Project system capability assessment method and device based on heterogeneous preference

Abstract

The invention provides a project system capability evaluation method and device based on heterogeneous preferences, and the method comprises the steps: obtaining a numerical evaluation matrix and a comparison evaluation matrix of a plurality of experts for a plurality of to-be-evaluated projects, converting the comparison evaluation matrix according to an information conversion operator, adding the converted comparison evaluation matrix to a numerical evaluation matrix given by the same expert, and obtaining the capability of a project system. The heterogeneous preference information of the experts in different fields is effectively aggregated, and the problem that the assessment information is difficult to unify due to different specialist fields of the experts is solved. According to the method, expert weights are determined based on a double-entropy target, a complete evaluation matrix of multiple experts is aggregated, and strategic target weights are determined based on the double-entropy target; according to the strategic target weight and the comprehensive evaluation matrix, the expected performance conditions of the multiple to-be-evaluated projects on the strategic target are determined, so that scientific evaluation of the project execution capability is realized, and a solid data support is provided for execution capability deviation evaluation and optimization adjustment of subsequent project combinations.

Description

Technical Field

[0001] This invention belongs to the field of project capability assessment, specifically relating to a method and apparatus for project system capability assessment based on heterogeneous preferences. Background Technology

[0002] In the implementation of military strategic planning, accurate assessment of project execution capabilities is a crucial prerequisite for ensuring the achievement of strategic objectives. However, due to the wide range of fields involved in strategic planning, there are significant differences in the assessment preferences of experts in different fields, which poses a significant challenge to the comprehensive assessment of project execution capabilities. How to effectively aggregate this heterogeneous preference information has become an urgent problem to be solved in the pre-assessment stage of strategic planning. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a method and apparatus for evaluating the capabilities of a project system based on heterogeneous preferences.

[0004] The present invention includes a method for evaluating the capabilities of a project system based on heterogeneous preferences, comprising: obtaining numerical evaluation matrices and comparative evaluation matrices from multiple experts for multiple projects to be evaluated, wherein the numerical evaluation matrices include the experts' evaluation values ​​of the projects to be evaluated on strategic objectives, and the comparative evaluation matrices include the experts' comparative priorities of the multiple projects to be evaluated on strategic objectives; transforming the comparative evaluation matrices according to an information transformation operator to obtain a supplementary matrix, and adding the supplementary matrix to the numerical evaluation matrix given by the same expert to obtain a complete evaluation matrix for each expert; determining expert weights based on a dual-entropy objective, and aggregating the complete evaluation matrices of multiple experts to obtain a comprehensive evaluation matrix; determining strategic objective weights based on the comprehensive evaluation matrix and the dual-entropy objective; and determining the expected performance of the multiple projects to be evaluated on the strategic objectives based on the strategic objective weights and the comprehensive evaluation matrix.

[0005] Optionally, the aforementioned dual-entropy objectives include the minimum total entropy objective and the maximum symmetric cross-entropy objective.

[0006] Optionally, the above information conversion operators include:

[0007] in, For the above information transformation operator, it represents the first... The fuzzy evaluation values ​​of the projects to be evaluated are used to construct the aforementioned supplementary matrix; This represents the total number of the aforementioned items to be evaluated. For the first The items to be evaluated and the first The membership degree between the items to be evaluated is calculated based on the evaluation language term set and the language scale function.

[0008] Optionally, the above method for determining expert weights based on a dual-entropy objective includes: calculating the fuzzy entropy and hesitation entropy of each of the above complete evaluation matrices, and combining the fuzzy entropy and hesitation entropy to obtain the total entropy of each of the above complete evaluation matrices; and calculating the symmetric cross entropy of each of the above complete evaluation matrices; and determining the expert weights based on the total entropy and symmetric cross entropy of each of the above complete evaluation matrices, wherein the total entropy is inversely correlated with the expert weights, and the symmetric cross entropy is positively correlated with the expert weights.

[0009] Optionally, the formula for calculating the expert weights mentioned above includes:

[0010]

[0011]

[0012] in, For the first The aforementioned expert weights for the complete evaluation matrix described above. The first sub-weight is calculated based on the total entropy. The second sub-weight is calculated based on the symmetric cross-entropy. This is a compromise factor. The total number of the above complete evaluation matrices, For the first The total entropy of the above complete evaluation matrices For the first The above complete evaluation matrix and the first Symmetric cross-entropy between the above complete evaluation matrices.

[0013] Optionally, the determination of strategic target weights based on the comprehensive evaluation matrix and the dual-entropy objective includes: constructing a set of target formulas for strategic target weights that minimizes the total entropy of the comprehensive evaluation matrix and maximizes the symmetric cross entropy of the comprehensive evaluation matrix; and solving the set of target formulas using the Lagrange multiplier method to obtain the strategic target weights.

[0014] Optionally, the above-mentioned determination of the expected performance of multiple projects to be evaluated on the strategic objectives based on the strategic objective weights and the comprehensive evaluation matrix includes: determining a comprehensive preference index among each project to be evaluated based on the strategic objective weights and the comprehensive evaluation matrix; determining the positive and negative ranking flows of each project to be evaluated based on the comprehensive preference index; determining the net flow of each project to be evaluated based on the positive and negative ranking flows; ranking each project to be evaluated based on the net flow; and determining the expected performance of each project to be evaluated among all projects to be evaluated.

[0015] Based on the same inventive concept, this invention also provides a project system capability assessment device based on heterogeneous preferences, comprising: an acquisition module for acquiring numerical evaluation matrices and comparative evaluation matrices of multiple experts for multiple projects to be evaluated, wherein the numerical evaluation matrices include the experts' evaluation values ​​of the projects to be evaluated on strategic objectives, and the comparative evaluation matrices include the experts' comparative priorities of the multiple projects to be evaluated on strategic objectives; an information conversion module for converting the comparative evaluation matrices according to an information conversion operator to obtain a supplementary matrix, and adding the supplementary matrix to the numerical evaluation matrix given by the same expert to obtain a complete evaluation matrix for each expert; an aggregation module for determining expert weights based on a dual-entropy objective, and aggregating the complete evaluation matrices of multiple experts to obtain a comprehensive evaluation matrix; a weight calculation module for determining strategic objective weights based on the comprehensive evaluation matrix and the dual-entropy objective; and an evaluation module for determining the expected performance of the multiple projects to be evaluated on the strategic objectives based on the strategic objective weights and the comprehensive evaluation matrix.

[0016] Based on the same inventive concept, the present invention also provides an electronic device, comprising: a processor; a memory for storing executable instructions; wherein the processor is configured to read the executable instructions from the memory and execute the executable instructions to implement any of the methods described above.

[0017] Based on the same inventive concept, the present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the processor causes the processor to implement any of the methods described above. The beneficial effects of this invention are as follows: This invention focuses on the pre-assessment stage of strategic planning and conducts in-depth research on a method for assessing the execution capability of a project system based on heterogeneous preferences. The method provided in this invention constructs an expert trust network and designs a preference information transformation operator to effectively aggregate heterogeneous preference information from experts in different fields, solving the problem of inconsistent assessment information caused by differences in expert expertise. Simultaneously, based on entropy theory and bi-objective programming, it determines expert weights and strategic goal weights, thereby achieving a scientific assessment of project execution capability and providing solid data support for subsequent performance evaluation and optimization of project portfolios. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a project system capability assessment method based on heterogeneous preferences, provided for an embodiment of the present invention; Figure 2 A flowchart illustrating another project system capability assessment method based on heterogeneous preferences provided in an embodiment of the present invention; Figure 3 An evaluation matrix based on probabilistic language preference information is provided in this embodiment of the invention; Figure 4 A language preference relationship evaluation matrix provided for embodiments of the present invention; Figure 5 A numerical information matrix based on a language preference relation evaluation matrix is ​​provided in this embodiment of the invention; Figure 6 An evaluation information matrix provided for embodiments of the present invention; Figure 7 This is a portion of the complete evaluation matrix provided for embodiments of the present invention; Figure 8 This is another part of the complete evaluation matrix provided in the embodiments of the present invention; Figure 9 A symmetric cross-entropy matrix provided in an embodiment of the present invention; Figure 10 A standardized comprehensive evaluation matrix provided for embodiments of the present invention; Figure 11A schematic diagram of a project system capability assessment device based on heterogeneous preferences provided in an embodiment of the present invention; Figure 12 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0021] To better understand the above-mentioned objectives, features, and advantages of the embodiments of the present invention, the solutions of the embodiments of the present invention will be further described below. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other.

[0022] Many specific details are set forth in the following description in order to provide a full understanding of the embodiments of the present invention, but the embodiments of the present invention may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some embodiments of the embodiments of the present invention, and not all embodiments.

[0023] Strategic planning, as a large-scale strategic undertaking, is characterized by four features: large project scale, wide scope, long execution period, and multiple sources of risk. This presents the first challenge to the formulation and implementation of strategic planning: the different areas of expertise among evaluation experts make it difficult to comprehensively assess the execution capabilities of strategic objectives across different areas within the plan. Simply put, it is difficult to reach a consensus among experts from different fields.

[0024] Language terminology sets, as a representational model approximating human cognitive expression, facilitate approximate and direct description of evaluations. Therefore, this invention proposes constructing an expert trust network, allowing experts to evaluate the expected effects of strategic objectives in different domains using language terminology sets. However, most related expert trust network evaluation methods based on language terminology sets only consider one aspect of probabilistic language preference information or probabilistic language preference relationships. This ignores the fact that experts have different areas of expertise, resulting in heterogeneous preference information when evaluating different attributes of a solution.

[0025] To address the problem of aggregating heterogeneous preference information to assess project execution capabilities, this invention provides a method for evaluating project system capabilities based on heterogeneous preferences, such as... Figure 1 and Figure 2 As shown, it includes: S1. Obtain numerical evaluation matrices and comparative evaluation matrices from multiple experts for multiple projects to be evaluated. The numerical evaluation matrix includes the experts' evaluation values ​​of the projects to be evaluated in terms of strategic objectives, and the comparative evaluation matrix includes the experts' comparative priorities of the multiple projects to be evaluated in terms of strategic objectives.

[0026] Specifically, in strategic planning, when experts from different fields assess the project execution capability, they can provide numerical evaluation information based on probabilistic language preference information (i.e., the numerical evaluation matrix mentioned above) for strategic goals corresponding to fields they are familiar with. However, for strategic goals corresponding to fields they are not familiar with, they can only provide comparative evaluation information based on probabilistic language preference relationships (i.e., the comparative evaluation matrix mentioned above).

[0027] Specifically, the aforementioned numerical evaluation matrix can include the expert's precise evaluation value for each objective of the evaluated project. For example, the expert might evaluate the probability of Project 1's completion on objectives A, B, and C as 98%, 82%, and 75%, respectively. Alternatively, the numerical evaluation matrix can be obtained by establishing a precise scoring system. For instance, the expert evaluation results can be limited to grades A, B, C, and D, with rules such as a score of 100 for grade A, 75 for grade B, 50 for grade C, and 25 for grade D. Even if the initial numerical evaluation information given by the expert is not numerical, it can be quantified into precise numbers. The aforementioned comparative evaluation matrix includes the expert's assessment of the priority of one project over another on a certain objective. Although it does not provide precise evaluation values ​​for each project, the comparative evaluation matrix can also be quantified by defining the numerical relationships. For example, if the expert states that Project 1 is more effective than Project 2 on objective A, this can be recorded as the mathematical relationship A1 > A2. Based on the specific numerical values ​​included in the aforementioned numerical evaluation matrix, these specific values ​​can be further calculated using the mathematical relationships provided by the comparative evaluation matrix. For example, if expert A gives Project 1 a score of 75 on target A, and expert B believes that Project 1 performs better than Project 2 on target A, then Project 1's score on target A can be considered a quantifiable value between 75 and 100. For more complex numerical evaluation matrices and comparative evaluation matrices, the methods provided in this embodiment of the invention can be used for calculation and information integration.

[0028] In a more specific embodiment, the numerical evaluation matrix and comparative evaluation matrix obtained above may be non-intuitive data such as linguistic descriptions, but they can still be converted into quantifiable data using methods such as the probabilistic linguistic terminology set and linguistic scaling function provided in the following embodiments of the present invention. For example, the linguistic evaluation of an event as "very likely," "possible," or "unlikely" corresponds to a quantifiable probability of the event occurring at 90%, 50%, or 30%, respectively.

[0029] S2. Based on the information transformation operator, the comparison evaluation matrix is ​​transformed to obtain the supplementary matrix, and the supplementary matrix is ​​added to the numerical evaluation matrix given by the same expert to obtain the complete evaluation matrix of each expert.

[0030] S3. Determine expert weights based on the dual-entropy objective, and aggregate the complete evaluation matrix of multiple experts based on the expert weights to obtain the comprehensive evaluation matrix.

[0031] S4. Based on the comprehensive evaluation matrix, determine the strategic target weights based on the dual-entropy target.

[0032] S5. Based on the strategic objective weights and the comprehensive evaluation matrix, determine the expected performance of multiple projects to be evaluated in terms of strategic objectives.

[0033] This invention focuses on the pre-assessment stage of strategic planning, and delves into methods for evaluating the execution capabilities of project systems based on heterogeneous preferences. The method provided in this invention constructs an expert trust network and designs a preference information transformation operator to effectively aggregate heterogeneous preference information from experts in different fields, solving the problem of inconsistent evaluation information caused by differences in expert expertise. Simultaneously, based on entropy theory and bi-objective programming, it determines expert weights and strategic goal weights, thereby achieving a scientific assessment of project execution capabilities and providing solid data support for subsequent evaluation and optimization of project portfolio execution capability deviations.

[0034] Specifically, firstly, this invention divides experts into different subgroups according to different "areas of expertise," thus obtaining the same number of expert subgroups as the planned "strategic directions." Each subgroup contains experts in the same strategic direction, who will evaluate the execution capabilities of all projects or project portfolios in these familiar areas, providing relatively accurate numerical evaluation information based on "probabilistic language preference information." Simultaneously, the information they cannot accurately provide also pertains to the same strategic direction, meaning they are unfamiliar with the same areas. For strategic objectives within these strategic directions, experts will provide comparative evaluation information based on "probabilistic language preference relationships." Thus, this invention constructs an expert trust network divided into subgroups based on "strategic directions." Each subgroup in this network exhibits consistency, meaning it consists of experts of the same type, who will provide numerical evaluation matrix information and comparative evaluation matrix information for strategic objectives in the same strategic direction. The next step is to aggregate the heterogeneous preference information from each subgroup to form a complete evaluation matrix for each expert.

[0035] For familiar strategic directions, experts provide numerical evaluation information based on "probabilistic language preference information." However, for unfamiliar strategic directions, experts struggle to provide precise numerical evaluation information based on "probabilistic language preference information." Instead, they can only compare all projects or combinations of projects across these strategic objectives, providing comparative evaluation information based on "probabilistic language preference relationships." Decision-making methods in related technologies only consider one aspect of these two factors, constructing a general numerical evaluation matrix or comparative evaluation matrix and then adjusting it, without aggregating the two. This leads to problems such as the evaluation information being difficult to adjust reasonably or the matrix adjustment results being inconsistent with reality. Therefore, this chapter innovatively proposes a "preference information transformation operator" to transform the comparative evaluation information based on probabilistic language preference relationships, supplementing the missing parts of the numerical evaluation matrix—specifically, the parts corresponding to strategic directions unfamiliar to experts—thereby achieving the aggregation of heterogeneous preference evaluation information. The next step is to consider the entropy of each expert's complete evaluation matrix and determine the weights.

[0036] In some embodiments, the dual-entropy objective includes a minimum total entropy objective and a maximum symmetric cross-entropy objective.

[0037] This invention, based on heterogeneous evaluation information from all experts, considers both the "total entropy" and "symmetric cross entropy" of the evaluation matrix. It sets dual objectives of "maximizing symmetric cross entropy" and "minimizing total entropy" to aggregate heterogeneous information, thereby determining expert weights and strategic objective weights. This ultimately yields an evaluation of the project system's execution capability under all strategic objectives, significantly improving the rationality and internal consistency of the evaluation information and facilitating more accurate evaluation results. Specifically, the lower the uncertainty of the evaluation matrix provided by an expert (i.e., the lower the total entropy), the better the quality of the information reflected by the matrix, and therefore, the greater the weight should be assigned to that expert. Conversely, a larger symmetric cross entropy indicates that the evaluation results of that expert are closer to those of other experts, meaning that the information provided by that expert is closer to the information of the entire group. In this case, such an expert should be given a greater weight.

[0038] In specific implementation, this invention constructs a project system execution capability assessment model based on heterogeneous preferences. The ultimate goal of this model is to evaluate the execution capability of projects or project portfolios in strategic planning, thereby laying the foundation for subsequent project portfolio execution capability deviation correction and optimization. To better facilitate subsequent correction and optimization operations, such as... Figure 2 As shown, this model considers evaluating projects and project portfolios in strategic planning from two aspects: one is the cycle The project portfolio as a whole is aligned with strategic objectives. Execution capability threshold Secondly, throughout the entire planning process, the project... In response to strategic objectives Expected value of execution capability .

[0039] Whether it is solving still The final result of the model is a set of data in two dimensions: either a combination of several projects at a certain period and their impact on strategic objectives. The expected value of execution capability data, or the overall planning of a project's strategic objectives. Execution capability threshold Therefore, since the model solving principle is consistent, the model establishment and example study in this embodiment of the invention will focus on solving "the expected execution capabilities of several projects for each strategic objective throughout the planning process". For example, each strategic direction in the example contains only one strategic objective. Also, for ease of solution, each strategic direction in the example contains only one strategic objective.

[0040] To facilitate subsequent modeling steps, this embodiment demonstrates the basic definitions and functions required for the model.

[0041] For a given strategic plan, suppose the set of projects to be evaluated is... The set of strategic objectives is The strategic objectives have the following weights: The evaluation experts are Expert weight is Furthermore, the weights of strategic objectives and experts are both unknown.

[0042] For a set of linguistic terms, let A set of linguistic terms consisting of an odd number of linguistic variables, where It is a positive integer. The granularity is called S.

[0043] The probabilistic language terminology set is defined as follows:

[0044] in, Representation and Probability Related language terms , for The number of all different language terms.

[0045] Each evaluation expert establishes their own probabilistic linguistic information matrix. As the aforementioned numerical evaluation matrix and probabilistic linguistic relation matrix As the comparison and evaluation matrix mentioned above, where, The total number of experts.

[0046] Specifically, each expert uses a pre-defined set of language terms. Each option is evaluated using a probabilistic linguistic terminology set. Then, experts construct a probabilistic linguistic preference information matrix. Its elements The representative stated that the experts had positive opinions on the candidate projects. About attributes The evaluation value. Then the probability preference information matrix. It can be represented as:

[0047] Each expert uses a pre-defined set of language terms. The above uses a probabilistic language terminology set for pairwise comparisons. Then, experts construct a probabilistic language preference relation matrix. Its elements Experts on the representative side believe that, in terms of a certain attribute, the alternative items Compared to The priority of these factors. Then the probability preference matrix... It can be represented as:

[0048] in, Based on language term set The probabilistic language terminology set. It satisfies the following conditions: (1) , ,and , , . They represent and The number of distinct elements in the text.

[0049] (2) For ,have ;for ,have . and The first Each language term and its corresponding probability.

[0050] A linguistic scale function is a tool for quantifying linguistic terms. It establishes a one-to-one correspondence between a set of linguistic terms and a set of numbers with elements between 0 and 1, thus numerically representing the utility of linguistic terms. For a linguistic scale function, let... It is a set of linguistic terms, expressed through a linguistic scale function. Get and language items Equivalent membership : ;

[0051] Furthermore, the representation and membership degree are obtained through the following function. Language items of equivalent information : ;

[0052] For the distance function, let and These are two sets of probabilistic language terms, where: , and .but and The Euclidean distance between them is defined as:

[0053] For the expectation function, given a set of probabilistic linguistic terms , The expectation function is defined as .

[0054] For the PLWAM aggregation operator, given n probabilistic linguistic term sets , , ..., The aggregate value of the PLWAM operator is also a set of probabilistic language terms, as follows: ;

[0055] and:

[0056] For the hesitation index, let there be a probabilistic linguistic terminology set. .but The hesitation index is defined as: ,in, .

[0057] For the closeness of the probabilistic language terminology set, let For the positive ideal probability language terminology set, This is the terminology set for negative ideal probability language. According to the TOPSIS method, the terminology set for probability language... The closer ,at the same time The further away Then the probabilistic language terminology set The better, therefore, define the probabilistic language terminology set. The degree of closeness is:

[0058] in , and They correspond to , and The set of terms for a normalized ordered probabilistic language. and They are arrive and The Euclidean distance.

[0059] The probabilistic language preference function is used to compare the merits of two candidates on a certain attribute. It is based on the deviation between attribute values ​​and is expressed in the form of a probabilistic language terminology set.

[0060] set up and These are two alternative projects in the solution set. and They are respectively and About attributes Normalized ordered attribute values. Definition about The preference function is: ;

[0061] Wherein, parameter v and Both are greater than or equal to 0, representing the indifference threshold and the strict preference threshold, respectively. According to the above definition, the language scaling function... and Clearly, the constructed preference function is a set of probabilistic language terms.

[0062] In some embodiments, the information conversion operator includes:

[0063] in, For information transformation operators, let represent the th The fuzzy evaluation values ​​of the projects to be evaluated are used to construct a supplementary matrix; The total number of projects to be evaluated. For the first The items to be evaluated and the first The membership degree between the items to be evaluated is calculated based on the evaluation language terminology set and the language scale function. At this point, the information conversion operator... It can also be regarded as the first The total membership degree of the project to be evaluated to all other projects. Specifically, ,at this time To evaluate the language terminology set, the first The items to be evaluated and the first The probabilistic language preference relation language item compared to the first item to be evaluated can be numerically measured. The items to be evaluated and the first The merits and demerits of the projects to be evaluated.

[0064] The significance of the aforementioned information transformation operator function lies in transforming the evaluation based on probabilistic linguistic preferences by assessment experts into an evaluation based on probabilistic linguistic information. Firstly, it utilizes the linguistic scaling function... The process involves converting all language term information in the probabilistic language preference relation matrix into numerical information, and then calculating the total membership degree for each item. Assuming a total of For the candidate projects to be evaluated, considering a specific strategic objective, the linguistic terminology information comparing a project with other projects, after being converted into numerical information, determines its membership range. The membership degree of a given entity compared to itself should be 0.5. Therefore, for a given strategic objective, the first... The fuzzy evaluation numerical information of a project based on probabilistic linguistic information can be transformed using the aforementioned information conversion operator. We obtain and form the supplementary matrix described above.

[0065] In practice, the probability linguistic relation matrix is ​​transformed using the preference information transformation operator H. The information transformation in the middle is supplemented into a probabilistic language information matrix. The missing parts were used to obtain the complete evaluation matrix for each evaluation expert. Furthermore, the matrix also needs to be... Transform into the corresponding normalized ordered matrix .

[0066] In some embodiments, the process of determining expert weights based on a dual-entropy objective described above, such as Figure 2 As shown, it includes: The fuzzy entropy and hesitation entropy of each complete evaluation matrix are calculated, and the total entropy of each complete evaluation matrix is ​​obtained by combining the fuzzy entropy and hesitation entropy; and the symmetric cross entropy of each complete evaluation matrix is ​​calculated.

[0067] Expert weights are determined based on the total entropy and symmetric cross entropy of each complete evaluation matrix, where the total entropy is inversely correlated with the expert weights, and the symmetric cross entropy is positively correlated with the expert weights.

[0068] In practice, the fuzzy entropy function is first determined.

[0069] set up It is a set of linguistic terms. and It is a standardized, ordered set of linguistic terms, in which and .if If it satisfies the following properties, we call it the fuzzy entropy of the probabilistic language terminology set: (1) or ; (2) ,in, It is the expectation function; (3) If and satisfy or ,but ; (4) .

[0070] Based on these properties, the generalized fuzzy entropy of the probabilistic language terminology set is defined below.

[0071] Assumption It is a strictly concave function and satisfies the following conditions: (1) For any , ; (2) ; (3) exist Monotonically increasing, in Monotonically decreasing, then A set of probabilistic language terms The fuzzy entropy.

[0072] make The fuzzy entropy of the probabilistic language terminology set is defined as follows: .

[0073] Next, determine the hesitation entropy function. Let... It is a set of linguistic terms. and It is a standardized, ordered set of linguistic terms, in which and ,if It can be called the hesitation entropy of the probabilistic language terminology set if it satisfies the following properties: (1) ; (2) ; (3) If and but ; (4) If for any All but ,in, and It can be calculated using the fuzzy index mentioned above; (5) .

[0074] Given a normalized, ordered set of probabilistic language terms ,when At that time, the hesitation index was ,when At that time, the hesitation index was ,in, , It is a linguistic term. The subscript.

[0075] Combine fuzzy entropy and hesitation entropy with adjustment coefficient Combining these factors, the total entropy is determined to be... ,because , ,so .

[0076] set up It is a matrix containing a set of probabilistic language terms, defined as follows: The total entropy is .

[0077] Then, the symmetric cross-entropy function is calculated. In this part, cross-entropy is defined to measure the discriminative power between probabilistic language term sets. The axiomatic definition of the cross-entropy measure for probabilistic language term sets is given below.

[0078] set up It is a set of linguistic terms. and It is a standardized, ordered set of linguistic terms, in which and ,but and cross-entropy The following conditions should be met: (1) ; (2) If and only if and .

[0079] Given two normalized ordered probabilistic language terminology sets and ,but It is a cross-entropy.

[0080] As can be seen from the formula, cross-entropy... It is asymmetric. The symmetric cross-entropy between probabilistic language term sets can be obtained as follows: .

[0081] Furthermore, a symmetric cross-entropy between two probabilistic language matrices is defined.

[0082] set up and These are two matrices with a set of probabilistic language terms, defined as follows: and The symmetric cross-entropy between them is .

[0083] In some embodiments, the formula for calculating the expert weights includes:

[0084]

[0085]

[0086] in, For the first Expert weights for a complete evaluation matrix. The first sub-weight is calculated based on the total entropy. The second sub-weight is calculated based on the symmetric cross-entropy. This is a compromise factor. To fully evaluate the total number of matrices, For the first The total entropy of a complete evaluation matrix For the first The complete evaluation matrix and the first Symmetric cross-entropy between complete evaluation matrices.

[0087] Specifically, based on the aforementioned dual-entropy objective, if the uncertainty of the evaluation matrix provided by the evaluation expert is smaller (i.e., the total entropy is smaller), then the information quality reflected by the matrix is ​​better, and therefore the expert's weight should be allocated larger. Based on this criterion, by minimizing the total entropy of the evaluation matrix, this embodiment of the invention establishes the following plan for solving the evaluation expert weights:

[0088] in, yes The total entropy.

[0089] Solving the above equation using the Lagrange multiplier method yields the following result:

[0090] On the other hand, if an evaluation expert's assessment results are closer to those of other evaluation experts, it means that the information provided by that expert is closer to the information of the entire group. In this case, such an evaluation expert should be given greater weight. Based on this consideration, this embodiment of the invention constructs another planning model for determining the weight of evaluation experts based on symmetric cross-entropy:

[0091] Solving the above equation using the Lagrange multiplier method yields the following result:

[0092] Based on the results of the two models above, the final weight of the evaluation experts is determined. for:

[0093] in This is a compromise factor.

[0094] In some embodiments, the process of aggregating the complete evaluation matrices of multiple experts to obtain a comprehensive evaluation matrix includes: The comprehensive evaluation matrix is ​​obtained by aggregating the complete evaluation matrices of each expert using the PLWAM operator. This represents the comprehensive evaluation matrix, where, .

[0095] Then, the comprehensive evaluation matrix Standardization process is performed to obtain ,in .

[0096] Referring to the above embodiments, symmetric cross-entropy and total entropy can respectively reflect the differences between candidate items and the reliability of the evaluation value. Based on this, the embodiments of the present invention determine the strategic objective weights by "maximizing symmetric cross-entropy" and "minimizing total entropy". Therefore, when the strategic objective weight information is completely unknown, the above process of determining the strategic objective weights based on the comprehensive evaluation matrix and dual-entropy objectives includes: Construct a set of objective formulas for the strategic objective weights that minimize the total entropy of the comprehensive evaluation matrix and maximize the symmetric cross entropy of the comprehensive evaluation matrix: ;

[0097] Solving the above set of objective formulas using the Lagrange multiplier method yields the first... Weight of each strategic objective for: ;

[0098] In some embodiments, the step of determining the expected performance of multiple projects to be evaluated in terms of strategic objectives based on strategic objective weights and a comprehensive evaluation matrix includes: A comprehensive preference index is determined for each project to be evaluated based on the strategic objective weights and a comprehensive evaluation matrix. Specifically, this index is used to comprehensively evaluate the merits of candidate projects across all attributes. It is based on the weight of each attribute and the corresponding preference function value. This is done according to a probabilistic preference function. Define the comprehensive preference index as ,in yes The degree of closeness. Furthermore, a comprehensive preference matrix of the candidate projects in terms of strategic objectives can be obtained.

[0099] The positive ranking stream for each item to be evaluated is determined based on the comprehensive preference index. With negative sorting flow Specifically, , .

[0100] Determine the net flow of each item to be evaluated based on the positive and negative order flows. Specifically,

[0101] Each item to be evaluated is ranked according to its net flow, and the expected performance of each item to be evaluated is determined among all items to be evaluated.

[0102] The present invention also provides a specific embodiment to demonstrate the specific implementation effect of the above method.

[0103] To address a specific strategic plan, a panel of experts was organized to analyze and evaluate several potential projects. There are currently six strategic evaluation experts. The candidate projects were analyzed and evaluated. To facilitate the study and analysis of examples, six evaluation experts selected four strategic objectives from four strategic directions (military technology, joint operations, defense industry, and military-civilian integration). (Self-reliance on key technologies) Integrated command across the entire domain Equipment digitization and sharing of military and civilian resources (This is a strategic objective to be evaluated as an alternative objective.)

[0104] Of the six strategic assessment experts, five had opinions on strategic objectives. Not very familiar with the strategic objectives, Expert 1 and Expert 3... , Not very familiar, Expert 4 on strategic objectives Not very familiar with the strategic objectives, Expert 2 and Expert 6. Not very familiar with it.

[0105] Based on the assessment, there are four candidate projects. The proposal was submitted and awaits evaluation by six strategic assessment experts. Following the experts' evaluation of the strategic objectives of the four projects within their respective "areas of expertise," an evaluation matrix based on probabilistic linguistic preference information (i.e., the numerical evaluation matrix mentioned above) was obtained, as follows: Figure 3 As shown. It is understandable that... Figures 3 to 10 For illustrative purposes only, the functions and values ​​in the matrix are simulated data and have no practical meaning.

[0106] After comparing the strategic objectives of the four projects in the "unfamiliar areas" of the experts, a language preference relationship evaluation matrix based on the six experts was obtained (i.e., the comparative evaluation matrix mentioned above). For example, the language preference relationship evaluation matrix of expert 2 is as follows: Figure 4 As shown. Then, through the language scale function The linguistic terminology information in the probabilistic linguistic relation matrix is ​​converted into numerical information. For example, the numerical information converted from the linguistic preference relation evaluation matrix of Expert 2 is as follows: Figure 5 As shown.

[0107] Furthermore, using the aforementioned preference information transformation operator, the fuzzy evaluation values ​​H of the four items in the strategic objectives of the experts' "unfamiliar domain" are calculated. This transforms the evaluation information of the six experts based on the language preference relationship matrix into evaluation information based on the language preference information matrix, i.e., the supplementary matrix. For example, the evaluation information matrix of expert 2 based on the language preference information matrix is ​​as follows: Figure 6 As shown.

[0108] Therefore, this embodiment transforms the information in the probabilistic linguistic relation matrix using a preference information transformation operator. Next, the missing parts of the probabilistic linguistic information moments are supplemented to obtain the complete evaluation matrix for each evaluation expert, as shown below. Figure 7 and Figure 8 As shown. Further, the heterogeneous matrix group is transformed into a corresponding normalized ordered matrix. Then, the original heterogeneous preference information and preference relation matrices can be aggregated into a matrix group.

[0109] Next, the total entropy and symmetric cross entropy of the complete evaluation matrix for each strategic evaluation expert are calculated. After calculation, the expert's complete evaluation matrix is ​​obtained. The total entropy are respectively , , , , , The calculated symmetric cross-entropy is as follows: Figure 9 As shown. Assume a compromise factor. Solving for the weight vectors of the six strategic assessment experts yields... .

[0110] Using the PLWAM operator, individual complete evaluation moments are aggregated into a collective comprehensive evaluation matrix. And transform it into a normalized ordered matrix. ,like Figure 10 As shown.

[0111] Calculate the comprehensive evaluation matrix for each item to be evaluated. The total entropy and the symmetric cross entropy are taken as... Afterwards, projects to be evaluated Total entropy Projects to be evaluated Total entropy Projects to be evaluated Total entropy Projects to be evaluated Total entropy Projects to be evaluated Symmetric cross-entropy Projects to be evaluated Symmetric cross-entropy Projects to be evaluated Symmetric cross-entropy Projects to be evaluated Symmetric cross-entropy According to the bi-objective programming method, the weight vector of the strategic objectives is determined as follows: .

[0112] Based on the strategic objective weights, the preference function value for each project to be evaluated is calculated, and the positive and negative flows of the candidate projects are calculated. Finally, the net flow of each project to be evaluated is obtained. , , , .because Therefore, alternative projects It is the best project.

[0113] After data processing, we can obtain the expected values ​​of the four projects' execution capabilities for the four strategic objectives. And the execution capability thresholds of the four projects for the four strategic objectives. Specifically, regarding the execution capability threshold... Since the focus is primarily on the overall project portfolio, and strategic objectives can change over time, the project portfolio execution capability threshold... The performance of a project portfolio is measured from two dimensions: strategic objectives and time horizon. The expected value for execution capability score... Considering that each project targets different strategic objectives, the "expected value" should remain stable across different periods. Therefore, the expected value of the project's execution capability score is... It measures a project's "expected performance," or theoretical performance, from the two dimensions of strategic goals and project.

[0114] It should be noted that the method of this embodiment can be executed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method of this embodiment, and the multiple devices will interact with each other to complete the above method.

[0115] It should be noted that the above description describes some embodiments of the present invention. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than that shown in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0116] Based on the same inventive concept, corresponding to any of the methods in the above embodiments, the present invention also provides a project system capability assessment device based on heterogeneous preferences, such as... Figure 11 As shown, it includes: The acquisition module 10 is used to acquire numerical evaluation matrices and comparative evaluation matrices of multiple experts for multiple projects to be evaluated. The numerical evaluation matrix includes the experts' evaluation values ​​of the projects to be evaluated on strategic objectives, and the comparative evaluation matrix includes the experts' comparative priorities of the multiple projects to be evaluated on strategic objectives.

[0117] The information conversion module 20 is used to convert the comparison evaluation matrix according to the information conversion operator to obtain a supplementary matrix, and add the supplementary matrix to the numerical evaluation matrix given by the same expert to obtain the complete evaluation matrix of each expert.

[0118] The aggregation module 30 is used to determine expert weights based on the dual-entropy objective, aggregate the complete evaluation matrices of multiple experts, and obtain a comprehensive evaluation matrix.

[0119] The weight calculation module 40 is used to determine the strategic target weights based on the dual-entropy target according to the comprehensive evaluation matrix.

[0120] Evaluation module 50 is used to determine the expected performance of multiple projects to be evaluated in terms of strategic objectives based on the strategic objective weights and the comprehensive evaluation matrix.

[0121] This invention focuses on the strategic planning pre-assessment stage, and delves into methods for evaluating the execution capability of project systems based on heterogeneous preferences. The apparatus provided in this invention constructs an expert trust network and designs a preference information transformation operator to effectively aggregate heterogeneous preference information from experts in different fields, solving the problem of inconsistent evaluation information caused by differences in expert expertise. Simultaneously, based on entropy theory and bi-objective programming, it determines expert weights and strategic goal weights, thereby achieving a scientific evaluation of project execution capability and providing solid data support for subsequent evaluation and optimization of project portfolio execution capability deviations.

[0122] For ease of description, the above apparatus is described in terms of its functions, divided into various modules. Of course, in implementing this invention, the functions of each module can be implemented in one or more software and / or hardware components.

[0123] The apparatus described above is used to implement the corresponding project system capability assessment method based on heterogeneous preferences in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0124] Figure 12 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention is shown.

[0125] like Figure 12 As shown, the electronic device may include a processor 1101 and a memory 1102 storing computer program instructions.

[0126] Specifically, the processor 1101 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of the present invention.

[0127] Memory 1102 may include a mass storage device for information or instructions. For example, and not limitingly, memory 1102 may include a hard disk drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 1102 may include removable or non-removable (or fixed) media. Where appropriate, memory 1102 may be internal or external to the integrated gateway device. In a particular embodiment, memory 1102 is a non-volatile solid-state memory. In a particular embodiment, memory 1102 includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable programmable PROM (EEPROM), an electrically alterable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0128] The processor 1101 reads and executes computer program instructions stored in the memory 1102 to perform the steps of the project system capability assessment method based on heterogeneous preferences provided in the embodiments of the present invention.

[0129] In one example, the electronic device may also include a transceiver 1103 and a bus 1104. Wherein, as... Figure 12 As shown, the processor 1101, memory 1102 and transceiver 1103 are connected via bus 1104 and communicate with each other.

[0130] Bus 1104 may include hardware, software, or both. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 1104 may include one or more buses. Although specific buses are described and illustrated in the embodiments of the present invention, the present invention is contemplated by any suitable bus or interconnect.

[0131] The following are embodiments of a computer-readable storage medium provided in this invention. This computer-readable storage medium and the project system capability assessment method based on heterogeneous preferences in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the computer-readable storage medium, please refer to the embodiments of the project system capability assessment method based on heterogeneous preferences described above.

[0132] This embodiment provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a project system capability assessment method based on heterogeneous preferences.

[0133] Of course, the computer-executable instructions provided in the embodiments of the present invention are not limited to the above-described method operations, but can also perform related operations in the project system capability assessment method based on heterogeneous preferences provided in any embodiment of the present invention.

[0134] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer cloud platform (which can be a personal computer, server, or network cloud platform, etc.) to execute the project system capability assessment method based on heterogeneous preferences provided in the various embodiments of the present invention.

[0135] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the aforementioned element.

[0136] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments described above, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for assessing the capabilities of a project system based on heterogeneous preferences, characterized in that: include: Obtain numerical evaluation matrices and comparative evaluation matrices from multiple experts for multiple projects to be evaluated. The numerical evaluation matrices include the experts' evaluation values ​​of the projects to be evaluated in terms of strategic objectives, and the comparative evaluation matrices include the experts' comparative priorities of the multiple projects to be evaluated in terms of strategic objectives. The comparison evaluation matrix is ​​transformed according to the information transformation operator to obtain a supplementary matrix, and the supplementary matrix is ​​added to the numerical evaluation matrix given by the same expert to obtain the complete evaluation matrix of each expert. Based on the dual-entropy objective, expert weights are determined, and the complete evaluation matrices of multiple experts are aggregated to obtain a comprehensive evaluation matrix. Based on the comprehensive evaluation matrix, the strategic objective weights are determined according to the dual-entropy objective. Based on the strategic objective weights and the comprehensive evaluation matrix, the expected performance of multiple projects to be evaluated in relation to the strategic objectives is determined.

2. The method as described in claim 1, characterized in that, The dual-entropy objective includes the minimum total entropy objective and the maximum symmetric cross-entropy objective.

3. The method as described in claim 1, characterized in that, The information conversion operator includes: in, Let be the information transformation operator, representing the first... The fuzzy evaluation values ​​of the items to be evaluated are used to construct the supplementary matrix; The total number of items to be evaluated. For the first The items to be evaluated and the first The membership degree between the items to be evaluated is calculated based on the evaluation language term set and the language scale function.

4. The method as described in claim 2, characterized in that, The method for determining expert weights based on a dual-entropy objective includes: The fuzzy entropy and hesitation entropy of each complete evaluation matrix are calculated, and the total entropy of each complete evaluation matrix is ​​obtained by combining the fuzzy entropy and hesitation entropy; and the symmetric cross entropy of each complete evaluation matrix is ​​calculated. The expert weights are determined based on the total entropy and symmetric cross entropy of each complete evaluation matrix, wherein the total entropy is inversely correlated with the expert weights, and the symmetric cross entropy is positively correlated with the expert weights.

5. The method as described in claim 4, characterized in that, The formula for calculating the expert weight includes: in, For the first The expert weights of the complete evaluation matrix. The first sub-weight is calculated based on the total entropy. The second sub-weight is calculated based on the symmetric cross-entropy. This is a compromise factor. The total number of the complete evaluation matrices, For the first The total entropy of the complete evaluation matrix. For the first The complete evaluation matrix mentioned above and the first Symmetric cross-entropy between the complete evaluation matrices.

6. The method as described in claim 2, characterized in that, The step of determining the strategic target weights based on the comprehensive evaluation matrix and the dual-entropy target includes: Construct a set of objective formulas for strategic objective weights that minimize the total entropy of the comprehensive evaluation matrix and maximize the symmetric cross entropy of the comprehensive evaluation matrix; The strategic objective weights are obtained by solving the set of objective formulas using the Lagrange multiplier method.

7. The method as described in claim 1, characterized in that, The step of determining the expected performance of multiple projects to be evaluated on the strategic objectives based on the strategic objective weights and the comprehensive evaluation matrix includes: A comprehensive preference index is determined for each of the items to be evaluated based on the strategic objective weights and the comprehensive evaluation matrix. The positive and negative ranking flows for each item to be evaluated are determined based on the comprehensive preference index. The net flow of each item to be evaluated is determined based on the positive sorting flow and the negative sorting flow; Based on the net flow, each of the items to be evaluated is sorted, and the expected performance of each item to be evaluated among all the items to be evaluated is determined.

8. A project system capability assessment device based on heterogeneous preferences, characterized in that, include: The acquisition module is used to acquire numerical evaluation matrices and comparative evaluation matrices of multiple experts for multiple projects to be evaluated. The numerical evaluation matrix includes the experts' evaluation values ​​of the projects to be evaluated on strategic objectives, and the comparative evaluation matrix includes the experts' comparative priorities of the multiple projects to be evaluated on strategic objectives. The information conversion module is used to convert the comparison evaluation matrix according to the information conversion operator to obtain a supplementary matrix, and add the supplementary matrix to the numerical evaluation matrix given by the same expert to obtain the complete evaluation matrix of each expert. The aggregation module is used to determine expert weights based on the dual-entropy objective, aggregate the complete evaluation matrices of multiple experts, and obtain a comprehensive evaluation matrix. The weight calculation module is used to determine the strategic target weights based on the dual-entropy target according to the comprehensive evaluation matrix. The evaluation module is used to determine the expected performance of multiple projects to be evaluated on the strategic objectives based on the strategic objective weights and the comprehensive evaluation matrix.

9. An electronic device, characterized in that, include: processor; A memory for storing executable instructions; wherein the processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The storage medium stores a computer program that, when executed by a processor, causes the processor to implement the method described in any one of claims 1 to 7.

Images