A project health monitoring analysis method and device based on behavior analysis

Abstract

The application discloses a project health degree monitoring analysis method and equipment based on behavior analysis, belongs to the technical field of project management, and is used for solving the technical problems that the overall and comprehensive management cannot be performed throughout the life cycle of an electronic product project, the health degree of the electronic product project is difficult to monitor, and the project management efficiency is low in the stage management of the electronic product project. The method comprises the following steps: calculating the task completion progress of each stage of the project to obtain the current task completion degree of each stage of the project; data matching is performed between the current task completion degree and a preset ideal task completion degree to obtain a data mapping relationship; cross mapping matching is performed on the current task completion degree according to the optimized data mapping relationship to determine the health degree of each stage of the project; the health degree of each stage of the project is input into a trained neural network model to obtain evaluation indexes of each stage of the project; an index system is constructed through the evaluation indexes; and the comprehensive correlation degree grade and the index system are visually displayed.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of project management, and in particular to a project health degree monitoring and analysis method and device based on behavior analysis. BACKGROUND

[0002] The project health degree is generally monitored and judged from seven aspects, including the relevant person in charge, project scope controllability, progress controllability, project target, risk awareness, team organization, and convertible benefit. The monitoring and pre-evaluation of the seven aspects should be throughout the project, from the start to the end of the project. The project manager can track the health degree effectively to manage and control the progress of the project in real time and realize the management of the project process.

[0003] With the increasing demand for electronic information products, there is a higher requirement for the health degree monitoring of the overall life cycle of the electronic product project.

[0004] The current traditional project management method is to manage a certain link or a certain part of the project engineering by using a project management mode based on the whole life cycle, without penetrating the entire cycle of the electronic product project. For the project quality management and the behavior management of the related personnel in the project engineering, it is difficult to form a perfect management mechanism, which easily causes the total quality of the project engineering to be poor and the health degree monitoring of the electronic product project to be not comprehensive. At the same time, in the electronic product project, the data display of the entire project life cycle, such as demand analysis, design, development, testing, acceptance and feedback, is not comprehensive, and it is difficult to effectively evaluate each stage. For new managers, it is difficult to get started with the project, and it is difficult to effectively manage the process of the electronic product project, which reduces the management efficiency of the electronic product project. SUMMARY

[0005] The embodiments of the present application provide a project health degree monitoring and analysis method and device based on behavior analysis, which are used to solve the following technical problems: in the current stage management of the electronic product project engineering, it is difficult to manage the entire life cycle comprehensively, and the health degree monitoring of the electronic product project is difficult, which reduces the project management efficiency.

[0006] The embodiments of the present application adopt the following technical solutions:

[0007] In one aspect, the embodiment of the application provides a project health degree monitoring analysis method based on behavior analysis, comprising: calculating real-time task completion progress of each stage of a project according to an initial project health parameter value of each stage of the project obtained in advance, to obtain a current task completion degree of each stage of the project; performing data matching related to a rough set on the current task completion degree and a preset ideal task completion degree, to obtain a data mapping relationship between the current task completion degree and the ideal task completion degree; performing correction processing on the data mapping relationship, to obtain an optimized data mapping relationship, and respectively performing cross mapping matching on current task completion degrees of a plurality of subtasks in the current task completion degree according to the optimized data mapping relationship, to determine health degrees of each stage of the project; wherein the current task completion degree comprises current task completion degrees of a plurality of subtasks; inputting the health degrees of each stage of the project into a trained neural network model, to obtain evaluation indexes of each stage of the project; constructing an index system based on a project full life cycle through the evaluation indexes of each stage of the project; wherein the project full life cycle is a total stage cycle of each stage of the project; performing correlation degree level judgment on the project full life cycle according to the index system, to obtain a comprehensive correlation degree level of the project full life cycle.

[0008] The embodiment of the application realizes monitoring of the project full life cycle by monitoring and analyzing health degrees of each stage of the project, in combination with an index system of the project full life cycle and a comprehensive correlation degree level, and helps a project manager to better manage a project process, simplifies a complicated management process, accurately warns the health degrees of each stage of the project, helps the manager to timely regulate relevant stages, intelligently reminds the manager of problems to be handled, further improves the efficiency of project management, perfects management of the project engineering, and reduces hidden risks possibly existing in the project management process.

[0009] In an implementation, the calculation of the real-time task completion progress of each stage of the project according to the initial project health parameter values in each stage of the project obtained in advance obtains the current task completion degree of each stage of the project, and specifically includes: obtaining the stage information of each stage in the whole life cycle of the project; wherein the stages at least include: a project requirement stage, a project design stage, a project development stage, a project test stage, a project acceptance stage and a project feedback stage; extracting the initial project health parameter values in the stages; wherein the initial project health parameters at least include: a project progress value, a project cost value and a project quality value; performing cost discount value processing on the project cost according to a NPV present value model to obtain an initial cost value; performing ratio conversion on the time value and the task amount value in the project progress to obtain an initial progress value; constructing a database for the project quality value, the initial cost value and the initial progress value to obtain an initial database; obtaining the real-time task amount of each stage; wherein the real-time task amount at least includes: a current task progress value, a current task cost value and a current task quality value; inputting the current task cost value into the initial database to determine the current task completion degree of each stage of the project.

[0010] The embodiments of the present application are beneficial to timely grasping the real-time task progress of each stage in the project and providing accurate judgment conditions for the health degree of each stage of the project by determining the current task completion degree of each stage of the project.

[0011] In an implementation, the current task completion degree is matched with a preset ideal task completion degree in relation to a rough set to obtain a data mapping relationship between the current task completion degree and the ideal task completion degree, and specifically includes: obtaining a current task completion set corresponding to the current task completion degree; the current task completion set includes a current task completion sub-set, and the current task completion sub-set at least includes a current task progress set, a current task cost set and a current task quality set; the current task completion set is extended by a preset distance according to a boundary domain of the current task completion set to obtain an inner approximation set and an outer approximation set of the current task completion set; the inner approximation set corresponds to the current task completion set before extension, and the outer approximation set corresponds to a set corresponding to the inner approximation set after extension; the inner approximation set and the outer approximation set are overlapped in an approximate area to determine a rough set of the current task completion set, and the rough set is taken as a decision attribute set; current task information in the current task completion sub-set is used to determine an ideal task completion set in a database; the ideal task completion set is a set of the current task completion set in an ideal state, and the ideal task completion set is taken as a condition attribute set; the ideal task completion set includes an ideal task completion sub-set, and the ideal task completion sub-set at least includes an ideal task progress set, an ideal task cost set and an ideal task quality set; a decision attribute feature in the decision attribute set is mapped and matched with a condition attribute feature in the condition attribute set to obtain the data mapping relationship; the decision attribute feature and the condition attribute feature are information features in an attribute set.

[0012] The embodiments of the present application are advantageous in that the current task completion degree is mapped with the ideal task completion degree, the data in the current task completion degree is one-to-one corresponding to the data in the ideal state bar, a mutual mapping relationship between the two is obtained, and a judgment basis is provided for determining the health degree of each stage of the project.

[0013] In an implementation, the decision attribute features of the decision attribute set are matched with the condition attribute features of the condition attribute set to obtain the data mapping relationship, specifically including: extracting decision information features in the decision attribute features; and constructing a decision attribute structure tree according to root nodes of reference attributes in the decision information features; wherein the decision information features include subtask names, subtask features and subtask progress; extracting condition information features in the condition attribute features; and constructing a condition attribute structure tree according to root nodes of reference attributes in the condition information features; wherein the condition information features correspond to subtask information in an ideal state, and include subtask names, subtask features and reference subtask progress; fusing similar information features between the decision attribute structure tree and the condition attribute structure tree according to a preset similarity threshold to obtain a similarity structure tree; determining a containing mapping relationship between the decision information features and the condition information features according to the similarity structure tree; wherein the containing mapping relationship is a superior-inferior containing relationship between the decision information features and the condition information features; matching the decision attribute features with the condition attribute features according to the containing mapping relationship and the similarity structure tree, and generating a mapping matching table; and determining the data mapping relationship between the decision attribute features and the condition attribute features according to the mapping matching table.

[0014] In an implementation, the current task completion degrees of the subtasks in the current task completion degree are respectively cross-matched according to the optimized data mapping relationship to determine the health degrees of the stages of the project, specifically including: discretizing an overlapping data feature region between the current task completion degree and the ideal task completion degree by a preset cross-gradient function; reversely fitting the discretized overlapping data feature region with a minimum objective function of the cross-gradient function to obtain a non-overlapping data feature region between the current task completion degree and the ideal task completion degree; wherein the non-overlapping data feature region is a cross-non-overlapping region of the overlapping data feature region; mapping matching the optimized data mapping relationship with the non-overlapping data feature region to obtain a matching result; wherein the matching result at least includes matched paths, a matched number, unmatched paths and an unmatched number; matching the current task completion degree with the ideal task completion degree according to the matching result to obtain a matching estimation value; and predicting the matching estimation value to determine the health degrees of the stages of the project.

[0015] The health degrees of the stages of the project are determined by the health degrees of the stages of the project, which is beneficial to multi-level monitoring of each stage, determining health degrees of each level, quantifying completion of each stage of the project, and reflecting risks of the current task.

[0016] In an implementation, before the health degree of each stage of the project is input into the trained neural network model to obtain the evaluation index of each stage of the project, the method further comprises: dividing the health degree of each stage of the project by a weight threshold to obtain a weight threshold; adjusting the node weight of the output layer of the error function corresponding to the weight threshold to obtain a weight correction amount, and correcting the weight threshold in real time by using the weight correction amount to obtain a weight of the health degree, and taking the weight of the health degree as the weight of the neural network model.

[0017] In an implementation, the health degree of each stage of the project is input into the trained neural network model to obtain the evaluation index of each stage of the project, specifically comprising: combining the trained neural network model with a preset PSO particle swarm model to obtain an index evaluation model; inputting the health degree of each stage of the project into the index evaluation model, and optimizing the weight and the position vector by repeated iteration and update of the index evaluation model to obtain an optimized weight and an optimized position vector; wherein the position vector is an output quantity of the preset PSO particle swarm model; performing Pearson correlation analysis on the optimized weight and the optimized position vector to obtain a conflict degree index; wherein the conflict degree index represents the correlation degree between the optimized weight and the optimized position vector; integrating the conflict degree index and the health degree of each stage of the project to obtain the evaluation index; wherein the evaluation index represents the compliance of each task in each stage of the project.

[0018] In an implementation, according to the index system, the correlation degree level of the project full life cycle is judged to obtain the comprehensive correlation degree level of the project full life cycle, specifically comprising: obtaining the evaluation index of each stage in the project full life cycle; according to a preset stage monitoring index level, the evaluation index is judged by domain evaluation to obtain the stage correlation membership level of each stage; wherein the stage correlation membership level comprises: poor correlation degree, general correlation degree, good correlation degree and excellent correlation degree; according to a preset full life cycle monitoring index level, the evaluation index in each stage is judged by classical domain evaluation to obtain the quality correlation membership level of the project full life cycle; wherein the quality correlation membership level comprises: high correlation degree, medium correlation degree and low correlation degree; the stage correlation membership level and the quality correlation membership level are combined to obtain the comprehensive correlation degree level of the project full life cycle.

[0019] The project whole life cycle comprehensive correlation degree grade is beneficial to judging the correlation degree of tasks in each stage of the project, performing correlation analysis on the whole life cycle of the project, reflecting the coordination between each stage of the project through the comprehensive correlation degree grade, and being beneficial to the manager to timely understand the correlation relationship of the whole project.

[0020] In a feasible implementation, after the correlation degree grade of the whole life cycle of the project is determined according to the index system, the method further includes: constructing a whole-process monitoring table of the project by the comprehensive correlation degree grade of each stage of the project and the whole life cycle of the project and the index system; delivering the whole-process monitoring table of the project to the management system of the related personnel to perform visual display of the whole-process monitoring table of the project; performing data threshold determination on the whole-process monitoring table of the project according to a preset threshold, and marking and warning the data less than the preset threshold to facilitate the related personnel to manage the process of the project.

[0021] In another aspect, the embodiment of the present application also provides a project health degree monitoring and analysis device based on behavior analysis, which includes: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, so that the at least one processor can execute the project health degree monitoring and analysis method based on behavior analysis in any of the above embodiments.

[0022] The present application provides a project health degree monitoring and analysis method and device based on behavior analysis, which realizes monitoring of the whole life cycle of the project by monitoring and analyzing the health degree of each stage of the project, combining the index system and comprehensive correlation degree grade of the whole life cycle of the project, and displaying the index completion situation of the whole cycle of the project and the correlation degree of each stage to the project manager, which is helpful for the manager to better manage the project process, simplifies the complicated management process, accurately warns the health degree of each stage of the project, helps the manager to timely regulate the related stage, intelligently reminds the manager to deal with the problems, improves the efficiency of project management, perfects the management of the project, and reduces the hidden risks in the project management process. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and those skilled in the art can also obtain other drawings according to these drawings without creating any creative labor. In the drawings:

[0024] Figure 1 A flow chart of a project health monitoring analysis method based on behavior analysis is provided for the embodiments of the present application.

[0025] Figure 2 A rough set boundary field schematic diagram is provided for the embodiments of the present application.

[0026] Figure 3 A structural schematic diagram of a project health monitoring analysis device based on behavior analysis is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0027] In order to make the person skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by the person skilled in the art without creative labor should be within the scope of protection of the present application.

[0028] The embodiments of the present application provide a project health monitoring analysis method based on behavior analysis, as shown in Figure 1 The project health monitoring analysis method based on behavior analysis specifically includes steps S101-S106:

[0029] S101, according to the initial project health parameter values in each stage of the project obtained in advance, the real-time task completion progress of each stage of the project is calculated to obtain the current task completion degree of each stage of the project.

[0030] Specifically, the information of each stage in the whole life cycle of the project is obtained. Each stage at least includes: project demand stage, project design stage, project development stage, project test stage, project acceptance stage and project feedback stage.

[0031] In an embodiment, for a software APP applied to a user, from the life cycle of the software project, the APP generally needs to go through the demand analysis stage related to the market, then go through the market research, preliminarily plan and design the APP, then organize professional personnel to develop the software, then perform various tests and acceptance, preliminarily put into the market, then collect the feedback results of the users, and then facilitate the subsequent improvement and update of the APP. The initial planning information of these stages is obtained, that is, the planning information of each stage is obtained.

[0032] Further, the initial project health parameter values in each stage are extracted. The initial project health parameters at least include: project schedule value, project cost value and project quality value. Specifically, the initial project health parameter values of each step, board and task quantity of the initial plan information of each stage are obtained, and the initial project health parameter values are the specific reference values in the ideal state corresponding to each stage in the entire project operation.

[0033] Further, according to the NPV present value model, the project cost is processed by cost discount value to obtain the initial cost value. The time value and the task quantity value in the project schedule are converted by ratio to obtain the initial schedule value. The project quality value, the initial cost value and the initial schedule value are constructed into a database to obtain the initial database.

[0034] In one embodiment, according to the NPV present value model in the whole life cycle, that is, wherein, is the cost discount value in the ith stage, is the present value cost value in the ith stage, is the time cost value in the ith stage, is the task quantity cost value in the ith stage, is the net recovery discount value in the ith stage. According to wherein, P is the initial cost value, is the planned project cost in the ith stage, t is the project schedule time, is the initial project health parameter value in the ith stage, the time cost value in the ith stage is ratio-operated with the initial project health parameter value in the ith stage to obtain the time value in the project schedule, the task quantity cost value in the ith stage is ratio-operated with the initial project health parameter value in the ith stage to obtain the task quantity value in the project schedule, and then the time value and the task quantity value in the project schedule are correspondingly combined to obtain the initial schedule value in the ith stage. Then, the initial database in the ith stage is constructed according to the project quality value in the initial project health parameter value in the ith stage, the initial cost value in the ith stage and the initial schedule value to obtain the initial database in the ith stage.

[0035] Further, the real-time task quantity in each stage is obtained. The real-time task quantity at least includes: current task schedule value, current task cost value and current task quality value. The current task cost value is input into the initial database to determine the current task completion degree of each stage of the project.

[0036] As a feasible implementation, the real-time task amount in a stage is acquired in real time, wherein the data of the real-time task amount and the initial project health parameter value are in a corresponding relationship, i.e., the project health parameter value of the current stage, and then the current task cost in the real-time task amount is input into the initial database constructed, compared with the corresponding initial cost value, and combined with the current task progress value and the current task quality value to determine the current task completion degree of the stage.

[0037] In S102, the current task completion degree is matched with the preset ideal task completion degree in relation to the data of the rough set, to obtain the data mapping relationship between the current task completion degree and the ideal task completion degree.

[0038] Specifically, the current task completion set corresponding to the current task completion degree is acquired. The current task completion set includes a current task completion sub-set, and the current task completion sub-set at least includes a current task progress set, a current task cost set and a current task quality set.

[0039] As a feasible implementation, the completion degree of each board or step in the current task in a stage is taken as a set amount to construct the current task completion set, and the set at least includes the set related to the current task progress, the set related to the current task cost and the set related to the current task quality, wherein the current task quality is the proportion of the defects of the current task engineering.

[0040] Further, the current task completion set is extended by a preset distance to obtain the inner approximate set and the outer approximate set of the current task completion set. The inner approximate set corresponds to the current task completion set before the extension, and the outer approximate set corresponds to the set corresponding to the current task completion set after the extension except the inner approximate set.

[0041] In an embodiment, first, the boundary domain of the current task completion set is determined according to the current task completion set, and then the boundary line of the boundary domain is extended by a preset distance to obtain the inner and outer two-part regions of the boundary line, i.e., according to , the boundary domain of the current task completion set is obtained , wherein X is the current task completion set, is the equivalence relation amount of the current task completion set. Then according to , the inner approximate set of the current task completion set is obtained , according to , the outer approximate set of the current task completion set is obtained , wherein U is the set region outside the boundary domain.

[0042] Further, the inner approximation set and the outer approximation set are overlapped to determine a rough set of the current task completion set.

[0043] In one embodiment, Figure 2 A boundary field diagram of the rough set is provided for the embodiment of the present application, as shown in Figure 2 As shown in , the overlapping approximation degree precision is obtained, and then the range of the overlapping approximation region of the current task completion set X is determined according to the overlapping approximation degree precision , the overlapping approximation degree precision satisfies , and |X| is the cardinality of X, , the current task completion set X is a rough set, and the rough set at this time is positioned as the decision attribute set for the decisive judgment.

[0044] Further, according to the current task information in the current task completion sub-set, an ideal task completion set is determined in the database. The ideal task completion set is a set of the current task completion set in an ideal state, and the ideal task completion set is determined as a condition attribute set, that is, an attribute set used for reference and comparison with the decision attribute set. The ideal task completion set includes an ideal task completion sub-set, and the ideal task completion sub-set at least includes an ideal task progress set, an ideal task cost set and an ideal task quality set.

[0045] As a feasible implementation, the current task progress set in the most ideal state in a stage, that is, the ideal task completion set, is obtained in the database. The ideal task completion set has the ideal task completion sub-set which is the optimal task completion set under the condition of the plan, and is used as the comparison set of the current task completion set.

[0046] Further, the decision attribute features in the decision attribute set and the condition attribute features in the condition attribute set are mapped and matched to obtain a data mapping relationship. The decision attribute features and the condition attribute features are information features in the attribute set.

[0047] First, decision information features in the decision attribute features are extracted, and a decision attribute structure tree is constructed according to the root node of the reference attribute in the decision information features. The decision information features include sub-task names, sub-task features and sub-task progress. Condition information features in the condition attribute features are extracted, and a condition attribute structure tree is constructed according to the root node of the reference attribute in the condition information features. The condition information features correspond to the sub-task information in the ideal state, and include the sub-task names, the sub-task features and the reference sub-task progress.

[0048] In one embodiment, first, the decision information features in the decision attribute features are determined respectively, that is, the reference attributes of the decision attribute structure tree are obtained, including the reference attributes related to the subtask name, subtask feature and subtask progress of the subtask, and then the root nodes of the reference attributes in the decision information features are determined according to the preset order of the blocks or steps in a stage, and the decision attribute structure tree is constructed. Then the reference attributes in the condition information features are obtained, including the subtask name of each subtask information in the ideal state, the subtask feature and the reference subtask progress, and then the condition attribute structure tree is established according to the root nodes of the reference attributes in the condition information features.

[0049] Then, according to the preset similarity threshold, the similar information features between the decision attribute structure tree and the condition attribute structure tree are fused to obtain a similarity structure tree.

[0050] In one embodiment, if each root node of the reference attributes in the decision information features and each root node of the reference attributes in the condition information features satisfy the maximum similarity threshold, the similar information features between the decision attribute structure tree and the condition attribute structure tree are fused to obtain a fused similarity structure tree, and the similarity structure tree contains two kinds of attributes in the decision information features and the condition information features.

[0051] Then, according to the similarity structure tree, the inclusion mapping relationship between the decision information features and the condition information features is determined. The inclusion mapping relationship is the superior-inferior inclusion relationship between the decision information features and the condition information features.

[0052] In one embodiment, for the fused root nodes in the similarity structure tree, there is also an inclusion mapping relationship of attribute features, that is, the decision attribute features are higher than the condition attribute features, and then the mapping relationship of this part, that is, the inclusion mapping relationship, is determined.

[0053] Finally, according to the inclusion mapping relationship and the similarity structure tree, the decision attribute features and the condition attribute features are mapped and matched, and a mapping and matching table is generated. According to the mapping and matching table, the data mapping relationship between the decision attribute features and the condition attribute features is determined.

[0054] In one embodiment, according to the similarity mapping relationship corresponding to the similarity structure tree, and in combination with the superior-inferior inclusion mapping relationship, the data mapping relationship between the current task completion degree and the ideal task completion degree in the similarity level and the inclusion relationship level is determined, that is, the data mapping relationship between the decision attribute features and the condition attribute features is determined, and a direct mapping and matching table is generated accordingly and stored in the backend database.

[0055] S103, correcting the data mapping relationship to obtain an optimized data mapping relationship, and respectively performing cross mapping matching on the current task completion degrees of the sub-tasks in the current task completion degree according to the optimized data mapping relationship to determine the health degrees of the various stages of the project.

[0056] Specifically, first, data in a mapping matching table corresponding to the data mapping relationship is cleaned, bad data is corrected, and error or garbled data is deleted to obtain an accurate and high data mapping relationship.

[0057] Further, the overlapping data feature region between the current task completion degree and the ideal task completion degree is discretized by a preset cross gradient function. The discretized overlapping data feature region and the minimum objective function of the cross gradient function are reversely fitted in similarity to obtain a non-overlapping data feature region between the current task completion degree and the ideal task completion degree. The non-overlapping data feature region is a non-overlapping region of the overlapping data feature region.

[0058] In one embodiment, the overlapping data feature region between the current task completion degree and the ideal task completion degree is discretized by a preset cross gradient function, the overlapping data feature region is constrained by a simulated cross gradient, the data corresponding to the current task completion degree and the ideal task completion degree are associated, the association is data cross, the discretized overlapping data feature region is captured by the cross gradient function, and a cross rule between the two is obtained, that is , to obtain a cross rule between the current task completion degree and the ideal task completion degree , wherein represents the current task completion degree attribute feature, represents the ideal task completion degree attribute feature, then the cross rule is minimized by a minimum objective function of the cross gradient function, and the cross rule after the minimization is reversely fitted in linear similarity by a check factor in the cross rule and a gradient vector of the cross gradient function, so that an attribute feature region other than the cross rule is obtained, that is, a non-overlapping data feature region between the current task completion degree and the ideal task completion degree.

[0059] Further, the optimized data mapping relationship is mapped and matched with the non-overlapping data feature area to obtain a matching result. The matching result at least includes a matched path, a matched number, an unmatched path, and an unmatched number. Mapping and matching the optimized data mapping relationship with the non-overlapping data feature area can obtain all relationships between the task completion degree and the ideal task completion degree, including the mapping relationship that can be mapped and the non-mapping relationship that exists but cannot be mapped, and then obtain the matching result between the task completion degree and the ideal task completion degree in a stage of the project.

[0060] Further, according to the matching result, the current task completion degree and the ideal task completion degree are matched to obtain a matching estimation value. The matching estimation value is expected to be predicted to determine the health degree of each stage of the project.

[0061] In an embodiment, according to the matching result between the two, the matching degree is estimated to obtain a matching estimation value. Due to the instability of calculation, the matching estimation value is expected to be predicted, and the average value and the standard deviation of the expected prediction are used to inversely fine-tune the linear dispersion degree of the matching estimation value to obtain a final fine-tuned matching estimation value, that is, the health degree of each stage of the project.

[0062] S104, inputting the health degree of each stage of the project into the trained neural network model to obtain an evaluation index of each stage of the project.

[0063] Specifically, the health degree of each stage of the project is divided by a weight threshold to obtain a weight threshold. The error function corresponding to the weight threshold is output to adjust the node weight of the output layer to obtain a weight correction amount, and the weight threshold is corrected in real time through the weight correction amount to obtain a weight of the health degree. The weight of the health degree is used as the weight of the neural network model.

[0064] In an embodiment, the neural network model generally has an input layer, a hidden layer, and an output layer. The node weight of the output layer is adjusted through the error function corresponding to the weight threshold, so that the network adjustment gradually reaches a minimum value. The output layer is back-propagated through the combination of the error function and the excitation function to obtain a weight correction amount of the output layer node weight. Then, the weight threshold is corrected and adjusted in real time to obtain the weight correction amount.

[0065] Further, the trained neural network model is combined with a preset PSO particle swarm model to obtain an index evaluation model.

[0066] In one embodiment, first, the initialization algorithm parameters in the PSO particle swarm model are acquired. Then, the sample training data is input, and the fitness function value of each particle is determined. The particles update the speed and position of the particle swarm and perform iterative operations, and the optimal solution is repeatedly updated. Then, the position vector of the global optimal particle is combined with the weight and threshold process of the neural network to optimize the neural network model, and finally, the index evaluation model is obtained.

[0067] Further, the health degree of each phase of the project is input into the index evaluation model, and the weight and position vector are optimized through repeated iteration and update of the index evaluation model, to obtain the optimized weight and optimized position vector. The position vector is the output quantity of the preset PSO particle swarm model.

[0068] Further, the optimized weight and the optimized position vector are subjected to Pearson correlation analysis to obtain a conflict degree index. The conflict degree index represents the correlation degree between the optimized weight and the optimized position vector.

[0069] It should be noted that the Pearson correlation analysis method is a mathematical statistical method for the correlation degree of two continuous variables. Then, according to the target decision matrix, the optimized weight and the optimized position vector are subjected to target judgment, and then the Pearson conflict coefficient of the optimized weight and the optimized position vector is calculated to obtain the conflict index therebetween.

[0070] Further, the conflict degree index and the health degree of each phase of the project are integrated to obtain an evaluation index. The evaluation index represents the compliance of each task in each phase of the project.

[0071] In one embodiment, the conflict degree index of each phase of the project and the health degree of each phase of the project are combined to obtain an evaluation index that can reflect the compliance of each phase of the project.

[0072] S105, constructing an index system based on the whole life cycle of the project through the evaluation index of each phase of the project.

[0073] Specifically, the evaluation index of each phase of the project is acquired, and then the index system of the whole life cycle of the project is constructed according to the evaluation index of each phase of the project. Each phase of the whole life cycle of the project corresponds to the evaluation index thereof.

[0074] S106, according to the index system, making a correlation degree level judgment on the whole life cycle of the project to obtain a comprehensive correlation degree level of the whole life cycle of the project.

[0075] Specifically, the evaluation index of each phase of the whole life cycle of the project is acquired.

[0076] Further, according to the preset stage monitoring index level, the evaluation indexes are evaluated and judged in the field, and stage correlation membership levels of each stage are obtained. The stage correlation membership levels include poor correlation, general correlation, good correlation and excellent correlation.

[0077] Further, according to the preset whole life cycle monitoring index level, the evaluation indexes in each stage are evaluated and judged in the classic field, and a quality correlation membership level of the whole life cycle of the project is obtained. The quality correlation membership level includes high correlation, medium correlation and low correlation. The stage correlation membership level and the quality correlation membership level are combined to obtain a comprehensive correlation level of the whole life cycle of the project.

[0078] In one embodiment, first, the information of the current task in each stage of the project is evaluated and judged in a small range in the field, the correlation between the information is determined, that is, the stage correlation membership level of each stage, and then the whole stage of the whole life cycle of the project is evaluated and judged in a large range in the classic field, and the quality correlation membership level of the whole life cycle of the project is obtained.

[0079] Further, the comprehensive correlation level and the index system are visualized to complete the process management of the project by the related personnel. The comprehensive correlation level of each stage of the project and the whole life cycle of the project and the index system are used to construct a project whole process monitoring table. The project whole process monitoring table is transmitted to the management system of the related personnel to perform visual display of the project whole process monitoring table.

[0080] Further, according to the preset threshold, data threshold judgment is performed on the project whole process monitoring table, and data less than the preset threshold is marked and warned to facilitate the process management of the project by the related personnel.

[0081] In addition, the embodiment of the application also provides a project health degree monitoring and analyzing device based on behavior analysis, as shown in Figure 3 The project health degree monitoring and analyzing device 300 based on behavior analysis specifically includes:

[0082] At least one processor 301, and a memory 302 connected with the at least one processor 301. The memory 302 stores instructions executable by the at least one processor 301, so that the at least one processor 301 can perform:

[0083] According to the initial project health parameter value of each stage of the project obtained in advance, the real-time task completion progress of each stage of the project is calculated, and the current task completion degree of each stage of the project is obtained.

[0084] The current task completion degree is matched with the preset ideal task completion degree in relation to rough set data, to obtain a data mapping relationship between the current task completion degree and the ideal task completion degree;

[0085] The data mapping relationship is corrected to obtain an optimized data mapping relationship, and the current task completion degrees of the subtasks in the current task completion degree are respectively cross-mapped and matched according to the optimized data mapping relationship, to determine the health degrees of the stages of the project; wherein the current task completion degree includes current task completion degrees of the subtasks;

[0086] The health degrees of the stages of the project are input into the trained neural network model, to obtain evaluation indexes of the stages of the project;

[0087] The evaluation indexes of the stages of the project are used to construct an index system based on the whole life cycle of the project; wherein the whole life cycle of the project is the sum of the stage periods of the stages of the project;

[0088] The whole life cycle of the project is judged in relation to the index system, to obtain a comprehensive correlation degree grade of the whole life cycle of the project.

[0089] The application provides a project health degree monitoring and analysis method and device based on behavior analysis, which realizes monitoring of the whole life cycle of the project by monitoring and analyzing the health degrees of the stages of the project, in combination with the index system and the comprehensive correlation degree grade of the whole life cycle of the project, and helps the project manager to better manage the project process, simplifies the complicated management process, accurately warns the health degrees of the stages of the project, helps the manager to timely regulate the relevant stages, intelligently reminds the manager of the problems to be handled, improves the efficiency of project management, perfects the management of the project engineering, and reduces the hidden risks in the project management process.

[0090] Each of the embodiments in the application is described in a progressive manner, and the same or similar parts of each of the embodiments can be referred to each other, and each of the embodiments mainly describes the differences from other embodiments. Especially, the device and the non-volatile computer storage medium embodiments are basically similar to the method embodiments, so the description is relatively simple, and the related parts can be referred to the part of the description of the method embodiments.

[0091] The above-described embodiments of the application have several aspects, no single one of which is solely responsible for the application's desirable attributes. Without limiting the scope of this application, other aspects of the application will become apparent from consideration of the drawings and following detailed description, it being understood that such changes in the details are within the scope of this application. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation, such that the terminology or phraseology of the present specification is to be interpreted by those skilled in the art in light of the teachings and

[0092] The specific embodiments described herein have many advantages over conventional methods. Although only a limited number of embodiments of the application are described herein, it is to be understood that the application is not limited to this precise embodiments and that changes can be made to the embodiments without departing from the scope of the application. For example, the order of steps can be changed, or replaced, or one or more steps can be changed, or replaced, or one or more steps can be omitted. It is therefore intended that the application not be limited to the described embodiments, but that it include all modifications and alternatives within the scope and spirit of the present application.

Claims

1. A project health monitoring and analysis method based on behavioral analysis, characterized in that, The method includes: calculating the real-time task completion progress of each stage of the project based on the initial project health parameter values ​​obtained in advance, and obtaining the current task completion degree of each stage of the project; The current task completion degree is matched with the preset ideal task completion degree using rough set data to obtain the data mapping relationship between the current task completion degree and the ideal task completion degree; The data mapping relationship is corrected to obtain an optimized data mapping relationship. Based on the optimized data mapping relationship, the current task completion degree of several sub-tasks in the current task completion degree is cross-mapped and matched to determine the health of each stage of the project. The current task completion degree includes the current task completion degree of several sub-tasks. The health status of each stage of the project is input into the trained neural network model to obtain the evaluation index of each stage of the project. By using the evaluation indicators for each stage of the project, an indicator system based on the entire project lifecycle is constructed; wherein, the entire project lifecycle is the sum of the stage cycles of each stage of the project. Based on the aforementioned indicator system, the correlation level of the entire project lifecycle is determined to obtain the comprehensive correlation level of the entire project lifecycle. Based on the optimized data mapping relationship, the current task completion degree of several sub-tasks in the current task completion degree is cross-mapped and matched to determine the health of each stage of the project. Specifically, this includes: discretizing the overlapping data feature region between the current task completion degree and the ideal task completion degree by using a preset cross gradient function. The overlapping data feature regions after discretization are backfitted with the minimum objective function of the cross gradient function to obtain the non-overlapping data feature regions between the current task completion degree and the ideal task completion degree; wherein, the non-overlapping data feature regions are the non-overlapping regions where the overlapping data feature regions intersect. The optimized data mapping relationship is mapped and matched with the non-overlapping data feature regions to obtain the matching result; The matching results include at least: matched paths, number of matched paths, unmatched paths, and number of unmatched paths; Based on the matching results, the current task completion rate is matched with the ideal task completion rate to obtain a matching estimate; and the matching estimate is used to predict the expected value to determine the health of each stage of the project. The health status of each stage of the project is input into the trained neural network model to obtain the evaluation index of each stage of the project. Specifically, the trained neural network model is combined with the preset PSO particle swarm model to obtain the index evaluation model. The health status of each stage of the project is input into the indicator evaluation model, and the weights and position vectors are optimized through repeated iterative updates of the indicator evaluation model to obtain optimized weights and optimized position vectors; wherein, the position vector is the output of the preset PSO particle swarm model. Pearson correlation analysis is performed on the optimized weights and the optimized position vectors to obtain a conflict degree index; wherein, the conflict degree index represents the degree of correlation between the optimized weights and the optimized position vectors. The conflict level index is integrated with the health status of each stage of the project to obtain the evaluation index; wherein the evaluation index represents the achievement status of each task in each stage of the project. Based on the aforementioned indicator system, the correlation level of the entire project lifecycle is determined to obtain the comprehensive correlation level of the entire project lifecycle. Specifically, this includes obtaining the evaluation indicators for each stage of the entire project lifecycle. Based on the preset stage monitoring indicator levels, the evaluation indicators are evaluated and judged to obtain the stage association membership level of each stage; wherein, the stage association membership level includes: poor association degree, average association degree, good association degree, and excellent association degree. Based on the preset full life cycle monitoring index levels, several evaluation indicators in each stage are evaluated and judged using classical domain methods to obtain the quality association membership level of the project's full life cycle; wherein, the quality association membership level includes: high association, medium association, and low association. The stage association level and the quality association level are combined to obtain the comprehensive association level of the entire project life cycle.

2. The project health monitoring and analysis method based on behavior analysis according to claim 1, characterized in that, Based on the initial project health parameter values ​​obtained in advance for each stage of the project, the real-time task completion progress of each stage of the project is calculated to obtain the current task completion rate of each stage of the project. Specifically, this includes: obtaining information on each stage in the entire life cycle of the project; wherein, each stage includes at least: project requirements stage, project design stage, project development stage, project testing stage, project acceptance stage, and project feedback stage. Extract the initial project health parameter values ​​from each stage; wherein, the initial project health parameters include at least: project schedule value, project cost value, and project quality value; Based on the NPV present value model, the project cost is discounted to obtain the initial cost value; the ratio of time value to task quantity value in the project schedule is converted to obtain the initial schedule value. The initial database is constructed by combining the project quality value, the initial cost value, and the initial schedule value. Obtain the real-time task volume for each stage; wherein, the real-time task volume includes at least: the current task progress value, the current task cost value, and the current task quality value; The current task cost value is input into the initial database to determine the current task completion rate at each stage of the project.

3. The project health monitoring and analysis method based on behavioral analysis according to claim 1, characterized in that, The current task completion degree is matched with the preset ideal task completion degree using rough set data to obtain the data mapping relationship between the current task completion degree and the ideal task completion degree. Specifically, this includes: obtaining the current task completion set corresponding to the current task completion degree; wherein, the current task completion set contains a current task completion subset, and the current task completion subset includes at least: the current task progress set, the current task cost set, and the current task quality set. Based on the boundary region of the current task completion set, the current task completion set is expanded by a preset distance to obtain an inner approximate set and an outer approximate set of the current task completion set; wherein, the inner approximate set corresponds to the current task completion set before expansion, and the outer approximate set corresponds to the set after expansion other than the inner approximate set; The inner approximation set and the outer approximation set are overlapped to determine the rough set of the current task completion set, and the rough set is used as the decision attribute set. Based on the current task information in the current task completion subset, an ideal task completion set is determined in the database; wherein, the ideal task completion set is the set of the current task completion set under an ideal state, and the ideal task completion set is determined as a set of conditional attributes; The ideal task completion set includes an ideal task completion subset, which includes at least: an ideal task progress set, an ideal task cost set, and an ideal task quality set. The decision attribute features in the decision attribute set are mapped and matched with the condition attribute features in the condition attribute set to obtain the data mapping relationship; wherein the decision attribute features and the condition attribute features are information features in the attribute set.

4. The project health monitoring and analysis method based on behavioral analysis according to claim 3, characterized in that, The decision attribute features of the decision attribute set and the condition attribute features of the condition attribute set are mapped and matched to obtain the data mapping relationship. Specifically, this includes: extracting decision information features from the decision attribute features; and constructing a decision attribute structure tree based on the root node of the baseline attribute in the decision information features; wherein the decision information features include subtask name, subtask features, and subtask progress. Extract the condition information features from the condition attribute features; and construct a condition attribute structure tree based on the root node of the baseline attribute in the condition information features; wherein, the condition information features correspond to the information of each subtask under ideal conditions, including the name of each subtask, subtask features and reference subtask progress; Based on a preset similarity threshold, the similar information features between the decision attribute structure tree and the condition attribute structure tree are fused to obtain a similarity structure tree; Based on the similarity structure tree, the inclusion mapping relationship between the decision information feature and the condition information feature is determined; wherein, the inclusion mapping relationship is the hierarchical inclusion relationship between the decision information feature and the condition information feature; Based on the mapping relationship and the similarity structure tree, the decision attribute features and the condition attribute features are mapped and matched, and a mapping matching table is generated; Based on the mapping matching table, the data mapping relationship between the decision attribute features and the condition attribute features is determined.

5. The project health monitoring and analysis method based on behavioral analysis according to claim 1, characterized in that, Before inputting the health scores of each stage of the project into the trained neural network model to obtain the evaluation indicators of each stage of the project, the method further includes: dividing the health scores of each stage of the project into weight thresholds to obtain weight thresholds; The error function corresponding to the weight threshold is used to adjust the node weights of the output layer to obtain the weight correction amount. The weight threshold is then corrected in real time using the weight correction amount to obtain the health weight. The health weight is then used as the weight of the neural network model.

6. The project health monitoring and analysis method based on behavioral analysis according to claim 1, characterized in that, After determining the correlation level of the entire project lifecycle based on the indicator system and obtaining the comprehensive correlation level of the entire project lifecycle, the method further includes: constructing a project full-process monitoring table based on the various stages of the project, the comprehensive correlation level of the entire project lifecycle, and the indicator system. The project monitoring form is transmitted to the stakeholder management system for visualization. Based on a preset threshold, the data threshold of the project's full-process monitoring table is judged, and data below the preset threshold is marked with an early warning prompt to facilitate the stakeholders' process management of the project.

7. A project health monitoring and analysis device based on behavioral analysis, characterized in that, The device includes: at least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, enabling the at least one processor to execute a project health monitoring and analysis method based on behavioral analysis according to any one of claims 1-6.

Images