548508400000075
wherein:
Figure BDA0002548508400000076
is that
Figure BDA0002548508400000077
Average cloud model of (i ═ 1,2, …, n), VijIs that
Figure BDA0002548508400000078
And
Figure BDA0002548508400000079
the distance difference between them.
Step 23: calculating a composite value of the degree of influence
For each i ≠ j: the integrated value a of the extent of influence of the demand i on the demand jijComprises the following steps:
Figure BDA00025485084000000710
and step 3: product related service requirement ordering based on DEMATEL method
Step 31: generating a direct-relation matrix (direct-relation matrix) A
According to the above calculation result, an n × n direct relationship matrix a can be obtained, which is abbreviated as: a ═ aij](i＝1,2,...,n；j＝1,2,...,n)：
Figure BDA00025485084000000711
Step 32: calculating a normalized direct-relation matrix (normalized direct-relation matrix) X
The normalized direct relation matrix X may be obtained by the following equation to ensure that each value in the matrix X is between 0 and 1:
X＝k×A (23)
Figure BDA0002548508400000081
step 33: calculating a total-relation matrix (T)
By adding a direct influence matrix X and an indirect influence matrix (X)2,X3…), obtaining a full relation matrix T:
T＝X+X2+...+Xn＝X(I-X)-1＝[tij]n×n (25)
wherein: i is an n × n identity matrix.
Step 34: calculating influence degree, center degree (degree) and reason degree (relationship)
For each product-related service requirement, its corresponding row and D are calculatediAnd column and RjThey represent the influence and influence of the demand, respectively:
Figure BDA0002548508400000082
further, a centrality (D) of each product-related service requirement may be determinedi+Rj) And degree of cause (D)i–Rj) They show the importance and net effect, respectively, of the product-related service demand on the overall product-related service.
Step 35: generating a cause and effect graph (practical diagram)
According to the content of the paper by Chien et al (2014):
(1) when (D)i+Rj)>avg(Di+RjWhen i, j is 1,2, n, the product-related service requirement i is a high-impact requirement;
(2) when (D)i+Ri)<avg(Di+RiI, j ═ 1,2,. n), the product-related service requirement i is a low-impact requirement;
(3) when (D)i-Rj)>When the value is 0, the influence degree of the product related service demand i on other demands is large, and the product related service demand i is a reason demand;
(4) when (D)i-Rj)<0, indicating that the product-related service demand i has a small influence on other demands,is a result requirement;
thus, a causal graph can be made, as shown in FIG. 2. According to the quadrant positions of the service requirements related to different products in the cause-effect diagram, the design priority sequence of the service requirements related to each product can be determined: from the first quadrant (core product related service requirements), to the second quadrant (drive type product related service requirements), to the third quadrant (independent type product related service requirements), and finally to the fourth quadrant (impact type product related service requirements).
The beneficial effect of this application is:
(1) consider that: people generally prefer to make evaluations by fuzzy linguistic values (e.g., high and low degrees of influence) rather than exact mathematical values (e.g., 2.7 degrees of influence), and to make decisions more conveniently by interval values (e.g., degrees of influence between no to moderate influences) rather than by single values. The method of the present invention allows evaluators to freely express their judgment using the form of interval language values. Therefore, the method is a more convenient, simple and humanized evaluation method for evaluators; for product-related service designers, more original viewpoint data with high possibility, complexity and uncertainty can be obtained, and more real and accurate demand analysis results can be obtained.
(2) The method combines the advantages of the DEMATEL method, the cloud model theory and the rough set theory, and is a brand new system analysis method. In the data processing and analyzing process, various evaluation uncertainties such as personal evaluation uncertainty (interpersonal uncertainty), interpersonal evaluation uncertainty (interpersonal uncertainty), ambiguity, randomness, and the like can be simultaneously processed. The final analysis result is displayed in a cause and effect diagram form, so that cause and effect interaction relations among related service requirements of different products and final priority sequencing are clearer and more intuitive, and a more effective and accurate evaluation analysis result is obtained.
(3) The cloud model theory is used instead of the fuzzy set theory to process the personal evaluation uncertainty (fuzzy uncertainty), so that the use defect of the fuzzy set theory can be effectively avoided: a lot of prior information (such as membership functions, fuzzy rules, etc.) needs to be preset in advance, and the accuracy of subsequent results and the membership degree are determined values directly influenced by the suitability of the prior information setting. Because, in the cloud model theory, the membership degree is a series of points, the prior information does not need to be preset in advance.
(4) The method comprehensively considers the subjective value and the objective value of the influence degree, and can more comprehensively disclose the relative influence degree between the related service demands of the product.
(5) The method is also suitable for other actual problem fields needing to comprehensively analyze all composition factors of the complex system and determine key factors in a fuzzy environment.
Drawings
FIG. 1 is a comparison of membership in cloud model theory and fuzzy set theory;
FIG. 2 is a cause and effect diagram of product related service requirements;
fig. 3 is a causal graph of elevator product related service requirements of an embodiment of the present invention;
fig. 4 is an elevator product related service demand correlation diagram of one embodiment of the invention;
fig. 5 is the result of the 5 methods of one embodiment of the invention prioritizing elevator product related service demands.
Detailed Description
The preferred embodiments of the present application will be described below with reference to the accompanying drawings for clarity and understanding of the technical contents thereof. The present application may be embodied in many different forms of embodiments and the scope of the present application is not limited to only the embodiments set forth herein.
The conception, the specific structure and the technical effects of the present invention will be further described below to fully understand the objects, the features and the effects of the present invention, but the present invention is not limited thereto.
In one embodiment of the invention, company a is a leading elevator manufacturing enterprise in China, and manufactures various elevator related products, such as passenger elevators, freight elevators, escalators, elevator monitoring systems and the like. In the face of increasingly intense market competition environment, in order to improve the market competitiveness of products of the company, obtain higher customer satisfaction, increase market share and realize sustainable development, company a decides to further optimize relevant service contents of the products provided by the company for customers, such as pre-sale type consultation guidance, installation and debugging of elevator products, remote monitoring of elevator running states, regular maintenance and upgrading of elevator products and the like.
Through the discussion of the product-related service designer and product manager in company a, the product-related service requirements common to 13 elevator customers are finally extracted, as shown in table 3 below:
TABLE 3 Elevator product-related service requirements for company A
Figure BDA0002548508400000101
Then, an expert group is established, and based on the evaluation language value table shown in table 1, the mutual influence degree between the 13 product-related service demands is compared and judged pairwise. The expert group consists of 5 experts, including 1 product related service manager of company A, 1 product related service designer of company A, 1 elevator engineer of company A and 2 client representatives of company A. They all have at least 6 years of relevant working experience and are therefore very qualified to give personal assessment.
5.2 Elevator product related service demand analysis example
5.2.1 cloud model of degree of influence between elevator product related service demands
The evaluation result of the section language value of expert 1 is shown in table 4, for example, [ L, M ] in the third row and the first column indicates: the degree of influence of the expert 1 on the demand 1 by the demand 3 is judged as "low degree of influence to medium degree of influence". The evaluation results of the remaining 4 experts are not listed here due to space limitation.
The qualitative linguistic value evaluation results were converted to a quantitative interval cloud model form according to definitions 1,2 and equation (6), as shown in table 5. For example, the section cloud model of the third row and the first column ([0.25,0.5],0.034,0.013) indicates the degree of influence judgment value [ L, M ] of the requirement 1 on the requirement 3 by the expert 1:
[0.25,0.5 ]: is a mathematical expected interval value of the interval cloud model, and corresponds to the abscissa range of the index [ L, M ] in FIG. 1;
0.034: the dispersion degree of each cloud drop in the cloud model is represented, and the randomness and the fuzziness of the qualitative concept that the influence degree is low to be medium are reflected;
0.013: the thickness of the cloud model is represented, reflecting the uncertainty of the membership.
Then, in order to cope with personal evaluation uncertainty (interpersonal uncertainty) and interpersonal uncertainty (interpersonal uncertainty), the rough cloud models of all evaluation results were calculated using equations (7) to (12). Due to the limited space, table 6 only lists the rough cloud models of the influence of five experts on SR1 in SRi, and the specific numerical interpretation of each rough cloud model is similar to the above.
Finally, according to the formulas (13) to (14), the rough cloud model evaluation results of different experts are summarized into a group rough cloud model matrix, as shown in table 7. For example, the values ([0.238,0.435],0.037,0.015) in the third row and the first column represent the final interval cloud model obtained after scientific processing and summarizing judgment values of the degree of influence of 5 experts on the requirement 1 by the requirement 3, and represent the mutual influence degree among the requirements.
TABLE 4 Interval linguistic value evaluation result matrix of expert 1
Figure BDA0002548508400000111
TABLE 5 Interval cloud model matrix of expert 1 evaluation results
Figure BDA0002548508400000112
Figure BDA0002548508400000121
TABLE 6 Rough cloud model of 5 expert evaluation results (in SR)iFor SR1Degree of influence of (1) as an example)
Figure BDA0002548508400000122
TABLE 7 Rough cloud model matrix of influence degree between elevator product related service demands
Figure BDA0002548508400000123
Figure BDA0002548508400000131
5.2.2 numerical determination of the extent of influence between the related service demands of elevator-related products
First, a subjective value of the degree of influence is calculated by formula (15). Next, an objective value of the degree of influence is obtained based on the formulas (16) to (20). Finally, the comprehensive influence value is calculated by using the formula (21). Due to space limitations, Table 8 shows only the numerical determination of the extent to which SRi has an effect on SR 1. For example, data of the third line: 0.33625, and 0.423686278, the objective influence degree of demand 3 on demand 1, and the two data are multiplied to obtain the combined influence degree 0.142464511 of demand 3 on demand 1, which is the input of the subsequent calculation.
TABLE 8 numerical determination of the extent of impact (in SR) between elevator product-related service demandsiFor SR1Influence value of (1) as an example)
Figure BDA0002548508400000132
5.2.3 Elevator product-related service demand ranking based on DEMATEL method
From the previous calculations a direct relation matrix (a) of the relevant service demands of the elevator product can be derived, as shown in table 9. Then, the normalized direct relation matrix (X) is calculated by using the equations (23) to (24), as shown in table 10. Using equation (25), a full relationship matrix (T) for the elevator product related service requirements is obtained, as shown in table 11. Where each cell value represents the degree to which a demand affects another demand.
TABLE 9 direct relationship matrix (A) of elevator product related service requirements
Figure BDA0002548508400000133
Figure BDA0002548508400000141
TABLE 10 normalized direct relation matrix (X) of elevator product related service requirements
Figure BDA0002548508400000142
TABLE 11 full relationship matrix (T) for elevator product related service requirements
Figure BDA0002548508400000143
Figure BDA0002548508400000151
Based on the full-relation matrix T and the formula (26), the related service requirement SR of each elevator product is calculated respectivelyiDegree of influence of DiAnd degree of influence Rj(ii) a Further, the center degree (D) of the center is obtainedi+Rj) And degree of cause (D)i–Rj) The importance and net effect of the product-related service demand on the overall product-related service are indicated separately. The final calculation results are shown in table 12. Taking the first row of data as an example:
0.81293 represents the sum of the extent of influence of demand 1 on the other 12 demands;
0.42742 represents the sum of the impact of the other 12 demands on demand 1;
1.24034 reflects the centrality of demand 1 throughout the product-related service;
0.38551 reflects the net effect of demand 1 on the overall product-related service: i.e., whether the demand is "influencer" or "influencer" as a whole;
2 indicates that demand 1 is in the second quadrant position in the demand causality graph, and therefore its design importance bit is "second fleet".
TABLE 12 analysis results of elevator product-related service requirements
Figure BDA0002548508400000152
Figure BDA0002548508400000161
Finally, according to the centrality of the relevant service requirements of each elevator product (D)i+Rj) And degree of cause (D)i–Rj) A causal graph of the relevant service requirements of the elevator product is made, as shown in figure 3. Thus, the 13 elevator product-related service demands can be classified according to their quadrant positions in the cause and effect diagram, i.e. two categories, cause or effect and impact or affected. Therefore, the design priority order of the service requirements related to each product is as follows: from the first quadrant (core product related service requirements), to the second quadrant (driving product related service requirements), to the third quadrant (independent product related service requirements), and finally to the fourth quadrant (impact product related service requirements), i.e., SR3(periodic product maintenance and optimization upgrades)>SR6(Wide monitoring range of product running state)>SR2(Rapid logistics, product nondestructive)>SR13(customer opinion processing center)>SR5(training for product use)>SR1(professional preoccupation guidance service)>SR4(professional product installation and commissioning service)>SR11(sufficient stock of product spare parts)>SR12(Rapid product-related service response)>SR10(timely, reliable product maintenance service)>SR9(accurate product failure diagnosis)>SR7(product return guarantee)>SR8(ensure the safe and efficient operation of the product).
Fig. 4 shows the interaction correlation between the 13 elevator product related service demands, where only the key interaction influence relation exceeding the threshold value 0.1 in the calculation result of the mapping table 11 is represented by a connecting line segment, where the round end represents the "influencer" and the arrow end represents the "influencer". In order to correspond to the results of fig. 3, the quadrant positions of the requirements in the cause and effect diagram are represented using the same graphs. It is easy to find that the most critical interaction of the product-related service requirements of the first quadrant is the most complex, and it is fully verified that the requirements of the first quadrant should be designed and satisfied preferentially.
TABLE 13 comparison of different methods of analysis of the requirements
Figure BDA0002548508400000162
The results of an analysis of the relevant service requirements of an elevator product using these methods are shown in fig. 5 below:
the ranking results calculated by the calculation method based on the AHP are greatly different from other methods because the method does not consider the interaction influence relationship among the demands, but simply analyzes the importance degree of each demand. In the calculation result of the method, the related service demands of the elevator products in the top 3 are SR respectively8(ensure safe and efficient operation of the product), SR5(training for product use), SR10(timely, reliable product maintenance service). However, these three requirements are all ranked later in the method of the other 4, since they are all "affected classes" requirements.
Compared with the DEMATEL method existing in the academic field in the other 3, the sequencing result is approximately the same, so that the result validity of the method is verified. Compared with the method in the other 3, the method has the advantages that:
(1) consider that: people generally prefer to make evaluations by fuzzy linguistic values (e.g., high and low degrees of influence) rather than exact mathematical values (e.g., 2.7 degrees of influence), and to make decisions more conveniently by interval values (e.g., degrees of influence between no to moderate influences) rather than by single values. The approach presented in this study allows evaluators to freely express their judgment in the form of interval linguistic values. Therefore, the method is a more convenient, simple and humanized evaluation method for evaluators; for product-related service designers, more original viewpoint data with high possibility, complexity and uncertainty can be obtained, and more real and accurate demand analysis results can be obtained.
(2) The method combines the advantages of the DEMATEL method, the cloud model theory and the rough set theory, and is a brand new system analysis method. In the data processing and analyzing process, various evaluation uncertainties such as personal evaluation uncertainty (interpersonal uncertainty), interpersonal evaluation uncertainty (interpersonal uncertainty), ambiguity, randomness, and the like can be simultaneously processed. The final analysis result is displayed in a cause and effect diagram form, so that cause and effect interaction relations among related service requirements of different products and final priority sequencing are clearer and more intuitive, and a more effective and accurate evaluation analysis result is obtained.
(3) The cloud model theory is used instead of the fuzzy set theory to process the personal evaluation uncertainty (fuzzy uncertainty), so that the use defect of the fuzzy set theory can be effectively avoided: a lot of prior information (such as membership functions, fuzzy rules, etc.) needs to be preset in advance, and the accuracy of subsequent results and the membership degree are determined values directly influenced by the suitability of the prior information setting. Because, in the cloud model theory, the membership degree is a series of points, the prior information does not need to be preset in advance. A basic cloud model of three interval judgment values shown in figure 1 (the horizontal axis represents the influence degree, the maximum influence degree is 1, the minimum influence degree is 0, and the vertical axis represents the membership degree). For example, for 0.5, it is clear that in the cloud model, there are many cases of membership; in fuzzy set theory (here, a triangular membership function is used), only one non-zero membership is 1.
(4) The method comprehensively considers the subjective value and the objective value of the influence degree, and can more comprehensively disclose the relative influence degree between the related service demands of the product.
(5) The method is also suitable for other actual problem fields needing to comprehensively analyze all composition factors of a complex system and determine key factors in a fuzzy environment.
The foregoing detailed description of the preferred embodiments of the present application. It should be understood that numerous modifications and variations can be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the concepts of the present application should be within the scope of protection defined by the claims.

Claims (1)

1. A method for identifying an interaction association relationship between product related service requirements is characterized by comprising the following steps:
step 1: establishing a group rough cloud model of the influence degree among the related service requirements of the product;
and 2, step: determining a numerical value of the degree of influence between the product-related service demands;
and step 3: ordering the product-related service requirements based on a DEMATEL method;
the step 1 comprises the following steps:
step 11: obtaining an interval language value evaluation matrix of the mutual influence degree of the product related service demands;
step 12: converting each interval language evaluation value into an interval cloud model;
step 13: converting each of the interval cloud models into a corresponding rough cloud model;
step 14: summarizing the rough cloud models of different experts;
step 15: generating a group rough cloud model evaluation matrix;
the step 2 comprises the following steps:
step 21: calculating a subjective value of the degree of influence;
step 22: calculating an objective value of the degree of influence;
step 23: calculating a comprehensive value of the degree of influence;
the step 3 comprises the following steps:
step 31: generating a direct relation matrix A;
step 32: calculating a normalized direct relation matrix X;
step 33: calculating an overall relation matrix T;
step 34: calculating the influence degree, the influenced degree, the centrality degree and the reason degree;
step 35: generating a causal relationship graph;
in the step 11, the section language value evaluation matrix is expressed as follows:
Figure RE-FDA0003496891310000011
wherein: y iskRepresents the interval language value evaluation matrix given by expert K, K being 1,2.. K;
Figure RE-FDA0003496891310000012
representing the evaluation of the influence degree of an expert k on a demand j for a product-related service demand i, wherein i is 1,2. j ═ 1,2., n, where all values on the diagonal of the interval language value evaluation matrix are 0;
in step 12, the section cloud model of each section language evaluation value is:
Figure RE-FDA0003496891310000021
in step 13, each interval cloud model in the interval cloud model set is further converted into the rough cloud model, and the specific calculation steps are as follows:
(1) computing each of the interval cloud models
Figure RE-FDA0003496891310000028
Upper and lower approximation sets of:
Figure RE-FDA0003496891310000022
Figure RE-FDA0003496891310000023
(2) computing each of the interval cloud models
Figure RE-FDA0003496891310000029
Upper and lower approximation limits of (d):
Figure RE-FDA0003496891310000024
Figure RE-FDA0003496891310000025
(3) computing each of the interval cloud models
Figure RE-FDA00034968913100000210
The rough cloud model of (a):
Figure RE-FDA0003496891310000026
thereby, each of the interval cloud models is converted into the coarse cloud model form:
Figure RE-FDA0003496891310000027
in step 14, for each evaluation item, namely the influence degree of the requirement i on the requirement j, the rough cloud model evaluation values of all experts are aggregated by using an arithmetic mean method, and a specific calculation formula is as follows:
Figure RE-FDA0003496891310000031
wherein:
Figure RE-FDA0003496891310000032
the group rough cloud model represents the influence degree of the demand i on the demand j;
in the step 2, in the step of processing,
for each i ≠ j, subjective value of the degree of influence of requirement i on requirement j
Figure RE-FDA0003496891310000033
Comprises the following steps:
Figure RE-FDA0003496891310000034
for each i ≠ j, objective value of influence degree of demand i on demand j
Figure RE-FDA0003496891310000035
Comprises the following steps:
Figure RE-FDA0003496891310000036
Figure RE-FDA0003496891310000037
Figure RE-FDA0003496891310000038
wherein:
Figure RE-FDA0003496891310000039
is that
Figure RE-FDA00034968913100000310
I ═ 1,2, …, n, VijIs that
Figure RE-FDA00034968913100000311
And
Figure RE-FDA00034968913100000312
a distance difference therebetween;
for each i ≠ j, the comprehensive value a of the influence degree of the demand i on the demand jijComprises the following steps:
Figure RE-FDA00034968913100000313
in the step 3, the step of processing the image,
according to the combined value aijObtaining an n × n direct relation matrix a, which is abbreviated as: a ═ aij]，i＝1,2,...,n；j＝1,2,...,n：
Figure RE-FDA0003496891310000041
The normalized direct relation matrix X is obtained by ensuring that each value in the matrix X is between 0 and 1:
X＝k×A (16)
Figure RE-FDA0003496891310000042
by adding said normalized direct relation matrix X and indirect influence matrix X2,X3,…,XnAnd obtaining the full-relation matrix T:
T＝X+X2+...+Xn＝X(I-X)-1＝[tij]n×n (18)
wherein: i is an n multiplied by n identity matrix;
for each of said product-related service requirements, its corresponding row and D are calculatediAnd column and RjThey represent the influence and influence of the demand, respectively:
Figure RE-FDA0003496891310000043
further, determining a centrality D of each of said product-related service requirementsi+RjAnd degree of cause Di–Rj(ii) a And identifying the interactive incidence relation among the related service requirements of the products according to the interactive incidence relation, and determining the design priority order of the related service requirements of the products.
CN202010568648.4A 2020-06-19 2020-06-19 Method for identifying interaction association relation between related service demands of products Active CN111832905B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010568648.4A CN111832905B (en) 2020-06-19 2020-06-19 Method for identifying interaction association relation between related service demands of products

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010568648.4A CN111832905B (en) 2020-06-19 2020-06-19 Method for identifying interaction association relation between related service demands of products

Publications (2)

Publication Number Publication Date
CN111832905A CN111832905A (en) 2020-10-27
CN111832905B true CN111832905B (en) 2022-05-20

Family

ID=72897769

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010568648.4A Active CN111832905B (en) 2020-06-19 2020-06-19 Method for identifying interaction association relation between related service demands of products

Country Status (1)

Country Link
CN (1) CN111832905B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112907100B (en) * 2021-03-09 2024-05-14 北京光速斑马数据科技有限公司 Service demand measurement method and device and electronic equipment
CN117172619A (en) * 2023-11-02 2023-12-05 成都飞机工业（集团）有限责任公司 Complex equipment delivery demand evaluation method based on group AHP-cloud model

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104331532A (en) * 2014-09-12 2015-02-04 广东电网公司江门供电局 Power transformer state evaluation method based on rough set-cloud model
CN105574685A (en) * 2016-02-02 2016-05-11 浙江工业大学 Subjective and objective combination-based cloud service evaluation method
CN109857939A (en) * 2019-02-01 2019-06-07 北京航空航天大学 Accurate method for pushing towards intelligence manufacture service
CN110009415A (en) * 2019-04-02 2019-07-12 青海师范大学 The reputation prediction technique of new seller in a kind of e-commerce system
CN110399382A (en) * 2019-07-26 2019-11-01 中国民航大学 Civil aviaton's master data recognition methods and system based on cloud model and rough set
CN111178772A (en) * 2019-12-31 2020-05-19 河海大学常州校区 Disaster risk assessment method and system based on hesitation fuzzy set

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10878324B2 (en) * 2012-07-20 2020-12-29 Ent. Services Development Corporation Lp Problem analysis and priority determination based on fuzzy expert systems

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104331532A (en) * 2014-09-12 2015-02-04 广东电网公司江门供电局 Power transformer state evaluation method based on rough set-cloud model
CN105574685A (en) * 2016-02-02 2016-05-11 浙江工业大学 Subjective and objective combination-based cloud service evaluation method
CN109857939A (en) * 2019-02-01 2019-06-07 北京航空航天大学 Accurate method for pushing towards intelligence manufacture service
CN110009415A (en) * 2019-04-02 2019-07-12 青海师范大学 The reputation prediction technique of new seller in a kind of e-commerce system
CN110399382A (en) * 2019-07-26 2019-11-01 中国民航大学 Civil aviaton's master data recognition methods and system based on cloud model and rough set
CN111178772A (en) * 2019-12-31 2020-05-19 河海大学常州校区 Disaster risk assessment method and system based on hesitation fuzzy set

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
A rough DEMATEL-based approach for evaluating interaction between requirements of product-service system;Wenyan Song et al;<Computers & Industrial Engineering>;20170615;第110卷;第353-363页 *
A rough-fuzzy DEMATEL-ANP method for evaluating sustainable value requirement of product service system;Zhihua Chen et al;《Journal of Cleaner Production》;20190424;第228卷;第485-508页 *
基于云模型—DEMATEL法的云制造资源评价问题研究;杨欣 等;《数学的实践与认识》;20181231;第48卷(第24期);第115-125页 *

Also Published As

Publication number Publication date
CN111832905A (en) 2020-10-27

Similar Documents

Publication Publication Date Title
Borade et al. Software project effort and cost estimation techniques
Hilorme et al. Formation of risk mitigating strategies for the implementation of projects of energy saving technologies
Wang et al. Integration of fuzzy AHP and FPP with TOPSIS methodology for aeroengine health assessment
Zeng et al. Intuitionistic fuzzy social network hybrid MCDM model for an assessment of digital reforms of manufacturing industry in China
Curran et al. Review of aerospace engineering cost modelling: The genetic causal approach
Marques et al. Multi-criteria performance analysis for decision making in project management
Chen A fuzzy back propagation network for output time prediction in a wafer fab
Kim et al. Characterizing viability of small manufacturing enterprises (SME) in the market
Castellanos et al. ibom: A platform for intelligent business operation management
Ebrat et al. Construction project risk assessment by using adaptive-network-based fuzzy inference system: An empirical study
Kumbhar et al. A digital twin based framework for detection, diagnosis, and improvement of throughput bottlenecks
CN111832905B (en) Method for identifying interaction association relation between related service demands of products
Alencar et al. A model for selecting project team members using multicriteria group decision making
Haghighi et al. A new analytical methodology to handle time-cost trade-off problem with considering quality loss cost under interval-valued fuzzy uncertainty
WO2017071369A1 (en) Method and device for predicting user unsubscription
Prokopenko et al. Development of the comprehensive method to manage risks in projects related to information technologies
Qu et al. Multi-stakeholder’s sustainable requirement analysis for smart manufacturing systems based on the stakeholder value network approach
Wang et al. Value evaluation method of industrial product-service based on customer perception
Sharma et al. Prognosis agent technology: influence on manufacturing organizations
Burak et al. Intuitionistic fuzzy number based group decision making approach for personnel selection
CN114385121B (en) Software design modeling method and system based on business layering
Ing et al. Edge-cloud collaboration architecture for AI transformation of SME manufacturing enterprises
Serpell Improving conceptual cost estimating performance
Singh et al. Critical factors of multi-agent technology influencing manufacturing organizations: an AHP and DEMATEL-oriented analysis
Sokolov et al. Combined models and algorithms on modern proactive intellectual scheduling under Industry 4.0 environment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant