A multi-dimensional evaluation method for product solutions based on systems engineering
Through the graph theory abstract model of system engineering design expression tools, a multi-dimensional evaluation of product solutions is achieved, and the problem that evaluation methods in the existing technology cannot cover the entire design stage and lack of scalability is solved, which improves conflict detection efficiency and accuracy.
Patent Information
- Application Number
- CN202210151657.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-18
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2042-02-18
AI Technical Summary
The evaluation method of existing technology system engineering product solutions cannot cover the entire design stage process. The rule set structure is complex, the versatility is poor, and the lack of scalability and reusability is affected, which affects the verification efficiency of design conflict detection application and detection execution effect.
A multi-dimensional evaluation method for product solutions based on system engineering is proposed, including demand item conflict detection, demand item clarity evaluation, demand item relationship conflict detection, demand item mapping relationship conflict detection and demand item traceability relationship conflict detection, through the figure The method of abstract model is used to establish evaluations of demand term clarity, relationship clarity, mapping relationship clarity and traceability relationship clarity, and realize comprehensive evaluation of product solutions.
This method can be used from various perspectives of design expression tools, improving versatility, scalability and reusability, reducing engineering volume, significantly improving conflict detection efficiency, shortening completion time to the second and millisecond levels of the computer, avoiding Missed and misrated.
Smart Images

Figure CN114519525B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an evaluation method for MBSE product solutions, and more specifically to a multi-dimensional evaluation method for product solutions based on systems engineering. Background Art
[0002] With the continuous development of information technology, the complexity and scale of modern systems are increasing. Traditional document-centric system engineering can no longer meet the requirements of engineering design due to many defects such as the inability to ensure data consistency, difficulty in change, and difficulty in describing activities. Model-based system engineering is becoming the basis for complex system design. MBSE (Model-Based Systems Engineering) puts the model at the center of the entire system engineering process, and realizes the system design of the product through the continuous evolution and iterative increment of the model from the design stage. Compared with traditional document-centric system engineering, MBSE strengthens the communication between investors and team members, improves the ability to acquire knowledge, has better information traceability, enhances artifact reuse, and reduces development risks. The model-driven system design method "is expected to replace the document-centric method used by system engineers for many years, and will eventually be fully integrated into the system engineering process."
[0003] Currently, there are many languages, tools and technologies for system engineering modeling, such as behavioral diagrams, IDEF0, N2 diagrams, etc. However, these modeling methods use different symbols and semantics, and cannot interoperate and reuse each other. It is based on this background that the system engineering standard modeling language SysML (Systems Modeling Language) was proposed by the International System Engineering Association INCOSE and the Object Management Group OMG in 2004. It has been widely used in engineering practice and has achieved good application results. In particular, the application of interdisciplinary and complex product solutions has achieved outstanding results.
[0004] However, the MBSE tool methods currently used in the market for product solutions focus on sorting out and recording the design process, and lack the ability to automatically detect and evaluate the determined product solutions. For example, Chinese patent publication number CN112558498A discloses a method for joint design and verification of MBSE aircraft systems based on a data pool. Secondly, there is currently no evaluation method for MBSE product solutions, nor is there a ready-made solution. It is necessary to consider the applicability and complexity and propose a multi-dimensional evaluation method for product solutions.
[0005] Considering the applicability and complexity of the multi-dimensional evaluation method of product solutions, conflict detection technology can be introduced into the multi-dimensional evaluation scheme of product solutions. At present, there are four main conflict detection methods: (1) Using formal reasoning to transform the design into a logical description and characterize its object attributes, thereby completing the design conflict detection technology. The difficulty lies in the transformation of the design into a logical description and the construction of the reasoning engine; (2) Using state transition as the design modeling principle, using state model verification, transforming the design into state transition, and judging the design conflict problem by detecting the state. The difficulty lies in the design being converted into state transition without omission and objectively; (3) Using the goal as the design modeling principle, using the goal semantics and its heuristic rules to perform design conflict detection. The difficulty lies in the construction of the goal semantics and heuristic rules and the integrity of the semantics and rule set; (4) Using the graph grammar, by giving the properties and semantics of the graph, detecting and processing the conflict problem of the design description expressed by the graph. The difficulty lies in the accuracy and completeness of the definitions of various different business fields. The above four methods target different aspects of the expression of system engineering design. Based on the characteristics of the expression form of this aspect, models are established and rules are defined, and implementation is given. Therefore, modeling using the above four methods can only establish a limited number of expressions specified by the methodology. Each expression form requires the establishment of models and rules, and the relationship between them is fixed. Sometimes it is also necessary to establish a semantic library.
[0006] It can be seen that the limitations in the expression of system engineering design often lead to the lack of integrity of the expression of system engineering design, and there is no consideration of the standardization, systematicness and complexity of the expression of system engineering design. However, system engineering design analysis is a hierarchical analysis process, from the initial design stage, a relatively vague strategic concept is proposed, to the mid-design stage, the operational design is gradually formed, the capability design is launched, and finally to the final design stage, a relatively complete system design scheme is obtained. Throughout the entire process, there are requirements for obtaining and defining system engineering design, so the system engineering design conflict detection method must be able to cover the entire design phase process; secondly, the modeling granularity of the above four methods is too fine, and they even have to rely on third-party analysis and design tools. The design conversion and expression workload is large, the rule set is complex to construct, and the versatility is poor. The engineering workload from conception to application is equivalent to the development of a specialized tool, and it lacks scalability and reusability, which greatly affects the design conflict detection application and the verification of the detection execution effect; finally, the use of semantic libraries will be limited by the specific business to be expressed. Although conflict detection is limited to system engineering design, due to the differences in system engineering design business, the required semantic libraries are also very different, and there is no way to evaluate the correctness and completeness of the built semantic library on the product solution. Summary of the invention
[0007] The technical problem to be solved by the present invention is that the evaluation method of product solutions for system engineering in the prior art cannot cover the entire design stage process, the rule set has a complex structure, poor versatility, lacks scalability and reusability, and greatly affects the efficiency of design conflict detection application and verification of detection execution effects.
[0008] The present invention solves the above technical problems through the following technical means: a multi-dimensional evaluation method for product solutions based on system engineering, comprising the following steps:
[0009] a1. Requirement item conflict detection;
[0010] a2. Evaluate the clarity of requirements based on the requirements conflict detection results;
[0011] b1. Detection of conflict between requirement items;
[0012] b2. Based on the requirements item relationship conflict detection results, evaluate the clarity of the requirements item relationship;
[0013] c1. Detection of conflict in requirement item mapping relationships;
[0014] c2. Based on the conflict detection results of requirement item mapping relationships, evaluate the clarity of requirement item mapping relationships;
[0015] d1. Detection of conflict in demand item traceability relationship;
[0016] d2. Based on the conflict detection results of the requirement item traceability relationship, evaluate the clarity of the requirement item traceability relationship;
[0017] e. Comprehensive evaluation results of product solutions are obtained by comprehensively evaluating the clarity of requirement items, the clarity of requirement item relationships, the clarity of requirement item mapping relationships, and the clarity of requirement item traceability relationships.
[0018] The present invention abstracts and classifies the characteristics of currently popular system engineering design expression tools, summarizes and establishes four graph theory abstract models, namely, requirement item clarity evaluation, requirement item relationship clarity evaluation, requirement item mapping relationship clarity evaluation and requirement item traceability relationship clarity evaluation, thereby proposing a multi-dimensional evaluation method for product solutions based on conflict detection, covering the entire design stage process, which can not only realize a complete inspection and quantitative evaluation of the product solution content, but also the model abstracted by the graph theory can be used in various perspectives of the design expression tool, thereby improving the versatility, scalability and reusability, reducing the amount of engineering, greatly improving the efficiency of conflict detection, and shortening the completion time from manual hours and levels to computer seconds and milliseconds.
[0019] Furthermore, the step a1 comprises:
[0020] Design item set a = {a i |i=1,2,…,n}, where a i is the i-th design item, n is the number of design items, and design item a i The relationship vector matrix with other design items is (a i1 ,a i2 ,…,a in ), the values of each element in the relationship vector matrix follow the following rules:
[0021]
[0022] Establish the relationship vector matrix of each design item. If there is a design item that cannot establish a vector matrix, it means that there is a conflict in the design item; compare any two design item vectors a in turn i ,a j ,in is a i The vector matrix of is a j In the comparison process, in addition to the vector matrix elements corresponding to the two design items being compared, that is, Except for the corresponding position value, the other element values of the vector are the same, which means that there is a design conflict between the two design items. If it is not 0, then find out whether there are elements in other positions of the two design vectors that are not 0 at the same time. If so, it means that there is a conflict in the compared design items; Represent the design term vector a i ,a j The i-th column element of Represent the design vector a j ,a i The j-th column element of .
[0023] Furthermore, the step a2 comprises:
[0024] Using the formula Obtain the clarity evaluation results of the requirement items, M1 is the total number of requirement items in the requirement item conflict detection process, and m1 is the total number of requirement items that have requirement item conflicts.
[0025] Furthermore, the step b1 includes:
[0026] Establish an n×n-order design item relationship matrix A, where n is the number of design items. Determine the value of each element in the matrix A. If the design item represented by the i-th row (1≤i≤n) is the parent design of the design item represented by the j-th column (1≤j≤n), then A(i,j)=1, otherwise A(i,j)=0, thereby obtaining all element values in the design item relationship matrix A;
[0027] Determine in turn whether the symmetric elements in the matrix A are all 1 at the same time, that is, whether there is A(i,j)×A(j,i)=1 in the matrix. If so, it means that there is a conflict between the design item corresponding to i and the design item corresponding to j. Next, find the relationship conflict problem caused by the transitivity of the design, adjust the design item relationship matrix A, set the elements whose symmetric elements found in the previous step are all 1 to 0, that is, A(i,j)=A(j,i)=0, initialize the two-dimensional zero matrix A0 and the unit matrix A1 of the same order as the design item relationship matrix A, and loop n-1 times to calculate A1=A1×A and A0=A0+A1. Finally Among them A T is the transpose of A, It means that the elements at corresponding positions of the matrix are multiplied. If the final A0(i,j) is not 0, it means that there is still a conflict between the design item corresponding to i and the design item corresponding to j.
[0028] Furthermore, the step b2 comprises:
[0029] Using the formula Obtain the evaluation results of the clarity of the relationship between requirements, M2 is the total number of requirements in the process of detecting the conflict between requirements, and m2 is the total number of requirements with conflicting relationship between requirements.
[0030] Furthermore, the step c1 comprises:
[0031] Determine the value of each element in the design item relationship matrix. If the design item represented by the i-th row (1≤i≤n) is the parent design of the design item represented by the j-th column (1≤j≤n), then A(i,j)=1, otherwise A(i,j)=0. In this way, all the element values in the design item relationship matrix A are obtained, and another perspective matrix is constructed. The relationship between each design item is n B ×n B The two-dimensional full matrix B is obtained, and the values of each element in the matrix B are judged. If the design item represented by the i-th row (1≤i≤n) is the parent design of the design item represented by the j-th column (1≤j≤n), then B(i,j)=1, otherwise B(i,j)=0. In this way, all the element values in the matrix B are obtained to construct the n-th perspective of the two perspectives. A ×n B Mapping relationship matrix AB, judge the value of each element in the matrix AB, if the i-th row (1≤i≤n A ) represents the design item and the jth column (1≤j≤n B ) indicates that there is a design mapping relationship between the design items, then AB(i,j)=1, otherwise AB(i,j)=0, thereby obtaining all element values in the matrix AB;
[0032] Find any non-zero element in the mapping relationship matrix AB, such as (i, j); then calculate the two matrices, AB ij =A i ×B j and Among them A i It represents the column vector matrix of the i-th column extracted from matrix A, B j Represents the row vector matrix of the jth row extracted from matrix B, AB ij represents the matrix obtained by multiplying the column vector of the i-th column extracted from matrix A with the row vector of the j-th row extracted from matrix B. Indicates the multiplication of the elements at the corresponding positions of the matrix, traversing F AB For each element in F AB (i,j) = 1, and AB(i,j) = 1, it indicates that there is a conflict in the design mapping relationship corresponding to the position. According to the above steps, the other non-zero elements in the matrix AB are traversed in turn for calculation and judgment to obtain the results of all conflicts with mapping relationships.
[0033] Furthermore, the step c2 comprises:
[0034] Using the formula Obtain the clarity evaluation result of the requirement item mapping relationship, M3 is the total number of requirement items in the requirement item mapping relationship conflict detection process, and m3 is the total number of requirement items that have requirement item mapping relationship conflicts.
[0035] Furthermore, the step d1 comprises:
[0036] Select the same perspective or the same perspective, find two design items to be compared with k traceability chain perspective relationships, assuming H is the hypothesis space, H = {similar, dissimilar}, h∈H, μ i is the probability on k observation space O, where 1≤i≤k, for μ i The calculation method uses a depth-first method to traverse and compare the differences between the two perspective models. If the node content is the same or similar, it is 1, otherwise it is 0. Where a and b are the two nodes being compared, and the number of identical or similar nodes is The number of nodes is N, then Finally, the similarity calculation result of DS inference is obtained If m(A) is greater than the empirical threshold of 60%, it is determined that there is a conflict between the two design items.
[0037] Furthermore, the step d2 comprises:
[0038] Formula 1-m(A) is used to obtain the clarity evaluation result of the requirement item traceability relationship.
[0039] Furthermore, the step e comprises:
[0040] Using the formula Get comprehensive evaluation results of product solutions.
[0041] The advantages of the present invention are:
[0042] (1) The present invention abstracts and classifies the characteristics of currently popular system engineering design expression tools, summarizes and establishes four graph theory abstract models, namely, requirement item clarity evaluation, requirement item relationship clarity evaluation, requirement item mapping relationship clarity evaluation and requirement item traceability relationship clarity evaluation, thereby proposing a multi-dimensional evaluation method for product solutions based on conflict detection, covering the entire design stage process. Not only can it realize a complete inspection and quantitative evaluation of the product solution content, but the model abstracted by the graph theory method can also be used in various perspectives of the design expression tool, thereby improving the versatility, scalability and reusability, reducing the amount of engineering work, greatly improving the efficiency of conflict detection, and shortening the completion time from hours and levels manually to seconds and milliseconds by computer.
[0043] (2) In view of the disadvantage of existing evaluation methods that use semantic libraries, the present invention proposes a method of using matrices and evidence inference expressions to implement graph theory. Without the need for additional resources, the method can achieve multi-dimensional evaluation of design product solutions before the system design is run, relying only on the characteristics and laws of the elements in the system engineering design expression matrix from a perspective, thereby avoiding omissions and mis-evaluations in the multi-dimensional evaluation of product solutions. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 A flowchart of a multi-dimensional evaluation method for a product solution based on system engineering provided by an embodiment of the present invention;
[0045] Figure 2 A schematic diagram of a demand item conflict model in a multi-dimensional evaluation method for a system engineering-based product solution provided by an embodiment of the present invention; Figure 2 In the figure, a, b, and c are demand item conflict models for three different situations;
[0046] Figure 3 A schematic diagram of a demand item relationship conflict model in a multi-dimensional evaluation method for a system engineering-based product solution provided by an embodiment of the present invention;
[0047] Figure 4 A schematic diagram of a conflict model of requirement mapping relationships in a multi-dimensional evaluation method for a system engineering-based product solution provided by an embodiment of the present invention;
[0048] Figure 5A schematic diagram of a conflict model of requirement traceability relationships in a multi-dimensional evaluation method for a system engineering-based product solution provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0050] like Figure 1 As shown, a multi-dimensional evaluation method for product solutions based on system engineering includes the following steps:
[0051] a1. Requirement item conflict detection;
[0052] Design item set a = {a i |i=1,2,…,n}, where a i is the i-th design item, n is the number of design items, for a specific design item, the design item a i The relationship vector matrix with other design items is (a i1 ,a i2 ,…,a in ), the values of each element in the relationship vector matrix follow the following rules:
[0053]
[0054] Establish the relationship vector matrix of each design item. If there is a design item that cannot establish a vector matrix, it means that there is a conflict in the design item; compare any two design item vectors a in turn i ,a j ,in is a i The vector matrix of is a j In the comparison process, in addition to the vector matrix elements corresponding to the two design items being compared, that is, Except for the corresponding position value, the other element values of the vector are the same, which means that there is a design conflict between the two design items. If it is not 0, then find out whether there are elements in other positions of the two design vectors that are not 0 at the same time. If so, it means that there is a conflict in the compared design items; Represent the design term vector a i ,a j The i-th column element of Represent the design vector a j ,a i The j-th column element of .
[0055] a2. Evaluate the clarity of requirements based on the requirements conflict detection results;
[0056] Using the formula Obtain the clarity evaluation results of the requirement items, M1 is the total number of requirement items in the requirement item conflict detection process, and m1 is the total number of requirement items that have requirement item conflicts.
[0057] b1. Detection of conflict between requirement items;
[0058] Establish an n×n-order design item relationship matrix A, where n is the number of design items. Determine the value of each element in the matrix A. If the design item represented by the i-th row (1≤i≤n) is the parent design of the design item represented by the j-th column (1≤j≤n), then A(i,j)=1, otherwise A(i,j)=0, thereby obtaining all element values in the design item relationship matrix A;
[0059] Determine in turn whether the symmetric elements in the matrix A are all 1 at the same time, that is, whether there is A(i,j)×A(j,i)=1 in the matrix. If so, it means that there is a conflict between the design item corresponding to i and the design item corresponding to j. Next, find the relationship conflict problem caused by the transitivity of the design, adjust the design item relationship matrix A, set the elements whose symmetric elements found in the previous step are all 1 to 0, that is, A(i,j)=A(j,i)=0, initialize the two-dimensional zero matrix A0 and the unit matrix A1 of the same order as the design item relationship matrix A, and loop n-1 times to calculate A1=A1×A and A0=A0+A1. Finally Among them A T is the transpose of A, It means that the elements at corresponding positions of the matrix are multiplied. If the final A0(i,j) is not 0, it means that there is still a conflict between the design item corresponding to i and the design item corresponding to j.
[0060] b2. Based on the requirements item relationship conflict detection results, evaluate the clarity of the requirements item relationship;
[0061] Using the formula Obtain the evaluation results of the clarity of the relationship between requirements, M2 is the total number of requirements in the process of detecting the conflict between requirements, and m2 is the total number of requirements with conflicting relationship between requirements.
[0062] c1. Detection of conflict in requirement item mapping relationships;
[0063] Determine the value of each element in the design item relationship matrix. If the design item represented by the i-th row (1≤i≤n) is the parent design of the design item represented by the j-th column (1≤j≤n), then A(i,j)=1, otherwise A(i,j)=0. In this way, all the element values in the design item relationship matrix A are obtained, and another perspective matrix is constructed. The relationship between each design item is n B ×n B The two-dimensional full matrix B, where n B Represents the number of rows and columns of matrix B, and determines the value of each element in the matrix B. If the design item represented by the i-th row (1≤i≤n) is the parent design of the design item represented by the j-th column (1≤j≤n), then B(i,j)=1, otherwise B(i,j)=0. In this way, all element values in matrix B are obtained to construct n from two perspectives. A ×n B Mapping relationship matrix AB, where n A Represents the number of rows of matrix AB, n B Indicates the number of columns of matrix AB, and determines the value of each element in the matrix AB. If the i-th row (1≤i≤n A ) represents the design item and the jth column (1≤j≤n B ) indicates that there is a design mapping relationship between the design items, then AB(i,j)=1, otherwise AB(i,j)=0, thereby obtaining all element values in the matrix AB;
[0064] Find any non-zero element in the mapping relationship matrix AB, such as (i, j); then calculate the two matrices, AB ij =A i ×B j and Among them A i It represents the column vector matrix of the i-th column extracted from matrix A, B j Represents the row vector matrix of the jth row extracted from matrix B, AB ij represents the matrix obtained by multiplying the column vector of the i-th column extracted from matrix A with the row vector of the j-th row extracted from matrix B. Indicates the multiplication of the elements at the corresponding positions of the matrix, traversing F AB For each element in F AB (i,j) = 1, and AB(i,j) = 1, it indicates that there is a conflict in the design mapping relationship corresponding to the position. According to the above steps, the other non-zero elements in the matrix AB are traversed in turn for calculation and judgment to obtain the results of all conflicts with mapping relationships.
[0065] c2. Based on the conflict detection results of requirement item mapping relationships, evaluate the clarity of requirement item mapping relationships;
[0066] Using the formula Obtain the clarity evaluation result of the requirement item mapping relationship, M3 is the total number of requirement items in the requirement item mapping relationship conflict detection process, and m3 is the total number of requirement items that have requirement item mapping relationship conflicts.
[0067] d1. Detection of conflict in demand item traceability relationship;
[0068] Dempster-Shafer synthesis formula: Let m1 and m2 be two mass functions on H, then m(φ)=0, is a mass function, where is the normalization coefficient.
[0069] Select the same perspective or the same perspective, find two design items to be compared with k traceability chain perspective relationships, assuming H is the hypothesis space, H = {similar, dissimilar}, h∈H, μ i is the probability on k observation space O, where 1≤i≤k, for μ i The calculation method uses a depth-first method to traverse and compare the differences between the two perspective models. If the node content is the same or similar, it is 1, otherwise it is 0. Where a and b are the two nodes being compared, and the number of identical or similar nodes is The number of nodes is N, then Finally, the similarity calculation results of DS inference are obtained If m(A) is greater than the empirical threshold of 60%, it is determined that there is a conflict between the two design items.
[0070] d2. Based on the conflict detection results of the requirement item traceability relationship, the clarity of the requirement item traceability relationship is evaluated; specifically:
[0071] Formula 1-m(A) is used to obtain the clarity evaluation result of the requirement item traceability relationship.
[0072] e. Comprehensive evaluation results of product solutions are obtained by comprehensively evaluating the clarity of requirements, the clarity of relationships between requirements, the clarity of mapping relationships between requirements, and the clarity of traceability relationships between requirements. Specifically:
[0073] Using the formula Get comprehensive evaluation results of product solutions.
[0074] The following examples describe the working process of the present invention in detail:
[0075] 1. Calculation Figure 2 Requirements conflict detection and requirements clarity evaluation are as follows:
[0076] like Figure 2 As shown in (a), design items a2 and a3 are in a superior-subordinate relationship, so design items a2 and a3 may conflict; Figure 2 As shown in (b), design item a2 is the parent design of design items a4, a5 and a3, and design item a3 is the parent design of design item a5. Then design items a2 and a3 may conflict. Figure 2 As shown in (c), design items a2 and a3 are both the parent designs of design items a4 and a5, so design items a2 and a3 may conflict.
[0077] Figure 2 Take (a) as an example. Design item a2 and design item a3 are in a superior-subordinate relationship. Therefore, in the relationship vector of design item a2, its relationship value with design item a3 cannot be expressed. Similarly, in the relationship vector of a3, its relationship with a2 cannot be expressed. The two relationship vectors cannot be established according to the rule method. Therefore, it is possible that there is a design conflict between these two design items a2 and a3. The clarity of the requirement item is calculated to be 60%.
[0078] Figure 2 (b) as an example. The relationship vector of design item a2 is expressed as (-1 1 1 1 1), and the relationship vector of design item a3 is expressed as (-1 -1 1 0 1). 12 ,a2 13 ,a3 12 ,a3 13 If it is not 0, a2 is the superior design of a3 and a5, and a3 is the superior design of a5, which means that there is a possibility of design conflict between a2 and a3. The clarity of the requirement item is 60%.
[0079] Figure 2 Take (c) as an example. The relationship vector of design item a2 is (-1 1 0 1 1), and the relationship vector of design item a3 is (-1 0 1 11). Compare the two design items a2 and a3, and remove the vector element a2. 12 ,a2 13 ,a3 12 ,a3 13 Except for , the values of other vector elements are the same, indicating that a2 and a3 have the same upper and lower level designs, so there is a possibility of conflict. The clarity of its requirement items is 60%.
[0080] The comprehensive clarity is (60%+60%+60%) / 3=60%.
[0081] 2. Calculation Figure 3 The requirements item relationship conflict detection and requirement item relationship clarity evaluation are as follows:
[0082] like Figure 3 As shown, it is defined that design item a1 is the parent design of design item a3, and design item a3 is the parent design of design item a5. Then design item a5 should not be the parent design of design item a1, otherwise there will be a conflict between the design item relationships.
[0083] Figure 3 Take as an example, and establish the relationship matrix A of the demand items. Since A(1,2)=A(2,1)=1, it can be inferred that there is a conflict between design items a1 and a2. Its value is as follows:
[0084]
[0085] Furthermore, the design relationship conflicts caused by the design relationship transfer are found, the relationship matrix A between the design items is adjusted, and the zero matrix A0 and the unit matrix A1 of the same order as A are initialized, and their values are as follows:
[0086]
[0087] After 4 cycles of A1=A1×A and A0=A0+A1 calculations, Calculate and get the final result as
[0088] Therefore, there are design relationship conflicts between a1 and a5, a3 and a1, and a5 and a3. The requirement item relationship clarity is 40%.
[0089] 3. Calculation Figure 4 The requirements mapping relationship conflict detection and requirements mapping relationship clarity evaluation are as follows:
[0090] like Figure 4 As shown, there is a mapping relationship between design set A = {a1, a2, a3, a4, a5} and design set B = {b1, b2, b3, b4, b5}. If in set A, design item a2 is defined as the parent design of design item a4. In set B, design item b3 is defined as the parent design of design items b4 and b5. If design items a2 and b4 and b5 have a mapping relationship between the two sets, and design item a4 and b3 have a mapping relationship, then there is a conflict in the mapping relationship between a2 and a4 in set A and b3, b4 and b5 in set B.
[0091] Figure 3 For example, a 5×5 matrix A and a 5×5 matrix B representing the relationship between design items and a 5×5 matrix AB representing the mapping relationship between two perspectives are established, and their values are as follows:
[0092]
[0093] Find the elements in matrix AB that are not 0. Take AB(4,3) as an example and calculate AB 4,3 =A4×B3 and The specific calculation process is as follows:
[0094]
[0095]
[0096] A4 represents the column vector of the fourth column extracted from matrix A. B3 means extracting the row vector (0 00 1 1) of the third row from matrix B.
[0097] Because F 4,3 (2,4)=1 and AB(2,4)=1, F 4,3 (2,5)=1 and AB(2,5)=1, so it can be concluded that the mapping relationship between a4 and b3 and the mapping relationship between a2 and b4 and b5 conflict. According to the above steps, traverse the other non-zero elements in the matrix AB in turn to calculate and judge whether all mapping relationship items conflict. The clarity of the demand item relationship is 50%.
[0098] 4. Calculation Figure 5 The conflict detection of requirement item traceability relationship and the clarity evaluation of requirement item traceability relationship are as follows:
[0099] like Figure 5 As shown in the figure, under the same perspective, design items a1 and b1, a1 is composed of design items aa1, aa2, aa3 and aa4, aa2 is composed of design items aaa1, aaa2, aaa3 and aaa4, b1 is composed of design items aa1, bb2, aa3 and aa4, bb2 is composed of design items aaa1, aaa2, aaa3 and bbb4. By comparing the similarity of the composition structure and content between design items a1 and b1, it is inferred that there is a possibility of conflict between a1 and b1.
[0100] Figure 4 For example, design items a1 and b1 each have two traceable perspective model structures. Assume H is the hypothesis space, H = {similar, dissimilar}, h∈H, μ i is the probability on the 2 observation spaces O, where 1≤i≤k, k=2. i The calculation method uses the depth-first method to traverse and compare the two models. If the node content is the same or similar, it is 1, otherwise it is 0. Where a and b are the two nodes being compared, and the number of identical or similar nodes is The number of nodes is N, so but According to the DS synthesis formula, Its value is greater than 60%, so there is a possibility of conflict between design items a1 and b1. The clarity of the traceability relationship of the requirement item is 25%.
[0101] 5. Assumptions Figure 2 , Figure 3 , Figure 4 , Figure 5 For the same product solution, based on the four-dimensional evaluation indicators of requirement item conflict detection evaluation dimension, requirement item relationship conflict detection evaluation dimension, requirement item mapping relationship conflict detection evaluation dimension and requirement item traceability relationship conflict detection evaluation dimension, the comprehensive evaluation index of this product solution is (60%+40%+50%+25%) / 4=43.75%.
[0102] Through the above technical scheme, the present invention aims at the problems of the professionalism of system engineering design, the diversity of expression forms, the poor universality of the current multi-dimensional evaluation method of product solutions, and the long construction period and difficulty of the multi-dimensional evaluation semantic library of product solutions. By studying the characteristics of the currently popular system engineering design expression tool based on SysML, abstraction and classification, the abstract modeling based on graph theory is proposed, which simplifies the difficulty of design conversion expression and improves the intuitiveness of design expression. The model abstracted by graph theory can be used in various perspectives of design expression tools, which improves the universality, scalability and reusability of the multi-dimensional evaluation method of product solutions, reduces the amount of engineering, and accelerates the application of multi-dimensional evaluation of design product solutions and the verification of evaluation execution effect. In view of the disadvantages of the semantic library, a method of realizing graph theory by matrix and evidence inference expression is proposed. Without the need for additional resources, only relying on the characteristics and laws of each element entity in the system engineering design expression matrix under the perspective, the multi-dimensional evaluation of the design product solution is realized before the system design is run, avoiding the omission and misevaluation of the multi-dimensional evaluation of the product solution. This automatic detection and evaluation method not only enables a thorough inspection and quantitative evaluation of product solution contents, but also greatly improves the efficiency of conflict detection, shortening the completion time from hours and levels manually to seconds and milliseconds by computer.
[0103] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A multi-dimensional evaluation method for product solutions based on systems engineering, characterized in that: The following steps are involved: a1. Requirement item conflict detection; The step a1 comprises: Design Item Collection ,in is the i-th design item, n is the number of design items, and The relationship vector matrix with other design items is , the values of each element in the relationship vector matrix follow the following rules: ; Establish the relationship vector matrix of each design item. If there is a design item that cannot establish a vector matrix, it means that there is a conflict in the design item. Compare the vectors of any two design items in turn. ,in yes The vector matrix of yes In the comparison process, in addition to the vector matrix elements corresponding to the two design items being compared, that is, Except for the corresponding position value, the other element values of the vector are the same, which means that there is a design conflict between the two design items. If it is not 0, then find out whether there are elements in other positions of the two design vectors that are not 0 at the same time. If so, it means that there is a conflict in the compared design items; Represent the design term vectors No. i Column elements, Represent the design vectors No. j Column elements; a2. Evaluate the clarity of requirements based on the requirements conflict detection results; b1. Detection of conflict between requirement items; b2. Based on the requirements item relationship conflict detection results, evaluate the clarity of the requirements item relationship; c1. Detection of conflict in requirement item mapping relationships; c2. Based on the conflict detection results of requirement item mapping relationships, evaluate the clarity of requirement item mapping relationships; d1. Detection of conflict in demand item traceability relationship; The step d1 comprises: Select the same perspective or similar perspective and find the k The two design items of the traceability chain perspective relationship are assumed to be H is the hypothesis space, H ={similar, dissimilar}, h∈H , yes k Observation Space O The probability of ,for The calculation method uses a depth-first method to traverse and compare the differences between the two perspective models. If the node content is the same or similar, it is 1, otherwise it is 0. , where a and b are the two nodes being compared, and the number of identical or similar nodes is , where the number of nodes is N, then , and finally the similarity calculation result of DS inference is obtained ,like If it is greater than the empirical threshold of 60%, it is determined that there is a conflict between the two design items; d2. Based on the conflict detection results of the requirement item traceability relationship, evaluate the clarity of the requirement item traceability relationship; e. Comprehensive evaluation results of product solutions are obtained by integrating the results of requirement item clarity evaluation, requirement item relationship clarity evaluation, requirement item mapping relationship clarity evaluation and requirement item traceability relationship clarity evaluation; The step e comprises: Using the formula Obtain comprehensive evaluation results of product solutions; is the total number of requirements in the requirement conflict detection process, is the total number of requirement items with requirement item conflicts, is the total number of requirement items in the requirement item relationship conflict detection process, is the total number of requirement items with requirement item relationship conflicts, is the total number of requirement items in the requirement item mapping relationship conflict detection process, The total number of requirement items with requirement item mapping conflicts.
2. According to claim 1, a multi-dimensional evaluation method for product solutions based on system engineering is characterized in that: The step a2 comprises: Using the formula Obtain the clarity evaluation results of the requirement items.
3. The multi-dimensional evaluation method for product solutions based on system engineering according to claim 2 is characterized in that: The step b1 comprises: Establish The design item relationship matrix A of order 1 is n, where n is the number of design items. The values of each element in the matrix A are judged. If the design item represented by the i-th row is the parent design of the design item represented by the j-th column, then ,otherwise , thereby obtaining all element values of the design item relationship matrix A, , ; Sequential judgment matrix A Are the symmetric elements in 1 all at the same time, that is, does the matrix exist? If yes, please explain i The corresponding design items and j The corresponding design items have conflicts; next, find the relationship conflict problem caused by the transitivity of the design, adjust the design item relationship matrix A, and set the elements whose symmetric elements found in the previous step are all 1 to 0, that is, , initialize the two-dimensional zero matrix of the same order as the design term relationship matrix A and the identity matrix , cycle n -1 calculation and ,at last ,in yes The transpose of Indicates that the corresponding position elements of the matrix are multiplied. If the final result is If it is not 0, it means i The corresponding design items and j There are still conflicts between corresponding design items.
4. The multi-dimensional evaluation method for product solutions based on system engineering according to claim 3 is characterized in that: The step b2 comprises: Using the formula Obtain the evaluation results of the relationship clarity of requirement items.
5. The multi-dimensional evaluation method for product solutions based on system engineering according to claim 4 is characterized in that: The step c1 comprises: Determine the value of each element in the design item relationship matrix. If the design item represented by the i-th row is the parent design of the design item represented by the j-th column, then ,otherwise , thereby obtaining all element values of the design item relationship matrix A, and constructing another perspective matrix, the relationship between each design item The two-dimensional matrix B is used to determine the value of each element in the matrix B. , ; If the design item represented by the i-th row is the parent design of the design item represented by the j-th column, then ,otherwise , thereby obtaining all element values in matrix B and constructing two perspectives Mapping relationship matrix AB, determine the value of each element in the matrix AB, , ; If there is a design mapping relationship between the design item represented by the i-th row and the design item represented by the j-th column, then ,otherwise , thereby obtaining all element values in matrix AB, , ; Find the mapping relationship matrix AB Any non-zero element in ( i , j ) ; Then calculate the two matrices, and ,in Representation Matrix A Extract i a column vector matrix of columns, Represents the matrix B Extract j The row vector matrix of rows, Representation based on matrix A The extracted i Column vectors and matrices of columns B The extracted j The matrix obtained by multiplying the row vectors of the rows, Indicates the multiplication of elements at corresponding positions of the matrix, traversing If the elements in ,and , it indicates that there is a conflict in the design mapping relationship corresponding to the position. According to the above steps, traverse the matrix in turn. AB Calculate and judge other non-zero elements in the to obtain the results of all conflicts with mapping relationships.
6. The multi-dimensional evaluation method for product solutions based on system engineering according to claim 5 is characterized in that: The step c2 comprises: Using the formula Obtain the clarity evaluation results of the requirement item mapping relationship.
7. The multi-dimensional evaluation method for product solutions based on system engineering according to claim 1 is characterized in that: The step d2 comprises: Using the formula Obtain the clarity evaluation results of the traceability relationship of requirement items.
