Performance analysis method and system based on knowledge dynamic propagation integrated measurement model
By constructing a dynamic communication process model and a multi-stage DEA model between projects, considering knowledge feedback and cascading dissemination, the problem of inaccurate performance measurement caused by ignoring these factors in the existing technology is solved, and the accuracy and management of project portfolio performance is improved. Effect.
Patent Information
- Application Number
- CN202410470439.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-18
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-04-18
AI Technical Summary
When analyzing the dynamic dissemination of knowledge, the existing technology ignores the impact of knowledge feedback and cascading dissemination on project portfolio performance, resulting in inaccurate measurement of project portfolio performance.
The dynamic communication process model based on knowledge dynamic communication is used to construct a dynamic communication process model between projects, consider knowledge feedback and cascading communication, and analyze the impact of dynamic knowledge communication on project portfolio performance through a multi-stage DEA model.
It improves the accuracy of project portfolio performance measurement, alleviates the problem of measurement inaccuracy caused by ignoring the influence of dynamic propagation, and enhances the understanding and management of project portfolio performance.
Smart Images

Figure CN118505173B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of multi-project management performance technology, and in particular to a performance analysis method and system based on a knowledge dynamic propagation integrated measurement model. Background Art
[0002] The project portfolio contains a large number of projects and the relationships between projects are intricate. The success or failure of a project will have a series of impacts on other projects. Therefore, in the case of numerous projects, high dependencies and limited resources, how to control the project portfolio to maximize the overall success probability of the project portfolio is an important issue facing decision makers. Knowledge is the experience and lessons accumulated by the R&D team in the previous product development process. Knowledge dissemination means that the knowledge acquired by the R&D team in one project can be spread to other projects, which can then drive the success of a series of other projects. Through the dissemination of knowledge between projects, enterprises can continuously improve their R&D capabilities to better develop future products. In the project portfolio, the dynamic dissemination of knowledge means that after a project in the project portfolio succeeds, its accumulated knowledge spreads to other projects in the project portfolio that have dependencies with it, driving the success of a series of other projects, and thus achieving the sustainable development of the entire project portfolio. It can be seen that through the dissemination of knowledge between projects, enterprises can continuously improve their R&D capabilities and enhance their future success in similar projects, which is of great significance to improving the overall performance of the project portfolio.
[0003] The essence of knowledge dynamic dissemination is the process in which the "knowledge source" spreads the knowledge it possesses to the "knowledge receiver" so that the "knowledge receiver" can master such knowledge. It can be seen that the dynamic dissemination of knowledge is the interaction between the "knowledge source" and the "knowledge receiver". Agent-based modeling and simulation can well model the interaction between nodes in the network, and through the modeling of the dynamic dissemination process of knowledge between projects in the project portfolio network, the micro behavior of dynamic dissemination of knowledge is organically combined with the "emergence" behavior of improving the performance of the project portfolio brought about by dynamic dissemination of knowledge.
[0004] However, the traditional use of "agent-based modeling" to analyze the dynamic dissemination of knowledge in multiple projects ignores the influence of the following factors: 1) The impact of knowledge feedback on the dynamic dissemination of knowledge. Knowledge feedback refers to the evaluation given by the "knowledge recipient" on the quality of the knowledge and the effect of using the knowledge in its own project after receiving the knowledge passed to it by the "knowledge source". 2) The impact of cascade propagation. Cascade propagation means that when a project in a project portfolio obtains knowledge, it will pass the knowledge to other projects with which it has a dependent relationship, so that the project also obtains knowledge, which in turn causes a chain reaction of knowledge in the project portfolio. 3) The impact of dynamic knowledge dissemination on project portfolio performance. Whether using an infectious disease model or an agent-based modeling, the main focus is on the dynamic dissemination process of knowledge between projects and the knowledge level of each project in a stable state, while ignoring the impact of this dynamic dissemination on project portfolio performance, which in turn affects the accurate measurement of project portfolio performance. Summary of the invention
[0005] In order to solve the above technical problems existing in the prior art, the embodiment of the present invention provides a performance analysis method and system based on a knowledge dynamic propagation integrated measurement model. The technical solution is as follows:
[0006] On the one hand, a performance analysis method based on a knowledge dynamic propagation integrated measurement model is provided, including:
[0007] Taking each project in the target project portfolio as an entity and the knowledge diffusion and feedback relationship between the projects as a connection, a dynamic knowledge propagation process model between the projects in the target project portfolio is constructed; based on the dynamic propagation process model, the probability of knowledge propagation and the knowledge level between the projects are determined; based on the dynamic propagation process model, the probability of knowledge propagation and the knowledge level, a multi-stage DEA model is constructed; based on the multi-stage DEA model, a performance analysis of the target project portfolio is performed.
[0008] Optionally, the dynamic propagation process model includes:
[0009] When the first entity is a knowledge source, the first entity will pass the knowledge it possesses to a second entity with a dependency relationship in the network where the target project portfolio is located; the first entity and the second entity are different projects in the target project portfolio; when the second entity receives the knowledge diffusion from the first entity, the second entity judges the knowledge level of the first entity. If the first entity has the highest knowledge level among all projects with a dependency relationship with the second entity, the second entity accepts the knowledge diffusion of the first entity; otherwise, the knowledge of the first entity will not be transmitted to the second entity; when the first entity is a knowledge recipient, the first entity selects the project with the highest knowledge level among the dependent projects.
[0010] Optionally, determining the probability of knowledge diffusion between projects based on the dynamic diffusion process model includes:
[0011] Taking feedback into account, the direct probability of knowledge diffusion between projects includes:
[0012]
[0013] Considering feedback, the probability of secondary knowledge diffusion between projects includes:
[0014]
[0015] Where N is the number of projects in the project portfolio; KP i,m (t) represents the project p at time t m The knowledge can be diffused to project p i The probability of KP m,j (t) represents the project p at time t j The knowledge can be diffused to project p m probability;
[0016] Considering feedback, the probability of three knowledge diffusions between projects includes:
[0017]
[0018] Considering feedback, the probability of multiple knowledge diffusions between projects includes:
[0019]
[0020] Among them, KP i,j (t) is the direct probability of knowledge diffusion between projects when knowledge feedback is considered at time t, x represents the xth knowledge feedback between projects, KF i,j (t) is the learning index, and z = 1, 2, and 3 represent the first, second, and third transmission of knowledge between projects, respectively.
[0021] Optionally, based on the dynamic propagation process model, determining the knowledge level between projects includes:
[0022]
[0023] KF i,j (t) = 1-(1-KQ i,j (t))×(1-KE i,j (t))
[0024] Among them, KL i (t) is the knowledge level at time t, KLS i(t) is the amount of knowledge obtained by the project as a knowledge source at time t, KLR i (t) is the amount of knowledge acquired by the project as a knowledge recipient at time t, KF i,j (t) is the item p i For project p j Evaluation score of the knowledge transferred, SKP i,j (t) is the probability of multiple knowledge diffusions between projects;
[0025] KF i,j (t) is the learning index, which is the calculated comprehensive score of the knowledge receiver’s feedback to the knowledge source at time t; KQ i,j (t-1) is the project knowledge evaluation at time t-1;
[0026] KE i,j (t-1) is the evaluation of knowledge usage effect at time t-1.
[0027] Optionally, the multi-stage DEA model includes:
[0028]
[0029] In the formula,
[0030]
[0031] Among them, PP is the objective function, ST i (t) represents item p i The learning time invested by the team in stage t, N represents the number of projects in the project portfolio, and M represents the number of projects pointing to project p i The number of items, O represents the number of items p i The number of items pointed to; Y i (t) represents item p i project performance.
[0032] Optionally, based on the multi-stage DEA model, a performance analysis is performed on the target project portfolio, including:
[0033] The knowledge level of each project in the target project portfolio is used as input and input into the multi-stage DEA model, and the project performance of the target project portfolio is obtained based on the multi-stage DEA model.
[0034] On the other hand, a performance analysis system based on a knowledge dynamic propagation integrated measurement model is provided, comprising: a first construction module and a determination module, a second construction module and an analysis module; wherein,
[0035] The first construction module is used to construct a dynamic knowledge diffusion process model among the projects in the target project portfolio, taking each project in the target project portfolio as an entity and the knowledge diffusion and feedback relationship among the projects as a connection; the determination module is used to determine the knowledge diffusion probability and knowledge level among the projects based on the dynamic diffusion process model; the second construction module is used to construct a multi-stage DEA model based on the dynamic diffusion process model, the knowledge diffusion probability and the knowledge level; the analysis module is used to perform performance analysis on the target project portfolio based on the multi-stage DEA model.
[0036] Optionally, determining the probability of knowledge diffusion between projects based on the dynamic diffusion process model includes:
[0037] Taking feedback into account, the direct probability of knowledge diffusion between projects includes:
[0038]
[0039] Considering feedback, the probability of secondary knowledge diffusion between projects includes:
[0040]
[0041] Where N is the number of projects in the project portfolio; KP i,m (t) represents the project p at time t m The knowledge can be diffused to project p i The probability of KP m,j (t) represents the project p at time t j The knowledge can be diffused to project p m probability;
[0042] Considering feedback, the probability of three knowledge diffusions between projects includes:
[0043]
[0044] Considering feedback, the probability of multiple knowledge diffusions between projects includes:
[0045]
[0046] Among them, KP i,j (t) is the direct probability of knowledge diffusion between projects when knowledge feedback is considered at time t, x represents the xth knowledge feedback between projects, KF i,j (t) is the learning index, and z = 1, 2, and 3 represent the first, second, and third transmission of knowledge between projects, respectively.
[0047] Optionally, based on the dynamic propagation process model, determining the knowledge level between projects includes:
[0048]
[0049] KF i,j (t) = 1-(1-KQ i,j (t))×(1-KE i,j (t))
[0050] Among them, KL i (t) is the knowledge level at time t, KLS i (t) is the amount of knowledge obtained by the project as a knowledge source at time t, KLR i (t) is the amount of knowledge acquired by the project as a knowledge recipient at time t, KF i,j (t) is the item p i For project p j Evaluation score of the knowledge transferred, SKP i,j (t) is the probability of multiple knowledge diffusion between projects; KF i,j (t) is the learning index, which is the calculated comprehensive score of the knowledge receiver’s feedback to the knowledge source at time t; KQ i,j (t-1) is the project knowledge evaluation at time t-1; KE i,j (t-1) is the evaluation of knowledge usage effect at time t-1.
[0051] Optionally, the multi-stage DEA model includes:
[0052]
[0053] In the formula,
[0054]
[0055] Among them, PP is the objective function, ST i (t) represents item p i The learning time invested by the team in stage t, N represents the number of projects in the project portfolio, and M represents the number of projects pointing to project p i The number of items, O represents the number of items p i The number of items pointed to; Y i (t) represents item p i project performance.
[0056] The present invention provides a performance analysis method and system based on an integrated measurement model of knowledge dynamic propagation. When analyzing the knowledge dynamic propagation process by using agent-based modeling, the influence of knowledge feedback and cascade propagation on each project in a project portfolio is also considered. Further, based on the analysis of the knowledge dynamic propagation process, the present invention adopts a multi-stage DEA model to analyze the influence of knowledge dynamic propagation on project portfolio performance. This alleviates the technical problem existing in the prior art of ignoring the influence of dynamic propagation on project portfolio performance, thereby affecting the accurate measurement of project portfolio performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0058] Figure 1 It is a flow chart of a performance analysis method based on a knowledge dynamic propagation integrated measurement model provided by an embodiment of the present invention;
[0059] Figure 2 is a schematic diagram of a knowledge dissemination mode provided by an embodiment of the present invention;
[0060] Figure 3 is a schematic diagram of a knowledge dissemination mode considering feedback provided by an embodiment of the present invention;
[0061] Figure 4 It is a schematic diagram of multiple diffusion and feedback between projects provided by an embodiment of the present invention;
[0062] Figure 5 It is a schematic diagram of a dynamic propagation process of knowledge among multiple projects provided by an embodiment of the present invention;
[0063] Figure 6 It is a schematic diagram of an inter-project diffusion network model provided by an embodiment of the present invention;
[0064] Figure 7 is a schematic diagram of a knowledge quality evaluation BSM provided by an embodiment of the present invention;
[0065] Figure 8 is a schematic diagram of measuring a project knowledge level provided by an embodiment of the present invention;
[0066] Fig. 9 It is a schematic diagram of the basic principle of a multi-stage DEA model provided by an embodiment of the present invention;
[0067] Fig.10It is a schematic diagram of a project development stage provided by an embodiment of the invention;
[0068] Fig.11 It is a schematic diagram of the basic principle of a dynamic network DEA model provided by an embodiment of the present invention;
[0069] Fig.12 This is a schematic diagram of project input and output in stages provided by an embodiment of the present invention;
[0070] Fig.13 It is a schematic diagram of a performance analysis system based on a knowledge dynamic propagation integrated measurement model provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0071] The technical solution of the present invention is described below in conjunction with the accompanying drawings.
[0072] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.
[0073] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.
[0074] Embodiment 1
[0075] Figure 1 FIG. 1 is a flow chart of a performance analysis method based on a knowledge dynamic propagation integrated measurement model provided according to an embodiment of the present invention. Figure 1 As shown, the method specifically comprises the following steps:
[0076] Step S102 , taking each project in the target project portfolio as an entity and the knowledge diffusion and feedback relationship between the projects as a connection, a dynamic knowledge diffusion process model between the projects in the target project portfolio is constructed.
[0077] Step S104, based on the dynamic diffusion process model, determine the probability of knowledge diffusion and the knowledge level between projects.
[0078] Step S106, constructing a multi-stage DEA model based on the dynamic propagation process model, knowledge propagation probability and knowledge level.
[0079] Step S108: Perform performance analysis on the target project portfolio based on the multi-stage DEA model.
[0080] Specifically, the dynamic propagation process model includes:
[0081] When the first entity is a knowledge source, the first entity will transfer the knowledge it possesses to a second entity with a dependency relationship in the network where the target project portfolio is located; the first entity and the second entity are different projects in the target project portfolio;
[0082] When the second entity receives knowledge diffusion from the first entity, the second entity judges the knowledge level of the first entity. If the first entity has the highest knowledge level among all projects that have a dependency relationship with the second entity, the second entity accepts the knowledge diffusion of the first entity; otherwise, the knowledge of the first entity will not be spread to the second entity.
[0083] When the first entity is a knowledge recipient, the first entity selects the item with the highest knowledge level among the items with dependency relationships.
[0084] Figure 2 FIG. 1 is a schematic diagram of a knowledge dissemination mode provided according to an embodiment of the present invention. Figure 2 As shown in Figure (a), the process of knowledge dissemination involves two objects: knowledge source and knowledge receiver. The knowledge source is the project that has knowledge, which is the experience accumulated by the project team in practice; the knowledge receiver is the project that receives the knowledge. The process of knowledge dissemination can be expressed as the process of "knowledge source A" "diffusing" the knowledge it has to "knowledge receiver B".
[0085] Further, if Figure 2 As shown in Figure (b), after receiving the knowledge, "knowledge receiver B" will feed back the effect of knowledge use to "knowledge source A", thereby improving the knowledge level of "knowledge source A" and the probability that "knowledge source A" will spread its knowledge to other projects in the network. In short, feedback will affect the dynamic dissemination process of knowledge. Therefore, the present invention analyzes the dynamic dissemination process of knowledge based on "feedback".
[0086] Figure 3 FIG. 1 is a schematic diagram of a knowledge dissemination model considering feedback provided according to an embodiment of the present invention. Figure 3 As shown in the figure, the content of "knowledge feedback" includes two aspects: 1) the evaluation of the quality of knowledge delivered by the knowledge source, which is the evaluation of the quality of the knowledge itself; 2) the evaluation of the use effect of the received knowledge, which is the evaluation of the implementation effect of the knowledge in the receiving project. Both evaluations are given by the R&D team of the receiving knowledge project, and the values are 0-1. The higher the value, the higher the evaluation score. Figure 3 As shown in the middle right figure, when "knowledge source A" receives the feedback on knowledge from "knowledge recipient B" at time t, at time t+1, when "knowledge source A" diffuses knowledge, it will diffuse the updated knowledge to other projects in the project portfolio (such as knowledge recipients B and C).
[0087] In summary, the following issues should be noted when considering the analysis of the knowledge dissemination process with feedback: 1) At time t, when "knowledge source A" transfers its knowledge to "knowledge receiver B", it will receive feedback from "knowledge receiver B" about the knowledge; 2) The feedback process will prompt "knowledge source A" to continuously update its knowledge, so that the knowledge level KL of "knowledge source A" at time t+1 is A (t+1) is greater than the knowledge level KL at time t A (t); 3) If Figure 2 (a) and Figure 2 As shown in (b), "Knowledge Source A" will diffuse its knowledge to multiple projects in the network, so it will receive feedback from multiple projects. Each "knowledge receiver"'s feedback to "Knowledge Source A" will prompt it to update its knowledge. Then, when receiving feedback from multiple "knowledge receivers" to "Knowledge Source A", how to measure the knowledge level of "Knowledge Source A" is an important issue that needs to be solved.
[0088] In addition, the diffusion effect in the R&D project portfolio may not only occur between projects with direct connections, but may also occur multiple times due to the existence of "intermediaries", that is, the spread of diffusion. Figure 4 FIG. 1 is a schematic diagram of multiple diffusion and feedback between projects according to an embodiment of the present invention. Figure 4 As shown in the figure, diffusion propagation means that the risk of project p1 will spread to projects p2 and p5 that have dependencies on it, and the risk will further spread to more projects in the network through projects p2 and p5. The diffusion of risk from project p1 to projects p2 and p5 along the paths p1→p2 and p1→p5 in the project portfolio network is called primary diffusion; the diffusion of risk from project p1 to projects p3 and p6 along the paths p1→p2→p3 and p1→p5→p6 is called secondary diffusion; the diffusion of risk along the paths p1→p2→p3 and p1→p5→p6 is called secondary diffusion.
[0089] p1→p2→p3→p4 and p1→p5→p6→p7, the diffusion from project p1 to projects p4 and p7 is called tertiary diffusion; the risk along the path
[0090] p1→p2→p3→p4→…→p m and
[0091] p1→p5→p6→p7→...→p n , spreading from project p1 to project p m and p n The diffusion of is called m-th diffusion.
[0092] Furthermore, similar to the diffusion of risk in a project portfolio network, the knowledge feedback from “knowledge receivers” will also undergo multiple cascade propagations, e.g. Figure 4 As shown, p m Along the path p m →p m-1 The knowledge feedback is called “m-time feedback”.
[0093] In the "Agent-based Modeling of Dynamic Knowledge Diffusion Process", the agents are the projects in the project portfolio network. By describing the dynamic process of knowledge diffusion among the projects, the impact of dynamic knowledge diffusion on the project portfolio network is analyzed. Figure 5 is a schematic diagram of a dynamic propagation process of knowledge among multiple projects provided according to an embodiment of the present invention, such as Figure 5 As shown, in the project portfolio network constructed by the present invention, each project plays two roles: knowledge source and knowledge receiver, that is, while the project transmits its own knowledge to other projects in the network, it also accepts knowledge diffusion from other projects.
[0094] Specifically, Figure 5 As shown, when the project p i When it is a knowledge source, it will pass on the knowledge it has to other projects in the network that have dependencies on it (such as project p k ), when the project p k Received from project p i When the knowledge of i The knowledge level of the project is judged. i If project p has the highest knowledge level among all the projects with which it is dependent, then project p k Will accept project p i knowledge diffusion, when knowledge from project p i Propagate to project p k The probability of SKP k,i (t); otherwise item p i The knowledge will not be propagated to project p k When the project p i When it is a knowledge receiver, it will choose the project with the highest knowledge level among the dependent projects (such as project p j ), namely KL j (t)=max(KL1(t),KL2(t),…,KL n (t)), at this time, knowledge is derived from project p j Propagate to p i The probability of SKP i,j (t).
[0095] In short, the essence of dynamic knowledge dissemination is the process in which the "knowledge source" project transfers its knowledge (knowledge level) to the knowledge recipient with a certain probability (the probability of knowledge dissemination between projects). It can be seen that the dynamic dissemination of knowledge between multiple projects involves two key factors: 1) the probability of knowledge dissemination between projects; 2) the knowledge level of the project.
[0096] Specifically, based on the dynamic propagation process model, the probability of knowledge propagation between projects is determined, including:
[0097] Taking feedback into account, the direct probability of knowledge diffusion between projects includes:
[0098]
[0099] Considering feedback, the probability of secondary knowledge diffusion between projects includes:
[0100]
[0101] Where N is the number of projects in the project portfolio; KP i,m (t) represents the project p at time t m The knowledge can be diffused to project p i The probability of KP m,j (t) represents the project p at time t j The knowledge can be diffused to project p m The probability of
[0102] Considering feedback, the probability of three knowledge diffusions between projects includes:
[0103]
[0104] Considering feedback, the probability of multiple knowledge diffusions between projects includes:
[0105]
[0106] Among them, KP i,j (t) is the direct probability of knowledge diffusion between projects when knowledge feedback is considered at time t, x represents the xth knowledge feedback between projects, KF i,j (t) is the learning index, and z = 1, 2, and 3 represent the first, second, and third transmission of knowledge between projects, respectively.
[0107] In the embodiment of the present invention, knowledge feedback occurs between related projects in the project portfolio network. Therefore, when measuring the probability of knowledge diffusion taking feedback into account, a knowledge diffusion network between projects must be constructed first. Figure 6 is a schematic diagram of an inter-project diffusion network model provided according to an embodiment of the present invention. Figure 6 As shown in the figure, the knowledge diffusion network is a directed weighted graph constructed with projects as nodes and dependencies between projects as edges. The direction of the edge reflects the direction of the dependencies between projects. For example, the directed edge from project p1 to p2 indicates that project p1 depends on p2. The weight of the edge reflects the strength of the dependencies between projects. For example, the weight of the directed edge from project p1 to p2 is 0.3, which means that the strength of project p1's dependency on p2 is 0.3. Further, the inter-project knowledge diffusion network shown in the figure is described by the knowledge diffusion DSM matrix. Figure 7 FIG. 1 is a schematic diagram of a knowledge quality evaluation BSM provided according to an embodiment of the present invention. Figure 7 As shown in (a), in the knowledge diffusion DSM, the "columns" of the matrix represent the diffusion effect of the project on other projects (i.e., the output knowledge diffusion), the "rows" represent the diffusion effect of the project on other projects (i.e., the input knowledge diffusion), and the numbers on the off-diagonal lines represent the probability of technology diffusion between projects. For example, Figure 7 In Figure (a), DSM(2,1) = 0.3, indicating that the probability of direct knowledge diffusion from p1 to p2 is 0.3.
[0108] From the above analysis, we can see that the evaluation of the "knowledge recipient" to the "knowledge source" has two aspects: 1) the evaluation of the quality of knowledge delivered by the knowledge source. Based on the evaluation results, a knowledge quality evaluation DSM can be constructed. This matrix reflects the quality evaluation of the knowledge received by the project team, such as Figure 7 In (b), DSM(2,1) = 0.6, which means that the p1 project team believes that the quality of knowledge diffused from p1 to p2 is 0.6; 2) Receive the evaluation of the use effect of knowledge. Similarly, the knowledge use effect evaluation DSM can be constructed based on the evaluation results of the use effect, such as Figure 7 In Figure (c), DSM(2,1) = 0.2, indicating that the effect of knowledge diffusion from p1 to p2 in project p1 is 0.2. Figure 7 In (b) and (c), the comprehensive evaluation score fed back by the “knowledge recipient” to the “knowledge source” at time t can be measured as:
[0109] KF i,j (t) = 1-(1-KQ i,j (t))×(1-KE i,j (t)) (1)
[0110] Where KF i,j (t) is the calculated comprehensive score fed back by the “knowledge receiver” to the “knowledge source” at time t; KQ i,j (t-1) is the project knowledge evaluation DSM at time t-1, such as Figure 7 Shown in (b); KE i,j (t-1) is the knowledge use effect evaluation DSM at time t-1, such as Figure 7 As shown in (c).
[0111] Furthermore, when the "knowledge source" receives the evaluation of the "knowledge recipient" on the knowledge feedback (the calculation result of formula (1)), it will stimulate the learning effect of the "knowledge source" project team and affect the knowledge diffusion probability and knowledge level of the "knowledge source". According to the classic learning curve, the impact of knowledge feedback on the probability of knowledge diffusion of the "knowledge source" is:
[0112]
[0113] In the formula, KP i,j (t) is the direct probability of knowledge diffusion between projects when knowledge feedback is not considered at time t, and x represents the xth knowledge feedback between projects. The reason why the base of the exponential function in Formula 2 is x+1 is that when the xth knowledge feedback is given, it is the x+1th time that the R&D team of the project has carried out such work. KF i,j (t) is the calculation result of formula (1), which represents the learning index in the learning curve.
[0114] In summary, the direct probability of knowledge diffusion between projects under the condition of considering feedback is shown in formula (2). Figure 4 The integrated probability of knowledge dissemination between projects under the conditions of multiple diffusion and multiple feedback shown in the figure. The specific process is:
[0115] ① Calculate the quadratic probability of knowledge dissemination between projects at time t, considering knowledge feedback
[0116] The quadratic probability refers to the item p j The knowledge will be passed through the intermediate project p m Spread to project p i ,like Figure 6 As shown in Figure 1, in addition to direct knowledge diffusion from project p1 to p2, knowledge diffusion will also occur through the intermediate project p4 according to the path p1→p4→p2. The calculation of the secondary probability of knowledge diffusion between projects is based on the direct knowledge diffusion probability between projects taking into account "feedback" calculated by formula (2), that is:
[0117]
[0118] In the formula, KP m,j (t) represents p at time t j The knowledge diffuses directly to p m The probability of (calculated from formula (2)), Indicates that from project p j Diffusion to p through any intermediate node i The probability that all possible propagation paths do not occur at the same time.
[0119] ② Under three-stage diffusion, the diffusion probability between projects
[0120] like Figure 6 As shown, there are two paths for the three-way diffusion from p1 to p2, namely, diffusion without loop (p1→p4→p3→p2) and diffusion with loop (p1→p2→p4→p2). Therefore, the project p j Through any two intermediate items p m and p n Spread to project p i The probability of the third knowledge diffusion is:
[0121]
[0122] ③ Under the conditions of multiple diffusion and multiple feedback, the integrated probability of knowledge dissemination between projects When multiple diffusion and multiple feedback are used at the same time, the integrated diffusion probability (SKP) between projects is j Diffusion to p through various possible paths i The probability (see Figure 7 (c)). When considering multiple diffusion propagation, from p j to p i Integrated diffusion probability SKP i,j (t) is:
[0123]
[0124] In the formula, z=1, 2, 3 represent the primary, secondary, and tertiary transmission of knowledge between projects, respectively.
[0125] Specifically, based on the dynamic diffusion process model, the knowledge level between projects is determined, including:
[0126]
[0127] Among them, KL i (t) is the knowledge level at time t, KLS i (t) is the amount of knowledge obtained by the project as a knowledge source at time t, KLR i (t) is the amount of knowledge acquired by the project as a knowledge recipient at time t, KF i,j (t) is the item p i For project p j Evaluation score of the knowledge transferred, SKP i,j (t) is the probability of multiple knowledge diffusion between projects; KF i,j (t) is the learning index, which is the calculated comprehensive score of the knowledge receiver’s feedback to the knowledge source at time t; KQ i,j (t-1) is the project knowledge evaluation at time t-1; KE i,j(t-1) is the evaluation of knowledge usage effect at time t-1.
[0128] KF i,j (t) = 1-(1-KQ i,j (t))×(1-KE i,j (t))
[0129] Where KF i,j (t) is the calculated comprehensive score fed back by the “knowledge receiver” to the “knowledge source” at time t; KQ i,j (t-1) is the project knowledge evaluation DSM at time t-1, such as Figure 7 Shown in (b); KE i,j (t-1) is the knowledge use effect evaluation DSM at time t-1, such as Figure 7 As shown in (c).
[0130] In the model of dynamic knowledge dissemination, the knowledge level of the project (Agent) is constantly improving. The knowledge level (KL) of the project refers to the amount of knowledge owned by the R&D project team, which depends on the experience and lessons accumulated by the R&D team in the previous product development process, the ability to use the knowledge to solve problems, the education level of the team, and other factors. In the traditional project knowledge level model, the project p i The knowledge level KL of the team at time t i (t) depends on its knowledge level KL at time t-1 i (t-1) and the probability of knowledge diffusion of other projects that have dependencies with this project team, namely:
[0131]
[0132] Where P i,j (t) is the project p at time t j The knowledge can be diffused to project p i probability.
[0133] Formula (6) is about the project p at time t i The calculation of the knowledge level of the project p at time t-1 is integrated i The knowledge level and project p at time t i The amount of knowledge that can be obtained from other projects in the network. The measurement method of the project's knowledge level has been generally recognized by scholars. However, there are the following problems: 1) It does not distinguish between knowledge sources and knowledge recipients. The two have different ways of acquiring knowledge in the process of dynamic knowledge dissemination. Figure 8 is a schematic diagram of measuring a project knowledge level according to an embodiment of the present invention, such as Figure 8As shown in Figure 2, for the knowledge source project, the knowledge level at time t depends on the knowledge level of the project at time t-1 and the amount of knowledge gained from feedback to stimulate learning ability; for the knowledge recipient project, the knowledge level at time t depends on the knowledge level of the project at time t-1 and the amount of knowledge gained from the knowledge source at time t. i If a team acquires a certain type of knowledge, it will not acquire it from all the projects that have dependencies on it, that is, at time t, project p i The knowledge increment is not equal to the accumulation of knowledge gained by the project from other projects with which it is dependent in the network (such as the value of formula (6) ). Similar to student course selection, if a student needs to acquire knowledge about the course "Introduction to Management", and there are 5 teachers in the school offering this course, the student will not choose the courses of all 5 teachers at the same time to acquire the knowledge. He will only choose the one he thinks is the best among these 5 teachers. The same is true for projects. Therefore, the project p i Before acquiring certain knowledge, you first need to evaluate the knowledge levels of all projects that are dependent on it, and select the project with the highest knowledge level to acquire the knowledge.
[0134] In summary, the present invention proposes to measure the knowledge level of project pi at time t by considering the knowledge source and the knowledge recipient, as well as the knowledge level of dependent projects. The specific process is as follows:
[0135] ① Such as Figure 8 As shown, the present invention measures the knowledge level of the project at time t from the two dimensions of knowledge source and knowledge recipient. For the knowledge source project, the knowledge level at time t depends on the knowledge level of the project at time t-1 and the amount of knowledge gained by feedback to stimulate learning ability. Therefore, the knowledge level of the knowledge source project at time t is:
[0136]
[0137] In the formula, KL i (t-1) is the project p at time t-1 i Knowledge level, KL i (t) is the project p at time t i level of knowledge, The definition of is consistent with that of formula (2).
[0138] ②For the knowledge recipient project, the project p at time t i The knowledge level of the project depends on the knowledge level of the project at time t-1 and the amount of knowledge obtained from the knowledge source at time t, that is: KLR i (t) = KL i (t-1)+SKP i,j (t)×(KLj (t)-KL i (t-1))
[0139]
[0140] In the formula, SKP i,j (t) represents the integrated probability of knowledge diffusion between projects, which is calculated by formula (5); KL j (t)-KL i (t-1) represents the item p at time t and at time t-1 j and p i The difference in knowledge level; constraints (a)-(c) make project p i Ability to select the project with the highest level of knowledge from all other projects associated with it in the project portfolio network j Then, at this time, project p i Can be from project p j The amount of knowledge gained is: SKP i,j (t)×(KL j (t)-KL i (t-1)), which is the product of the difference in knowledge level between the "knowledge source" and the "knowledge recipient" project and the probability of knowledge diffusion from the "knowledge source" to the "knowledge recipient". Constraint (d) indicates that similar knowledge can only be diffused from projects with high knowledge levels to projects with low knowledge levels.
[0141] It is worth noting that the purpose of constraints (a)-(c) in equation (8) is to make the project p i The knowledge can be obtained by selecting the project with the highest knowledge level from all other projects in the project portfolio network that have dependencies on it. Figure 6 As shown in the knowledge diffusion network model shown in the figure, the project portfolio network is related to project p i There is more than one associated project, that is, project p i This knowledge can be obtained from multiple projects in the network. Therefore, if project p j The knowledge can be diffused to project p i , then it is necessary to satisfy the constraints (a)-(c) in equation (8). Constraint (c) can also be expressed as follows, finding the maximum product of the probability of knowledge diffusion and the amount of knowledge for each row:
[0142] SKP i,j (t)*KL j (t) = max{SKP i,1 (t)*KL1(t),SKP i,2 (t)*KL2(t),…,SKP i,n (t)*KL n(t)} (9)
[0144] ③ By Figure 6 It can be seen that each project in the project portfolio network is often both a knowledge source and a knowledge receiver. Therefore, the knowledge level of project pi at time t is:
[0145] KL i (t) = KLS i (t)+KLR i (t) (10)
[0146] In summary, in the agent-based modeling of the present invention, Agenti (i.e., project p) in the project portfolio i )’s knowledge level is:
[0147]
[0148] Specifically, the multi-stage DEA model includes:
[0149]
[0150] In the formula,
[0151]
[0152] Among them, PP is the objective function, ST i (t) represents item p i The learning time invested by the team in stage t, N represents the number of projects in the project portfolio, and M represents the number of projects pointing to project p i The number of items, O represents the number of items p i The number of items pointed to; Y i (t) represents item p i project performance.
[0153] Specifically, step S108 includes: taking the knowledge level of each project in the target project portfolio as input, inputting it into the multi-stage DEA model, and obtaining the project performance of the target project portfolio based on the multi-stage DEA model.
[0154] The multi-stage DEA model of the project portfolio constructed by the present invention not only evaluates the performance of the project at each stage, but also evaluates the overall performance of the R&D project, and at the same time, measures the overall performance of the project portfolio. Fig. 9 Schematic diagram of the basic principle of a multi-stage DEA model provided according to an embodiment of the present invention. Fig. 9 As shown, for a project portfolio, the input is the knowledge level KL of each project. i(t), calculated by formula (11); the output is project performance. How to measure the efficiency between the input "knowledge level" and the output "project performance" is a problem that needs to be solved by the DEA (Data envelopment analysis, DEA) model in this invention.
[0155] Fig.10 1 is a schematic diagram of a project development stage provided according to an embodiment of the present invention. Fig.10 As shown in the figure, since the project life cycle includes a series of stages, such as feasibility study, customer demand analysis, concept development, detailed design, prototyping, testing, trial operation, etc., the present invention measures the performance of project portfolios at different R&D stages under the condition of dynamic knowledge dissemination based on a multi-stage DEA (data development analysis, DEA) model.
[0156] In an embodiment of the present invention, DEA may be used to analyze the input-output efficiency of a series of decision making units (DMUs). Fig.11 FIG. 1 is a schematic diagram of the basic principle of a dynamic network DEA model provided according to an embodiment of the present invention. Fig.11 As shown, in the project portfolio performance measurement model constructed by the present invention, the decision-making units (DMUs) are the projects in the project portfolio, the input of the decision-making unit is the knowledge level of the project, and the output is the project performance. Among them, the knowledge level of the project at time t can be calculated by formula (11); the project performance at time t can be calculated by formula (14). Furthermore, the project life cycle includes a series of stages, such as feasibility study, customer demand analysis, concept development, detailed design, prototyping, testing, trial operation, etc. Therefore, the present invention takes this series of R&D stages as milestone nodes, and believes that the knowledge accumulated by the R&D project at a certain stage can be spread to other projects, which helps to shorten the time required for these projects in the corresponding R&D stages.
[0157] Project p i Take the example to illustrate the dynamic dissemination process of knowledge at each stage. Fig.12 1 is a schematic diagram of project input and output in stages according to an embodiment of the present invention. Fig.12 As shown, t = 1 stage project p i The input is project p i Initial knowledge level of the team KL i (t=0) and item p i The probability of being able to acquire knowledge from other projects in the network, i.e. (SKP i,j (t) calculated by formula (5)); the input of stage t = 2 to t = 7 is the knowledge level of the previous stage (calculated by formula (11)) and the project pi The probability of being able to acquire knowledge from other projects in the network, i.e. Project i The output is project performance Y i (t) (calculated by formula (12)).
[0158] For R&D projects, progress, cost and technical indicator measurement results are the three most critical elements in the project R&D process. Therefore, the present invention measures project performance by the gap between the expected "progress, cost and technical indicator measurement results" of the R&D project at time t and the actual results, that is:
[0159]
[0160] In the formula, t represents the stage of the R&D project, S r (t) represents the R&D project p in stage t i Expected progress, S a (t) represents the R&D project p in stage t i The actual progress of r (t) represents the R&D project p in stage t i Expected cost, C a (t) represents the R&D project p in stage t i The actual cost of r (t) represents the R&D project p in stage t i The expected technical indicators of the completion of T a (t) represents the R&D project p in stage t i The actual completion status of technical indicators.
[0161] In summary, according to Fig.12 The project input-output diagram shown in the figure, the multi-stage DEA model for calculating the project portfolio efficiency constructed by the present invention is:
[0162]
[0163] Where, the objective function is to maximize the target completion rate and knowledge level of the project portfolio in the overall R&D stage, μ i (t) and ω i (t) can be calculated by equations (14) and (15) respectively; further, Y i (t) can be calculated by formula (12), which reflects the project p i The difference between the expected progress, cost, and technical indicators at stage t and the actual situation. The smaller the value, the higher the target completion rate of the project. Therefore, the objective function adopts in the form of KL i (t) According to formula (11)
[0164] It can be calculated that represents the knowledge level of each project in the project portfolio. Constraint (1) γ i (t-1)×KL i (t-1)+ω i (t)×SKP i (t) represents the value of any project p in the project portfolio i For example, the sum of the information flowing into the project at stage t is 1. Fig.12 As shown, at stage t, the inflow into project p i The amount of information includes the t stage project p i The probability of being able to obtain knowledge from other projects in the network SKP i (t) and t-1 phase project p i Knowledge Level KL i (t-1);
[0165] Constraints (2)
[0166] (μ i (t)Y i (t)+γ i (t)×KL i (t))-(ω i (t)×SKP i (t)+γ i (t-1)×KL i (t-1))≤0 means that at stage t, the information flow outflowing from each project in the project portfolio is less than or equal to the information flow inflowing. i Outgoing information flows include items p i The target completion rate Y i (t) and t-stage project p i Knowledge Level KL i (t).
[0167]
[0168] Where ST i (t) represents item p i The learning time invested by the team in stage t.
[0169]
[0170] In the formula, SKP i,j (t) can be calculated by formula (5), which means that the project p j The knowledge of p diffuses to p through various possible paths i The integration probability of .
[0171]
[0172] Where KLi (t) can be calculated by formula (11) to obtain the item p at time t i level of knowledge.
[0173] Dynamic knowledge dissemination is conducive to improving the overall performance level of the project portfolio. How to analyze the dynamic dissemination process of knowledge between projects and measure the impact of dynamic knowledge dissemination on the project portfolio network are important issues that project portfolio managers are concerned about. The present invention constructs a model of the dynamic dissemination process of knowledge based on Agent modeling, and analyzes the impact of dynamic knowledge dissemination on project portfolio performance based on a multi-stage DEA model. In summary, the present invention uses Agent-based modeling and a multi-stage DEA model to study the dynamic dissemination of knowledge and its impact on the performance of the project portfolio network. It is specifically divided into three steps: 1) Treat each project in the project portfolio as an agent, take the project as a node, and take the knowledge diffusion and feedback relationship between projects as a connection to construct a dynamic dissemination process model of knowledge between projects in the project portfolio, and analyze the dynamic dissemination process of knowledge between projects; 2) Construct a knowledge dissemination probability and knowledge level model between projects considering knowledge feedback and cascade propagation, and analyze the impact of knowledge feedback and cascade propagation on the project portfolio network; 3) Construct a project portfolio performance measurement model under the condition of dynamic knowledge dissemination based on a multi-stage DEA model, and analyze the impact of dynamic knowledge dissemination on the performance of the project portfolio network.
[0174] Embodiment 2
[0175] Fig.13 FIG. 1 is a schematic diagram of a performance analysis system based on a knowledge dynamic propagation integrated measurement model provided according to an embodiment of the present invention. Fig.13 The system comprises: a first construction module 10 and a determination module 20 , a second construction module 30 and an analysis module 40 .
[0176] Specifically, the first construction module 10 is used to construct a dynamic knowledge diffusion process model among the projects in the target project portfolio, taking the projects in the target project portfolio as entities and the knowledge diffusion and feedback relationship among the projects as connections.
[0177] The determination module 20 is used to determine the probability of knowledge dissemination and the knowledge level between projects based on the dynamic dissemination process model.
[0178] The second construction module 30 is used to construct a multi-stage DEA model based on the dynamic propagation process model, the knowledge propagation probability and the knowledge level.
[0179] The analysis module 40 is used to perform performance analysis on the target project portfolio based on the multi-stage DEA model.
[0180] Specifically, the dynamic propagation process model includes:
[0181] When the first entity is a knowledge source, the first entity will transfer the knowledge it possesses to a second entity with a dependency relationship in the network where the target project portfolio is located; the first entity and the second entity are different projects in the target project portfolio;
[0182] When the second entity receives knowledge diffusion from the first entity, the second entity judges the knowledge level of the first entity. If the first entity has the highest knowledge level among all projects that have a dependency relationship with the second entity, the second entity accepts the knowledge diffusion of the first entity; otherwise, the knowledge of the first entity will not be spread to the second entity.
[0183] When the first entity is a knowledge recipient, the first entity selects the item with the highest knowledge level among the items with dependency relationships.
[0184] Based on the dynamic propagation process model, determining the probability of knowledge propagation between projects includes:
[0185] Taking feedback into account, the direct probability of knowledge diffusion between projects includes:
[0186]
[0187] Considering feedback, the probability of secondary knowledge diffusion between projects includes:
[0188]
[0189] Where N is the number of projects in the project portfolio; KP i,m (t) represents the project p at time t m The knowledge can be diffused to project p i The probability of KP m,j (t) represents the project p at time t j The knowledge can be diffused to project p m probability;
[0190] Considering feedback, the probability of three knowledge diffusions between projects includes:
[0191]
[0192] Considering feedback, the probability of multiple knowledge diffusions between projects includes:
[0193]
[0194] Among them, KP i,j (t) is the direct probability of knowledge diffusion between projects when knowledge feedback is considered at time t, x represents the xth knowledge feedback between projects, KF i,j(t) is the learning index, and z = 1, 2, and 3 represent the first, second, and third transmission of knowledge between projects, respectively.
[0195] Based on the dynamic diffusion process model, determine the knowledge level between projects, including:
[0196]
[0197] KFi, j (t) = 1-(1-KQ i,j (t))×(1-KE i,j (t))
[0198] Among them, KL i (t) is the knowledge level at time t, KLS i (t) is the amount of knowledge obtained by the project as a knowledge source at time t, KLR i (t) is the amount of knowledge acquired by the project as a knowledge recipient at time t, KF i,j (t) is the item p i For project p j Evaluation score of the knowledge transferred, SKP i,j (t) is the probability of multiple knowledge diffusion between projects; KF i,j (t) is the learning index, which is the calculated comprehensive score of the knowledge receiver’s feedback to the knowledge source at time t; KQ i,j (t-1) is the project knowledge evaluation at time t-1; KE i,j (t-1) is the evaluation of knowledge usage effect at time t-1.
[0199] Multi-stage DEA model, including:
[0200]
[0201] In the formula,
[0202]
[0203] Among them, PP is the objective function, ST i (t) represents item p i The learning time invested by the team in stage t, N represents the number of projects in the project portfolio, and M represents the number of projects pointing to project p i The number of items, O represents the number of items p i The number of items pointed to; Y i (t) represents item p i project performance.
[0204] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0205] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0206] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0207] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0208] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0209] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0210] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0211] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0212] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A performance analysis method based on a knowledge dynamic propagation integrated measurement model, characterized in that: include: Taking each project in the target project portfolio as an entity and the knowledge diffusion and feedback relationship between the projects as a connection, a dynamic knowledge diffusion process model between the projects in the target project portfolio is constructed; Determining the probability of knowledge diffusion and the knowledge level between projects based on the dynamic diffusion process model; Based on the dynamic propagation process model, the knowledge propagation probability and the knowledge level, construct a multi-stage DEA model; Based on the multi-stage DEA model, performing performance analysis on the target project portfolio; Wherein, based on the dynamic propagation process model, determining the probability of knowledge propagation between projects includes: Taking feedback into account, the direct probability of knowledge diffusion between projects includes: Considering feedback, the probability of secondary knowledge diffusion between projects includes: Considering feedback, the probability of three knowledge diffusions between projects includes: Considering feedback, the probability of multiple knowledge diffusions between projects includes: Among them, KP i,j (t) is the direct probability of knowledge diffusion between projects when knowledge feedback is considered at time t, x represents the xth knowledge feedback between projects, KF i,j (t) is the learning index, KP m,j (t) represents p at time t j The knowledge diffuses directly to p m The probability of z = 1, 2, and 3 respectively represents the first, second, and third transmission of knowledge between projects; Wherein, based on the dynamic propagation process model, determining the knowledge level between projects includes: KF i,j (t)=1-(1-KQ i,j (t))×(1-KE i,j (t)) Among them, KL i (t) is the knowledge level at time t, KLS i (t) is the amount of knowledge obtained by the project as a knowledge source at time t, KLR i (t) is the amount of knowledge acquired by the project as a knowledge recipient at time t, KF i,j (t) is the item p i For project p j Evaluation score of the knowledge transferred, SKP i,j (t) is the probability of multiple knowledge diffusion between projects; KF i,j (t) is the calculated comprehensive score fed back by the "knowledge receiver" to the "knowledge source" at time t; KQ i,j (t-1) is the project knowledge evaluation at time t-1; KE i,j (t-1) is the evaluation of knowledge utilization effect at time t-1; Wherein, based on the multi-stage DEA model, the performance analysis of the target project portfolio is performed, including: The knowledge level of each project in the target project portfolio is used as input and input into the multi-stage DEA model, and the project performance of the target project portfolio is obtained based on the multi-stage DEA model.
2. The method according to claim 1, characterized in that The dynamic propagation process model comprises: When the first entity is a knowledge source, the first entity will pass the knowledge it possesses to a second entity with a dependency relationship in the network where the target project portfolio is located; the first entity and the second entity are different projects in the target project portfolio; When the second entity receives knowledge diffusion from the first entity, the second entity determines the knowledge level of the first entity. If the first entity has the highest knowledge level among all projects that have a dependency relationship with the second entity, the second entity accepts the knowledge diffusion of the first entity; otherwise, the knowledge of the first entity will not be spread to the second entity. When the first entity is a knowledge recipient, the first entity selects the project with the highest knowledge level among the dependent projects.
3. The method according to claim 1, characterized in that The multi-stage DEA model includes: In the formula, Among them, PP is the objective function, ST i (t) represents item p i The learning time invested by the team in stage t.
4. A performance analysis system based on a knowledge dynamic propagation integrated measurement model, characterized in that: include: A first building module, a determination module, a second building module and an analysis module; wherein, The first construction module is used to construct a dynamic propagation process model of knowledge among the projects in the target project portfolio, taking the projects in the target project portfolio as entities and the knowledge diffusion and feedback relationship among the projects as connections; The determination module is used to determine the probability of knowledge dissemination and the knowledge level between projects based on the dynamic dissemination process model; The second construction module is used to construct a multi-stage DEA model based on the dynamic propagation process model, the knowledge propagation probability and the knowledge level; The analysis module is used to perform performance analysis on the target project portfolio based on the multi-stage DEA model; The knowledge propagation probability includes: Taking feedback into account, the direct probability of knowledge diffusion between projects includes: Considering feedback, the probability of secondary knowledge diffusion between projects includes: Considering feedback, the probability of three knowledge diffusions between projects includes: Considering feedback, the probability of multiple knowledge diffusions between projects includes: Among them, KP i,j (t) is the direct probability of knowledge diffusion between projects when knowledge feedback is considered at time t, x represents the xth knowledge feedback between projects, KF i,j (t) is the learning index, KP m,j (t) represents p at time t j The knowledge diffuses directly to p m The probability of z = 1, 2, and 3 respectively represents the first, second, and third transmission of knowledge between projects; The knowledge level includes: KF i,j (t)=1-(1-KQ i,j (t))×(1-KE i,j (t)) Among them, KL i (t) is the knowledge level at time t, KLS i (t) is the amount of knowledge obtained by the project as a knowledge source at time t, KLR i (t) is the amount of knowledge acquired by the project as a knowledge recipient at time t, KF i,j (t) is the item p i For project p j Evaluation score of the knowledge transferred, SKP i,j (t) is the probability of multiple knowledge diffusion between projects; KF i,j (t) is the calculated comprehensive score fed back by the "knowledge receiver" to the "knowledge source" at time t; KQ i,j (t-1) is the project knowledge evaluation at time t-1; KE i,j (t-1) is the evaluation of knowledge utilization effect at time t-1; The analysis module is specifically used to take the knowledge level of each project in the target project portfolio as input into the multi-stage DEA model, and obtain the project performance of the target project portfolio based on the multi-stage DEA model.
5. The system according to claim 4, characterized in that The multi-stage DEA model includes: In the formula, Among them, PP is the objective function, ST i (t) represents item p i The learning time invested by the team in stage t.
