CN112257979A - Domestic electronic device application effect evaluation method based on index system architecture - Google Patents

Domestic electronic device application effect evaluation method based on index system architecture Download PDF

Info

Publication number
CN112257979A
CN112257979A CN202010981022.6A CN202010981022A CN112257979A CN 112257979 A CN112257979 A CN 112257979A CN 202010981022 A CN202010981022 A CN 202010981022A CN 112257979 A CN112257979 A CN 112257979A
Authority
CN
China
Prior art keywords
evaluation
index
vector
matrix
application effect
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.)
Pending
Application number
CN202010981022.6A
Other languages
Chinese (zh)
Inventor
王晋婧
薛恩
李福秋
王小宁
角淑媛
郑紫霞
刘鼎
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.)
CHINA AEROSPACE STANDARDIZATION INSTITUTE
Original Assignee
CHINA AEROSPACE STANDARDIZATION INSTITUTE
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 CHINA AEROSPACE STANDARDIZATION INSTITUTE filed Critical CHINA AEROSPACE STANDARDIZATION INSTITUTE
Priority to CN202010981022.6A priority Critical patent/CN112257979A/en
Publication of CN112257979A publication Critical patent/CN112257979A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • 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

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)
  • Development Economics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Mathematical Analysis (AREA)
  • Educational Administration (AREA)
  • Computational Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Algebra (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

The invention relates to an application effect evaluation method of a domestic electronic device based on an index system architecture, and belongs to the technical field of autonomous control of components. The method comprises the following steps: (1) acquiring a universal index system architecture for the application effect of the domestic electronic device; (2) weighting the original scoring matrix by applying an optimal combined weighting method of the subjective and objective weight information aiming at each evaluation dimension, and carrying out evaluation to obtain an evaluation vector of each evaluation dimension; (3) taking the evaluation representation item as an evaluation object, taking the evaluation vector of each evaluation dimension as an index element of the evaluation object, calculating the subjective weight corresponding to the evaluation representation item based on the judgment matrix, and weighting the evaluation result of the evaluation dimension to obtain the evaluation vector; (4) and taking the evaluation effect as an evaluation object, taking the evaluation vector of each evaluation representation item as an index element of the evaluation representation item, calculating the subjective weight corresponding to the application effect based on the judgment matrix, and weighting the evaluation result of the evaluation representation item to obtain the application effect evaluation result of the domestic electronic device.

Description

Domestic electronic device application effect evaluation method based on index system architecture
Technical Field
The invention relates to an application effect evaluation method of a domestic electronic device based on an index system architecture, and belongs to the technical field of autonomous control of components.
Background
With the increasing requirement of the major national specialties of manned spaceflight, Beidou navigation, heavy carrier rockets, deep space exploration and the like on the autonomous controllability of electronic devices, the research and the application of the domestic electronic devices are greatly developed through the promotion of model specialties and the support of related specialties, and digital integrated circuits such as DSP, FPGA, FLASH, SRAM and other devices are initially scaled and form a series. On the basis of comprehensively completing ground verification, part of electronic devices gradually develop actual combat application on weaponry, satellites and airships.
In the engineering practice, many factors affecting the large-scale use of domestic components and devices exist, and in order to understand and master the application effect of components and devices used in models and meet the engineering use requirements, evaluation and evaluation need to be performed on the components and devices.
However, under the situation that the application range of the device is wider and the application objects are more and more, a general application effect evaluation system of the domestic electronic device is lacked, and the application effect of the device in the model cannot be objectively and accurately evaluated, so that each part applying the device cannot comprehensively obtain the application effect of the developed application item of the device, and the popularization and the use of the domestic device in various models are restricted.
Disclosure of Invention
The technical problem solved by the invention is as follows: the method overcomes the defects of the prior art, provides a domestic electronic device application effect evaluation method based on a universal index system, and provides quantitative data support for continuous popularization and application of devices.
The technical scheme for solving the technical problem is as follows: the domestic electronic device application effect evaluation method based on the index system architecture comprises the following steps:
(1) acquiring a universal index system architecture for the application effect of the domestic electronic device; the domestic electronic device application effect general index system architecture comprises four layers from top to bottom, wherein the top layer is an application effect, and the second layer is an evaluation representation item; the third layer is an evaluation dimension corresponding to each evaluation representation item, and the lowest layer is an index element related to each evaluation dimension corresponding to each evaluation representation item in the device development process and the application task process;
(2) executing the steps (2.1) to (2.2) aiming at each evaluation dimension to obtain an evaluation vector of each evaluation dimension;
(2.1) taking the evaluation dimension as an evaluation object, and acquiring an original scoring matrix and a judgment matrix corresponding to the evaluation object, wherein the original scoring matrix comprises the recognition degree of each index element given by n experts for the index element, which expresses the realization of the index element on a related product by the expert, and can be given in the forms of a percentage score, a relative ratio and the like; the judgment matrix comprises importance degree sequencing among index elements given by n experts, wherein n is more than or equal to 1;
(2.2) calculating a subjective weight vector corresponding to the evaluation object based on the judgment matrix of the evaluation object; calculating to obtain an objective weight vector through an information entropy method based on the evaluation object original scoring matrix; calculating to obtain main and objective evaluation results through the subjective weight vector, the objective weight vector and an original scoring matrix respectively, calculating a standard deviation of the main and objective evaluation results, if the standard deviation is greater than or equal to a preset threshold, performing combined weight calculation on the subjective weight vector and the objective weight vector to obtain a combined weight vector, and recording the combined weight vector as an evaluation vector of an evaluation dimension; otherwise, recording the subjective weight vector as an evaluation vector of an evaluation dimension;
(3) taking the evaluation representation item as an evaluation object, taking the evaluation vector of each evaluation dimension as an index element of the evaluation object, calculating a subjective weight vector corresponding to the evaluation representation item based on the judgment matrix, and weighting the evaluation result of the evaluation dimension to obtain an evaluation vector;
(4) and taking the evaluation effect as an evaluation object, taking the evaluation vector of each evaluation representation item as an index element of the evaluation representation item, calculating the subjective weight corresponding to the application effect based on the judgment matrix, and weighting the evaluation result of the evaluation representation item to obtain the application effect evaluation result of the domestic electronic device.
The construction method of the judgment matrix comprises the following steps:
(1) establishing an evaluation object judgment matrix element aijI is 1,2, a, n, j is 1,2, a, m, i is a matrix row number, j is a matrix column number, n is the number of experts, and m is the number of index elements; a isijIs any two index elements XiAnd XjA ratio scale value of;
(2) any two index elements XiAnd XjIs used for expressing any two index elements X of each expert about the evaluation objectiAnd XjThe relative description of the degree of importance of (c) is assigned according to the following table:
scale Means of
1 Indicates that two elements have the same importance compared
2 Means that the former is slightly more important than the latter when compared with the latter
3 Means that the former is significantly more important than the latter when compared with the two elements
4 Means that the former is more important than the latter in comparison with the two elements
5 Means that the former is extremely important than the latter in comparison with the two elements
1.5,2.5,3.5,4.5 Intermediate value representing the above-mentioned adjacent judgment
The subjective weight evaluation vector calculation step is as follows:
(1a) solving the maximum characteristic root of the judgment matrix and the characteristic vector corresponding to the maximum characteristic root;
(2a) normalizing the eigenvector corresponding to the maximum characteristic root of the judgment matrix to obtain a subjective weight vector;
(3a) and multiplying the original scoring matrix by the subjective weight vector to obtain a subjective weight evaluation vector.
The objective weight evaluation vector calculation step is as follows:
(1b) converting the original scoring matrix into a normalized matrix X ═ Xij)m×n；
(2b) And calculating the information entropy value of the index element according to the normalized matrix:
Figure BDA0002687523430000031
in the formula, when xijWhen 0, ln (x) is specifiedij)＝0；
(3b) Calculating the variation coefficient of the index element according to the information entropy;
dj＝1-hj,j＝1,2,...,n
(4b) calculating an objective weight vector of the index element:
Figure BDA0002687523430000032
(5b) and multiplying the original scoring matrix and the objective weight vector to obtain an objective weight evaluation vector.
The standard for constructing the normalized matrix is as follows:
when the index element is a benefit type index, i.e., the larger the index value, the better, the index element is normalized by the following formula:
Figure BDA0002687523430000041
when the index element is a cost-type index, namely the smaller the index value is, the better the index value is, the normalization is performed by using the following formula;
Figure BDA0002687523430000042
when the index element is an interval type index, the following formula is used for standardization:
Figure BDA0002687523430000043
wherein [ q ]j1,qj2]Is the interval range of interval type index.
The calculation method of the combined weight evaluation vector is as follows:
(1c) the subjective weight W is calculated1And objective weight W2The component block matrix W ═ W1, W2]；
(2c) Solving a non-negative fixed square matrix B of the normalized matrix X1The calculation method is as follows:
Figure BDA0002687523430000044
wherein x isi1～xinAnd
Figure BDA0002687523430000045
elements in the normalized matrix X;
(3c) calculating a symmetric matrix WTB1W；
(4c) Calculating a symmetric matrix WTB1The maximum characteristic root of W and the corresponding characteristic vector thereof;
(5c) normalizing the characteristic vector corresponding to the maximum characteristic root to obtain a normalized characteristic column vector T
(6c) Calculating a combined weighting coefficient Wc＝[W1,W2]*T
(7c) And a pair of combined weighting coefficients WcCarrying out normalization processing to obtain a combined weight vector;
(8c) and multiplying the original scoring matrix and the combined weight vector to obtain a combined weight evaluation vector.
Compared with the prior art, the invention has the beneficial effects that:
(1) the invention forms the construction method and the process of the index system by analyzing the relevant characteristics of the index system for evaluating the application effect of the device, thereby developing the index weighting and evaluation by applying the method of the optimal and most weighted subjective and objective weights, and providing a basis for the use, popularization and application of the device.
(2) The invention uses the optimal combined weighting method based on the sum of squared deviations, not only considers the preference cognition of each expert on the index attribute, but also strives to reduce the subjective randomness, so that the weighting on the index attribute can achieve the subjective and objective unification.
(3) The indexes established by the comprehensive evaluation index system for the application effect of the device not only reflect the functional performance of the device, but also reflect the adaptability and reliability of the device in different environments. The index system is an evaluation index system which is established by combining the whole life cycle of the device, the feelings of users at different stages and different use conditions of the device.
(4) The research method of the comprehensive evaluation system for the application effect of the device can be widely popularized and applied to evaluation of the use effect of similar single machines and systems.
Drawings
FIG. 1 is a method and a process for constructing an index system of a domestic electronic device according to an embodiment of the present invention;
FIG. 2 is an index system of the effect of domestic electronic device according to an embodiment of the present invention;
FIG. 3 is a comprehensive evaluation method and process for domestic electronic devices according to an embodiment of the present invention;
Detailed Description
The invention is further illustrated by the following examples.
1. Analysis of application effect evaluation characteristics of domestic devices
The comprehensive evaluation time dimension of the application effect of the domestic electronic device relates to the full life cycle of the device, and the application process of the domestic electronic device comprises design, production, test experiment and on-orbit operation, so that the comprehensive evaluation of the application effect needs to cover all the stages of the application of the domestic electronic device, and the application good degree and the application suitable degree of the device are comprehensively obtained and used as the basis for large-area popularization and application of the device.
After the components are comprehensively used and verified on the ground of a chip level, the components enter a model application stage on the basis that the stability and consistency of the performance level reach a certain level. The model uses a new device, and the performance and reliability performance of the device are obtained through a large amount of work in two stages of ground and on-orbit and are used as the basis of application effect evaluation. Generally, the using process of the device in the model comprises the stage processes of device guarantee, circuit design, production and manufacture, debugging test, on-track test and the like, so that the application effect evaluation is carried out aiming at a series of processes of the device applied in the model, and the purpose is to evaluate the effect performance of the device in different application processes. From this, it can be seen that the application effect evaluation has the following significant characteristics: (1) the whole process of application effect. The target of the application effect evaluation is whether the device meets the application requirement of the model working at each stage, and the stage working needs to carry out all-around evaluation on a plurality of application processes such as component guarantee, model design, model production, model ground verification, model flight verification and the like, so that a constructed index system needs to cover the processes such as component guarantee, model design, production, verification, application and the like; (2) cross-domain application of effects. The evaluation of the application effect of the electronic device comprises a plurality of stages, and inevitably spans a plurality of fields, such as device guarantee, model production stage, model design, circuit design, system electromagnetic compatibility design and the like, and the field of component reliability guarantee, project management, quality management and the like, and the field verification relates to reliability verification, anti-irradiation verification and the like. Therefore, in the process of constructing the index system, the experience and knowledge of experts in the related field need to be gathered widely.
On the basis of comprehensively analyzing the use characteristics of the device in the model, an applicable method is established and applied to the construction of an index system.
2. Characteristic analysis of application effect evaluation index system
The characteristics of the application effect evaluation of the electronic device determine that the constructed index system can meet the following requirements:
(1) complexity of the index system. The application of the device in the model relates to each chain for product development, and the selection of indexes not only highlights the stage characteristics, but also ensures the relative independence among the indexes, so the construction work of an index system is relatively complex;
(2) the index is hierarchical. The construction of the index system is not a single hierarchical structure, but an index system with a certain hierarchical structure. For example, the anti-radiation capability is an index to be verified in the ground test verification stage, and the anti-radiation capability needs to be analyzed from multiple levels and aspects such as the anti-radiation capability of the device, the system structure design, the anti-locking design of the system, the single-particle overturn prevention design and the like.
Through the analysis, the problems to be solved by the index system construction are as follows:
(1) and establishing an expert selection method. Screening expert groups meeting the requirement of constructing an index system, and providing effective intelligence support for identifying application effect evaluation influence factors and constructing the index system;
(2) evaluation elements are effectively identified. Through analysis, identifying a plurality of factors influencing the application of each process of the device, and sequencing the importance;
(3) independence of the indices. The mutual influence among all the elements is reduced as much as possible, and an application verification comprehensive evaluation index system with clear layers and simple structure is constructed.
3. Index system construction method and process
As shown in FIG. 1, the index system construction process comprises four steps:
(1) and determining an evaluation target. The evaluation target directly determines the selection of the index elements in the index system, and the index system and the evaluation target are related to each other.
(2) Evaluation dimension and influence factor analysis. And carrying out influence factor analysis in multiple dimensions and multiple angles according to an evaluation target from a device application task and process, and widely identifying and evaluating evaluation elements related to the dimensions.
(3) And constructing an index system. The evaluation dimension is associated with the identified influencing factor.
(4) The importance of the index elements and the influence domain balance analysis. And analyzing the importance, the influence magnitude, the quantitative evaluation condition of the elements and the like of the evaluation elements identified by combing on the evaluation target, and ensuring the independence of each index element and the quantitative evaluation of the elements as much as possible, thereby forming an evaluable index system.
By applying the method, the identified main influence factors for evaluating the application effect of the electronic device comprise:
(1) the major factors identified during the device assurance phase include: screening qualification rate, supply capacity and the like.
(2) The main factors identified in the model design phase include: redundancy design, derating design and electromechanical and thermal integrated design;
(3) the main factors identified during the model production phase include: process maturity, assembly complexity, production qualification rate.
(4) The main factors identified during the ground test validation phase include:
1) functional performance aspects: tolerance analysis and circuit matching;
2) environmental adaptability aspect: noise, sine + random vibration, temperature shock, temperature cycling, thermal balance, thermal vacuum, electromagnetic compatibility;
3) and (3) system adaptability aspect: power supply bias, input characteristics, output characteristics, frequency characteristics, parameter consistency, main parameter variation, main parameter consistency and software and hardware matching;
4) radiation resistance effect: total dose, single event upset, single event lockout.
(5) The main influence factors of the flight phase comprise three aspects of key parameter change, key parameter consistency and on-orbit failure.
Based on the above, the invention constructs the basic structure of the comprehensive evaluation index system for evaluating the application effect of the electronic device, as shown in fig. 2.
The domestic electronic device application effect general index system architecture comprises four layers from top to bottom, wherein the top layer is an application effect, and the second layer is an evaluation representation item; the third layer is an evaluation dimension corresponding to each evaluation representation item, and the lowest layer is an index element related to each evaluation dimension corresponding to each evaluation representation item in the device development process and the application task process;
the application effect evaluation characterization items of the domestic electronic device comprise the usability and the fitness. The application process of the device is evaluated from two dimensions of good use degree and adaptability of the device. The good utilization degree is mainly to evaluate the reasonability, completeness and the like of technical elements from the three aspects of design, production and reliability of application devices and represent the effect of the devices in each application link; the adaptability degree is mainly evaluated from the adaptive effect of the device on the internal environment and the external environment to the use capability and the use consistency of the device, and the realization condition of the self capability of the device under various use environment conditions is represented.
For the good-use degree, the evaluation dimension at least comprises a functional performance dimension, a production assembly dimension and a reliability dimension.
The index elements of the functional performance dimension at least comprise a design stage functional performance dimension and a use stage functional performance dimension;
index elements related to the functional performance dimension in the design stage at least comprise design complexity and functional performance index satisfaction;
the index elements related to the functional performance dimension of the use stage at least comprise interface matching and index satisfaction rate.
The production assembly dimensions comprise at least a process dimension and an assembly dimension;
the index elements related to the process dimension at least comprise the satisfaction rate and the maturity of the existing process;
the index elements related to the assembly dimension at least comprise the one-time production assembly qualification rate of the product.
For fitness, the evaluation dimension comprises an exogenous adaptation effect dimension and an endogenous adaptation effect dimension.
The exogenous adaptive effect dimension comprises at least an environmental adaptive dimension;
the index elements related to the environmental adaptability dimension at least comprise space environmental adaptability, electromagnetic compatibility adaptability and vibration environmental adaptability.
The intrinsic cause adaptation effect dimension at least comprises a performance consistency dimension;
the index elements related to the performance consistency dimension comprise the consistency of the performance of the batch products, the consistency of the performance before and after the test, and the performance consistency of the unit test and the comprehensive test.
Through the analysis of relevant characteristics of the index system for evaluating the application effect of the device, the index system for evaluating the application effect of the device is constructed, and a foundation is provided for evaluating the application effect of the domestic device.
4. Domestic electronic device application effect evaluation method based on index system architecture
As shown in fig. 3, the present invention further provides a domestic electronic device application effect evaluation method based on the index system architecture, which includes the following steps:
(1) acquiring a universal index system architecture for the application effect of the domestic electronic device; the domestic electronic device application effect general index system architecture comprises four layers from top to bottom, wherein the top layer is an application effect, and the second layer is an evaluation representation item; the third layer is an evaluation dimension corresponding to each evaluation representation item, and the lowest layer is an index element related to each evaluation dimension corresponding to each evaluation representation item in the device development process and the application task process;
(2) executing the steps (2.1) to (2.2) aiming at each evaluation dimension to obtain an evaluation vector of each evaluation dimension;
(2.1) taking the evaluation dimension as an evaluation object, and acquiring an original scoring matrix and a judgment matrix corresponding to the evaluation object, wherein the original scoring matrix comprises the recognition degree of each index element given by n experts for the index element, which expresses the realization of the index element on a related product by the expert, and can be given in the forms of a percentage score, a relative ratio and the like; the judgment matrix comprises importance degree sequencing among index elements given by n experts, wherein n is more than or equal to 1;
the construction method of the judgment matrix comprises the following steps:
(1) establishing an evaluation object judgment matrix element aijI is 1,2, a, n, j is 1,2, a, m, i is a matrix row number, j is a matrix column number, n is the number of experts, and m is the number of index elements; a isijIs any two index elements XiAnd XjA ratio scale value of;
(2) any two index elements XiAnd XjIs used for expressing any two index elements X of each expert about the evaluation objectiAnd XjThe relative description of the degree of importance of (c) is assigned according to the following table:
scale Means of
1 Indicates that two elements have the same importance compared
2 Means that the former is slightly more important than the latter when compared with the latter
3 Means that the former is significantly more important than the latter when compared with the two elements
4 Means that the former is more important than the latter in comparison with the two elements
5 Means that the former is extremely important than the latter in comparison with the two elements
1.5,2.5,3.5,4.5 Intermediate value representing the above-mentioned adjacent judgment
(2.2) calculating a subjective weight vector corresponding to the evaluation object based on the judgment matrix of the evaluation object; calculating to obtain an objective weight vector through an information entropy method based on the evaluation object original scoring matrix; calculating to obtain main and objective evaluation results through the subjective weight vector, the objective weight vector and an original scoring matrix respectively, calculating a standard deviation of the main and objective evaluation results, if the standard deviation is greater than or equal to a preset threshold, performing combined weight calculation on the subjective weight vector and the objective weight vector to obtain a combined weight vector, and recording the combined weight vector as an evaluation vector of an evaluation dimension; otherwise, recording the subjective weight vector as an evaluation vector of an evaluation dimension;
the subjective weight assessment vector is calculated as follows:
(1a) solving the maximum characteristic root of the judgment matrix and the characteristic vector corresponding to the maximum characteristic root;
(2a) normalizing the eigenvector corresponding to the maximum characteristic root of the judgment matrix to obtain a subjective weight vector;
(3a) and multiplying the original scoring matrix by the subjective weight vector to obtain a subjective weight evaluation vector.
The objective weight evaluation vector is calculated as follows:
(1b) converting the original scoring matrix into a normalized matrix X ═ Xij)m×n；
The standard for constructing the normalized matrix is as follows:
when the index element is a benefit type index, i.e., the larger the index value, the better, the index element is normalized by the following formula:
Figure BDA0002687523430000111
when the index element is a cost-type index, namely the smaller the index value is, the better the index value is, the normalization is performed by using the following formula;
Figure BDA0002687523430000112
when the index element is an interval type index, the following formula is used for standardization:
Figure BDA0002687523430000113
wherein [ q ]j1,qj2]Is the interval range of interval type index.
(2b) And calculating the information entropy value of the index element according to the normalized matrix:
Figure BDA0002687523430000114
in the formula, when xijWhen 0, ln (x) is specifiedij)＝0；
(3b) Calculating the variation coefficient of the index element according to the information entropy;
dj＝1-hj,j＝1,2,...,n
(4b) calculating an objective weight vector of the index element:
Figure BDA0002687523430000115
(5b) and multiplying the original scoring matrix and the objective weight vector to obtain an objective weight evaluation vector.
Standard deviation std of subjective and objective evaluation results:
Figure BDA0002687523430000116
u is the mean of the subjective weight assessment vector and the objective weight assessment vector
The combined weight estimate vector is calculated as follows:
(1c) the subjective weight W is calculated1And objective weight W2The component block matrix W ═ W1,W2]；
(2c) Solving a non-negative fixed square matrix B of the normalized matrix X1The calculation method is as follows:
Figure BDA0002687523430000121
wherein x isi1～xinAnd
Figure BDA0002687523430000122
is an element of the normalized matrix X;
(3c) calculating a symmetric matrix WTB1W；
(4c) Calculating a symmetric matrix WTB1The maximum characteristic root of W and the corresponding characteristic vector thereof;
(5c) normalizing the characteristic vector corresponding to the maximum characteristic root to obtain a normalized characteristic column vector T
(6c) Calculating a combined weighting coefficient Wc＝[W1,W2]*T
(7c) And a pair of combined weighting coefficients WcCarrying out normalization processing to obtain a combined weight vector;
(8c) and multiplying the original scoring matrix and the combined weight vector to obtain a combined weight evaluation vector.
(3) Taking the evaluation representation item as an evaluation object, taking the evaluation vector of each evaluation dimension as an index element of the evaluation object, calculating a subjective weight vector corresponding to the evaluation representation item based on the judgment matrix, and weighting the evaluation result of the evaluation dimension to obtain an evaluation vector;
(4) and taking the evaluation effect as an evaluation object, taking the evaluation vector of each evaluation representation item as an index element of the evaluation representation item, calculating the subjective weight corresponding to the application effect based on the judgment matrix, and weighting the evaluation result of the evaluation representation item to obtain the application effect evaluation result of the domestic electronic device.
Example 1
An example is given below to illustrate the application of the method.
Dividing the evaluation of the application effect of the electronic device into two evaluation dimensions of good use degree and fitness; the usability evaluation is evaluated from three levels of functional performance, production assembly and reliability, wherein 4 indexes are considered in the aspect of functional performance, 4 indexes are considered in the aspect of production assembly, and 5 indexes are considered in the aspect of reliability; the fitness evaluation is carried out from two levels of external factor adaptability and internal factor adaptability, wherein 4 indexes are considered in the aspect of the external factor adaptability, and 3 indexes are considered in the aspect of the internal factor adaptability. Firstly, the comprehensive application effect of the electronic device is taken as a total target, and the total target is decomposed according to a hierarchical structure based on a hierarchical analysis method, as shown in fig. 2. And then determining whether to adopt the combining weight based on the subjective and objective weight distribution calculation. The subjective weight calculation needs to construct a judgment matrix to show the comparison of relative importance between a certain factor at the previous level and the relevant factor at the current level. The objective weight needs to construct a normalized standard decision matrix, and the calculation result reflects the comparison of relative importance between a certain factor of the previous level and the relevant factor of the previous level by adopting entropy calculation.
The application of the evaluation method will be described by taking the reliability evaluation under the good utilization degree of fig. 2 as an example.
The raw scoring matrix is
Figure BDA0002687523430000131
Determine the matrix as
Figure BDA0002687523430000132
1. Subjective weight assessment vector calculation
Obtaining the maximum feature root and the corresponding feature vector according to the matrix as
λmax＝5.2386
[-0.6500 -0.6200 -0.2163 -0.3224 -0.2060]
Normalizing the feature vectors to obtain a subjective weight vector of
W1＝[0.3226 0.3078 0.1073 0.1600 0.1023]
The subjective weight evaluation vector is calculated as
Figure BDA0002687523430000141
2 calculation of Objective weight evaluation vector
Normalized matrix
Figure BDA0002687523430000142
The information entropy value obtained is
[0.8418 0.8515 0.8515 0.6704 0.8550]
Coefficient of variation of
[0.1582 0.1485 0.1485 0.3296 0.1450]
The objective weight vector is
W2＝[0.1701 0.1597 0.1597 0.3545 0.1560]
The objective weight evaluation vector calculation results are:
Figure BDA0002687523430000143
3. standard deviation of subjective weight evaluation vector and objective weight evaluation vector
[0.0111 0.0104 0.0106 0.0101]
The condition that std is 0.01 or more is satisfied, and therefore, the evaluation is performed by the combination weight method.
4. Combining weight estimate vector calculation
Calculating a non-negative fixed square matrix
Figure BDA0002687523430000144
Figure BDA0002687523430000151
Figure BDA0002687523430000152
The maximum feature root of the matrix is 0.0922, and the corresponding feature vector is
Figure BDA0002687523430000153
Normalized to obtain
Figure BDA0002687523430000154
From this, the optimal combined weighting factor can be calculated as:
Figure BDA0002687523430000155
the combined weight estimate vector is calculated as
Figure BDA0002687523430000156
And evaluating and executing functional performance, production assembly, external cause adaptation effect and internal cause adaptation effect according to reliability, and giving evaluation results of the several evaluation dimensions.
Functional performance evaluation result matrix:
Figure BDA0002687523430000157
production assembly evaluation result matrix:
Figure BDA0002687523430000158
reliability evaluation fruit solving matrix
Figure BDA0002687523430000159
An external cause adaptation effect evaluation result matrix:
Figure BDA00026875234300001510
internal cause adaptability evaluation result matrix
Figure BDA0002687523430000161
Calculation of the goodness evaluation vector:
establishing a judgment matrix according to the importance ranking and scale assignment table of experts on three evaluation factors of functional performance, production assembly and reliability
Degree of good use Functional performance Production assembly Reliability of
Functional performance 1 3 2
Production assembly 1/3 1 1/1.5
Reliability of 1/2 1.5 1
Obtaining the maximum characteristic root of the judgment matrix and the corresponding characteristic vector
λmax＝3.0000
[0.8571 0.2857 0.4286]
Normalizing the feature vector to obtain a subjective weight vector of [ 0.54550.18180.2727 ]
The goodness evaluation vector is calculated as:
Figure BDA0002687523430000162
similarly, the fitness refers to the good utilization degree to be calculated and executed, and the obtained evaluation result is
Figure BDA0002687523430000163
Comprehensively evaluating the application effect of the device through the good use degree and the adaptability:
establishing a judgment matrix according to the importance ranking and scale assignment table of experts on three evaluation factors of functional performance, production assembly and reliability
Figure BDA0002687523430000164
Figure BDA0002687523430000171
Obtaining the maximum characteristic root of the judgment matrix and the corresponding characteristic vector
λmax＝2.0000
[0.8321 0.5547]
Normalizing the feature vector to obtain a subjective weight vector of [ 0.60.4 ]
Figure BDA0002687523430000172
Therefore, the evaluation result of 4 experts in the application effect of the device is obtained, and the average value of the evaluation result can be used as the final evaluation result of the application effect of the device.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the present invention, and those skilled in the art can make variations and modifications of the present invention without departing from the spirit and scope of the present invention by using the methods and technical contents disclosed above.

Claims (6)

1. The domestic electronic device application effect evaluation method based on the index system architecture is characterized by comprising the following steps of:
(1) acquiring a universal index system architecture for the application effect of the domestic electronic device; the domestic electronic device application effect general index system architecture comprises four layers from top to bottom, wherein the top layer is an application effect, and the second layer is an evaluation representation item; the third layer is an evaluation dimension corresponding to each evaluation representation item, and the lowest layer is an index element related to each evaluation dimension corresponding to each evaluation representation item in the device development process and the application task process;
(2) executing the steps (2.1) to (2.2) aiming at each evaluation dimension to obtain an evaluation vector of each evaluation dimension;
(2.1) taking the evaluation dimension as an evaluation object, and acquiring an original scoring matrix and a judgment matrix corresponding to the evaluation object, wherein the original scoring matrix comprises the recognition degree of each index element given by n experts for the index element, which expresses the realization of the index element on a related product by the expert, and can be given in the forms of a percentage score, a relative ratio and the like; the judgment matrix comprises importance degree sequencing among index elements given by n experts, wherein n is more than or equal to 1;
(2.2) calculating a subjective weight vector corresponding to the evaluation object based on the judgment matrix of the evaluation object; calculating to obtain an objective weight vector through an information entropy method based on the evaluation object original scoring matrix; calculating to obtain main and objective evaluation results through the subjective weight vector, the objective weight vector and an original scoring matrix respectively, calculating a standard deviation of the main and objective evaluation results, if the standard deviation is greater than or equal to a preset threshold, performing combined weight calculation on the subjective weight vector and the objective weight vector to obtain a combined weight vector, and recording the combined weight vector as an evaluation vector of an evaluation dimension; otherwise, recording the subjective weight vector as an evaluation vector of an evaluation dimension;
(3) taking the evaluation representation item as an evaluation object, taking the evaluation vector of each evaluation dimension as an index element of the evaluation object, calculating a subjective weight vector corresponding to the evaluation representation item based on the judgment matrix, and weighting the evaluation result of the evaluation dimension to obtain an evaluation vector;
(4) and taking the evaluation effect as an evaluation object, taking the evaluation vector of each evaluation representation item as an index element of the evaluation representation item, calculating the subjective weight corresponding to the application effect based on the judgment matrix, and weighting the evaluation result of the evaluation representation item to obtain the application effect evaluation result of the domestic electronic device.
2. The evaluation method for domestic electronic device application effect based on index system architecture as claimed in claim 1, wherein the construction method of the judgment matrix is as follows:
(1) establishing an evaluation object judgment matrix element aijI 1,2, a, n, j 1,2, a, m, i is the serial number of the matrix row, j is the serial number of the matrix column, n is the number of experts, m is the indexThe number of elements; a isijIs any two index elements XiAnd XjA ratio scale value of;
(2) any two index elements XiAnd XjIs used for expressing any two index elements X of each expert about the evaluation objectiAnd XjThe relative description of the degree of importance of (c) is assigned according to the following table:
scale Means of 1 Indicates that two elements have the same importance compared 2 Means that the former is slightly more important than the latter when compared with the latter 3 Means that the former is significantly more important than the latter when compared with the two elements 4 Means that the former is more important than the latter in comparison with the two elements 5 Means that the former is extremely important than the latter in comparison with the two elements 1.5,2.5,3.5,4.5 Intermediate value representing the above-mentioned adjacent judgment
3. The method for evaluating the application effect of domestic electronic devices based on index architecture as claimed in claim 1, wherein said subjective weight evaluation vector is calculated by the following steps:
(1a) solving the maximum characteristic root of the judgment matrix and the characteristic vector corresponding to the maximum characteristic root;
(2a) normalizing the eigenvector corresponding to the maximum characteristic root of the judgment matrix to obtain a subjective weight vector;
(3a) and multiplying the original scoring matrix by the subjective weight vector to obtain a subjective weight evaluation vector.
4. The method for evaluating the application effect of domestic electronic devices based on index architecture as claimed in claim 1, wherein said objective weight evaluation vector is calculated by the steps of:
(1b) converting the original scoring matrix into a normalized matrix X ═ Xij)m×n；
(2b) And calculating the information entropy value of the index element according to the normalized matrix:
Figure FDA0002687523420000021
in the formula, when xijWhen 0, ln (x) is specifiedij)＝0；
(3b) Calculating the variation coefficient of the index element according to the information entropy;
dj＝1-hj,j＝1,2,...,n
(4b) calculating an objective weight vector of the index element:
Figure FDA0002687523420000031
(5b) and multiplying the original scoring matrix and the objective weight vector to obtain an objective weight evaluation vector.
5. The method for evaluating the application effect of domestic electronic devices based on index system architecture as claimed in claim 1, wherein the normalized matrix construction standard is:
when the index element is a benefit type index, i.e., the larger the index value, the better, the index element is normalized by the following formula:
Figure FDA0002687523420000032
when the index element is a cost-type index, namely the smaller the index value is, the better the index value is, the normalization is performed by using the following formula;
Figure FDA0002687523420000033
when the index element is an interval type index, the following formula is used for standardization:
Figure FDA0002687523420000034
wherein [ q ]j1,qj2]Is the interval range of interval type index.
6. The method for evaluating the application effect of domestic electronic devices based on index architecture as claimed in claim 1, wherein the calculation method of the combined weight evaluation vector is as follows:
(1c) the subjective weight W is calculated1And objective weight W2Component block matrix W = [ W =1,W2,]；
(2c) Solving a non-negative fixed square matrix B of the normalized matrix X1The calculation method is as follows:
Figure FDA0002687523420000035
wherein x isi1～xinAnd
Figure FDA0002687523420000041
elements in the normalized matrix X;
(3c) calculating a symmetric matrix WTB1W；
(4c) Calculating a symmetric matrix WTB1The maximum characteristic root of W and the corresponding characteristic vector thereof;
(5c) normalizing the characteristic vector corresponding to the maximum characteristic root to obtain a normalized characteristic column vector T
(6c) Calculating a combined weighting coefficient Wc=[W1,W2}*T
(7c) And a pair of combined weighting coefficients WcCarrying out normalization processing to obtain a combined weight vector;
(8c) and multiplying the original scoring matrix and the combined weight vector to obtain a combined weight evaluation vector.
CN202010981022.6A 2020-09-17 2020-09-17 Domestic electronic device application effect evaluation method based on index system architecture Pending CN112257979A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010981022.6A CN112257979A (en) 2020-09-17 2020-09-17 Domestic electronic device application effect evaluation method based on index system architecture

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010981022.6A CN112257979A (en) 2020-09-17 2020-09-17 Domestic electronic device application effect evaluation method based on index system architecture

Publications (1)

Publication Number Publication Date
CN112257979A true CN112257979A (en) 2021-01-22

Family

ID=74233046

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010981022.6A Pending CN112257979A (en) 2020-09-17 2020-09-17 Domestic electronic device application effect evaluation method based on index system architecture

Country Status (1)

Country Link
CN (1) CN112257979A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113219287A (en) * 2021-05-21 2021-08-06 山东电工电气集团有限公司 Method for carrying out capacitance compatibility rating and weak link positioning on state net core FTU

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107122894A (en) * 2017-04-14 2017-09-01 华中师范大学 The Education Informatization Level appraisal procedure and system of a kind of combination weighting
CN108399481A (en) * 2018-01-15 2018-08-14 上海机电工程研究所 Weapons Systems Effectiveness appraisal procedure based on combination weighting and evaluation method
CN109754161A (en) * 2018-12-14 2019-05-14 中国航天标准化研究所 Military industry developing ability evaluation method based on financial pest analysis frame

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107122894A (en) * 2017-04-14 2017-09-01 华中师范大学 The Education Informatization Level appraisal procedure and system of a kind of combination weighting
CN108399481A (en) * 2018-01-15 2018-08-14 上海机电工程研究所 Weapons Systems Effectiveness appraisal procedure based on combination weighting and evaluation method
CN109754161A (en) * 2018-12-14 2019-05-14 中国航天标准化研究所 Military industry developing ability evaluation method based on financial pest analysis frame

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
陈国华;刘爽;王新华;: "基于组合赋权法的塔式起重机定量风险评估及应用", 起重运输机械, no. 04 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113219287A (en) * 2021-05-21 2021-08-06 山东电工电气集团有限公司 Method for carrying out capacitance compatibility rating and weak link positioning on state net core FTU

Similar Documents

Publication Publication Date Title
US20060010089A1 (en) Hybrid neural network generation system and method
US20070061144A1 (en) Batch statistics process model method and system
Panoutsos et al. A neural-fuzzy modelling framework based on granular computing: Concepts and applications
Asad et al. An approach for software reliability model selection
Ma et al. A hybrid dynamic equivalent using ANN-based boundary matching technique
CN110543932A (en) air conditioner performance prediction method and device based on neural network
CN111582350A (en) Filtering factor optimization AdaBoost method and system based on distance weighted LSSVM
Liu et al. Regression models for estimating product life cycle cost
CN112257979A (en) Domestic electronic device application effect evaluation method based on index system architecture
Abdul Kadhar et al. A stopping criterion for decomposition-based multi-objective evolutionary algorithms
CN111695260B (en) Material performance prediction method and system
CN107705160A (en) A kind of combination econometrics and heuristic intelligent automobile Method for Sales Forecast method and system
Al Nawaiseh et al. Evaluate database management system quality by analytic hierarchy process (ahp) and simple additive weighting (saw) methodolog
Rigoni et al. Metamodels for fast multi-objective optimization: trading off global exploration and local exploitation
CN115081856A (en) Enterprise knowledge management performance evaluation device and method
Barnett et al. Significance tests for regression model hierarchies
Dai et al. Multi-objectives design optimization based on multi-objectives Gaussian processes for System-in-Package
CN110942149B (en) Feature variable selection method based on information change rate and condition mutual information
CN117235477B (en) User group evaluation method and system based on deep neural network
CN112884286B (en) Comprehensive evaluation method for ground test scheme of aviation equipment
Bacarreza et al. Robust Multilevel Design Optimization of Composite Panels
Su et al. The Strategy on Artificial Neural Networks for Predicting Advanced Packaging Reliability Under Small Dataset
Hidayat et al. Sensitivity Analysis of Decision Support Systems for Selection of Achievement Students Using the TOPSIS Method
CN118014445B (en) Technical method and system for evaluating space equilibrium state of water network system
An et al. The Generation Method of Simulation Scenario Sample Space Based on Sensitivity Analysis of Meta-model

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