CN106779366A - project management process performance model and its construction method and performance model management system - Google Patents
project management process performance model and its construction method and performance model management system Download PDFInfo
- Publication number
- CN106779366A CN106779366A CN201611097082.1A CN201611097082A CN106779366A CN 106779366 A CN106779366 A CN 106779366A CN 201611097082 A CN201611097082 A CN 201611097082A CN 106779366 A CN106779366 A CN 106779366A
- Authority
- CN
- China
- Prior art keywords
- model
- project
- performance model
- variable
- defect
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Stored Programmes (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of construction method of project management process performance model, including：(1) multiple variables are drawn by unit of analysis feature, unit project data situation, variable is divided into the fluctuation factor and the defective effect factor；(2) situational variables data, in confirming that can it include model construction；(3) model is built：A, Normality Analysis：Relation to each two variable x is analyzed, and is confirmed whether to belong to normal distribution；B, correlation analysis：Correlation analysis are carried out to each x and y, confirms that the correlation of x and y is strong and weak；C, variance analysis/regression analysis：Variance analysis or regression analysis being carried out respectively for qualitative or quantitative variable, y and all x being analyzed, project management process performance model formula is extracted using matlab.The performance model and the project performance model management system including the model built by methods described are also disclosed.The purpose that the present invention can realize predicting project objective and defect distribution is estimated by the performance model.
Description
Technical field
The present invention relates to project management process performance appraisal technical field, more particularly to a kind of project management process performance
Model and its construction method.
Background technology
Current existing project management process performance model be mainly manifested in it is following some：
(1) existing Project Process performance model biases toward evaluation in application aspect and nonanticipating its Main Function is to evaluate
Performance income of finished item etc., by analyzing the influence factors such as manpower, tissue, evaluates various factors in project performance
Role and influence degree, lays particular stress on post-project evaluating；Existing performance model is slightly weak in terms of the project control and project forecast, and typically
It is qualitatively to predict.
(2) be currently without the special existing performance model instrument of software project process performance modeling tool it is general,
Focus is the factor for collecting each tissue each side from shareholder, such as Prism Model of the Performance instrument, is from all
The angle of stakeholder pattern is set out the appraisement system carried out to project.Objective immutable factor distinctive for software project is then
Do not account for, such as the difference of software type and field can cause production capacity also to differ.
(3) existing performance model does not support that the existing performance model of sustained improvement is directed to specific finished item and carries out
Analytic statistics, the project performance that model was only adapted within the time period of a kind of scale, i.e., one substantially in actual applications is qualitative
Evaluation and prediction, can not with scale of the project increase, the increase of data volume and data type self adaptation, model can not hold
It is continuous to improve.
As can be seen here, above-mentioned existing project management process performance model application it is upper, it is clear that still suffered from it is inconvenient with it is scarce
Fall into, and be urgently further improved.How to found a kind of new project management process performance model and its construction method and
One of performance model management system, real current important research and development problem of category.
The content of the invention
The technical problem to be solved in the present invention is to provide a kind of construction method of project management process performance model, makes it can
Realize estimating the prediction of project objective and defect distribution, so as to overcome the shortcomings of existing project management process performance model.
In order to solve the above technical problems, the present invention provides a kind of construction method of project management process performance model, including
Following steps：
(1) defined variable：Multiple variables are drawn by analyzing our unit's feature, unit project data situation, and according to item
Mesh feature, the fluctuation factor and the defective effect factor are divided into by the variable；
(2) variable analysis：The variable data is analyzed using statistical method, confirms that can the variable data
Include in model construction, retain if it can include in model construction, otherwise remove；
(3) model is built：Specifically include：
A, Normality Analysis：The relation of each two variable x is analyzed using statistical analysis technique, is confirmed whether to belong to
Normal distribution, if belonging to normal distribution, into step B；
B, correlation analysis：Correlation analysis are carried out to each variable x and y using statistical method, the phase of x and y is confirmed
Closing property is strong and weak, if correlation by force if include model construction, otherwise reject；
C, variance analysis/regression analysis：According to our unit's data characteristics, variance analysis is used for qualitatively variable, it is right
Regression analysis is used in quantitative variable, the variance analysis is with regression analysis to y and by all x after the correlation analysis
It is analyzed, and the project management process performance model formula prototype is extracted using matlab.
Used as a modification of the present invention, the project management process performance model formula prototype includes：
1. defect concentration object module formula
For qualitatively variable, obtaining defect concentration object module formula by the variance analysis is：
Y=CLx* (1+u*a* ∑s X1i*c)(1-v*b*∑X2i*c)
Wherein, X1It is the controllable fluctuation factor, X2It is the uncontrollable fluctuation factor；I.e. process performance baseline median value, takes
From historical data；U, v are weight coefficient, u=0.04, v=0.1；A, b, c are regulation coefficient；
For quantitative variable, obtaining defect concentration object module formula by the regression analysis is：
Y=CLx*M* (1+a*X11+b*X12+c*X13+d*X14+e*BX15+f*X21+g*X22)
Wherein, X1It is the controllable fluctuation factor, X2It is the uncontrollable fluctuation factor；I.e. process performance baseline median value, takes
From historical data；M is baseline regulation coefficient, is calculated by bringing history item data Y and Xi into.a、b、c、d、e、f、g：Point
The empirical coefficient of controllable factor and the uncontrollable factor Wei not corresponded to；
2. throughput objectives model formation
The throughput objectives model formation is with the defect concentration object module formula, and the throughput objectives model is public
Formula is different from the regulation coefficient in the defect concentration object module formula；
3. defect injection rate model formation
Wherein, CLXi is each phase process performance baseline median value, takes from historical data；Xi represents each stage；Pi is
Anderson-Darling assays；∑ S=∑s (CLXi* ∑ Pi).
4. Defect removal rate model formation
Wherein, Xij is stage problem number, Xi1 12J is X1J to X12J, represents 12 problem numbers in stage.
Fluctuation factor variable is applied to the throughput objectives model formation and defect concentration target mould in the step
In type formula, defective effect factor variable is applied in the defect injection rate and Defect removal rate model formation.
Further improve, variable includes personnel, environment, standard procedure cutting, complexity and work in the step (1)
Product scale.
Further improve, step (1) medium wave reason and the defective effect factor are also respectively divided into controllable factor and not
Controllable factor.
Further improve, the amount of degree of correlation and change direction between description variable is also calculated in the step B correlation analyses
Count, i.e. Pearson correlation coefficient r, its computing formula is：
Wherein, X, Y represent two independent variables, and X is the correlation factor of object module, and Y is target, and Xi and Yi represent two
I-th value of individual variable,It is the average value of all Xi,It is the average value of all Yi, n is the number of the X and Y of statistics.r
Absolute value show that x is more related to y more greatly, it is contemplated that the model for obtaining is more accurate.
Further improve, the construction method also includes the inspection to the project management process performance model, the inspection
Testing step includes：
Whether A, individual values type variable are close to normal distribution；
Whether each other strong correlation between B, each independent variable；
Whether the mistake in C, model is random, and belong to normal distribution；
D, check whether the situation that there is too strong influence on result in the presence of certain project or certain several project；
The regression accuracy r of E, inspection equation2, i.e. degree of fitting inspection, regression accuracy illustrates that fitting effect is got over closer to 1
It is good；
F, inspection F distributions and t distribution significance probabilities, when probable value is less than or equal to 0.05, it is believed that model result is aobvious
Write.
Further improve, the construction method also includes the improvement step to the project management process performance model, institute
Stating improvement step is：When project data is accumulated to a certain extent, mould is carried out using Monte Carlo simulation instrument Crystal Ball
Intend, using the grown form of model formation and the data volume of accumulation as input quantity, more accurate predicts coefficient value, realizes
Correction to model formation.
The present invention also provides a kind of project management built by above-mentioned project management process performance model construction method
Journey performance model.
The present invention also provides a kind of project performance model management system of the above-mentioned project management process performance model of application,
The model formation of the project management process performance model is used in the project performance model management system by two ways,
First way, the coefficient for X in formula will not change with data accumulation, using the side of built-in formula
Method, the parameter of flexible defining factor X；
The second way, coefficient for X in formula is constantly updated with data accumulation, using the side of in non-built formula
Method, the parameter of flexible defining factor X.
Further improve, the project performance model management system includes performance model management module and performance model application
Module,
The performance model management module includes fluctuation factor maintenance module, defective effect factor maintenance module, productivity ratio
Object module maintenance module, defect concentration object module maintenance module, defect injection rate model maintenance module and Defect removal rate
Model maintenance module；
The performance model application module includes project objective prediction module and defect distribution estimation module, the project mesh
Mark prediction module, for calculating project construction rate, defect concentration, defect injection rate and Defect removal rate；The defect distribution
Estimation module, defect injection rate, Defect removal rate and defect number for calculating each stage stage by stage.
After such design, the present invention at least has advantages below：
Project management process performance model of the present invention is placed in the project achievement by the form of built-in formula or in non-built formula
In effect model management system, by the application to its performance model management module and performance model application module, it is capable of achieving automatic
The project objective prediction of project construction rate, defect concentration, defect injection rate and Defect removal rate is calculated, sublevel can also be realized
The defect distribution of the defect injection rate, Defect removal rate and defect number in section displaying each stage estimates that these are predicted the outcome should
In project integration and quality performance tracking, supported for the whole project control provides necessary data.
Project management process performance model of the present invention also gathers item base data by continuous, according to season, time
Performance model is corrected Deng chronomere, the continuous correction to project performance model management system of the present invention is realized with this,
Finally give and more accurately predict the outcome.
Brief description of the drawings
Above-mentioned is only the general introduction of technical solution of the present invention, in order to better understand technological means of the invention, below
With reference to accompanying drawing, the present invention is described in further detail with specific embodiment.
Fig. 1 is the structure flow chart of project management process performance model of the present invention.
Fig. 2 is the application structure figure of project management process performance model of the present invention.
Fig. 3 is the applicating flow chart of project management process performance model of the present invention.
Fig. 4 is project data collection-item base maintenance of information schematic diagram of the present invention.
Fig. 5 is project data collection-project performance data maintenance figure of the present invention.
Fig. 6 is editor's figure that performance model of the present invention safeguards the fluctuation factor.
Fig. 7 is that performance model of the present invention safeguards that defective effect factor editor schemes.
Fig. 8 is that performance model of the present invention safeguards that throughput objectives model Back ground Information editor schemes.
Fig. 9 is that performance model of the present invention safeguards throughput objectives model component figure.
Figure 10 is performance model application item target prediction figure of the present invention.
Figure 11 is performance model applied defect distribution estimation figure of the present invention.
Figure 12 is performance model application quality performance tracing figure of the present invention.
Figure 13 is performance model application quality performance tracking post figure of the present invention.
Specific embodiment
Project management process performance model of the present invention mainly by the basis of existing model increase variable's attribute and
Addition variable classification, realizes the application function estimated project objective prediction and defect distribution, its specific model building method
It is as follows.
Project management process performance model construction method of the present invention comprises the following steps：
(1) defined variable：Variable is drawn by analyzing our unit's feature, unit project data situation.Variable in the present invention
Both existing model attributes variable is included, such as personnel, environment etc., always according to software project feature, with the addition of standard procedure sanction
Cut, three new variable's attributes of complexity and work product scale；According to software project feature, with the addition of variable classification and because
Subclassification step, the fluctuation factor and the defective effect factor are divided into by variable, and the factor is divided into controllable factor and the uncontrollable factor, with
The predictive ability of accurate different target.
(2) variable analysis：The step uses existing statistical method, for example, draw variance, the standard deviation of data, passes through
Analysis result, confirms that data are in needing to remove or include model construction.
(3) model is built：Specifically include：
A, Normality Analysis：The step is analyzed using existing statistical analysis technique to the relation of each two x, it is determined that
Whether normal distribution is belonged to, if can carry out formula simulation.By the Anderson-Darling of matlab instruments check come
Carry out, the assay can obtain a P value, whether relation has statistical significance between representing x.The value of P such as following table
1。
The P value criterions of table 1
B, correlation analysis：The step carries out correlation analysis to each x and y using existing statistical method, confirm x with
Whether the correlation of y is strong and weak, i.e., related enough, and model construction is included if correlation enough, otherwise rejects.The first step is filtering
Scatterplot, for the point data that peels off, deletes, it is ensured that the accuracy of data and the accuracy of final mask before modeling.Second
Step is the amount number for calculating degree of correlation and change direction between description variable, i.e. Pearson correlation coefficient r, and computing formula is following formula
(1)：
The bigger explanation x of absolute value of r is more related to y, it is contemplated that the model prediction ability for obtaining is more accurate.Wherein, X, Y are represented
Two independent variables, X is the correlation factor of object module, and Y is target, and Xi and Yi represents two i-th values of variable,
It is the average value of all Xi,It is the average value of all Yi, n is the number of the X and Y of statistics.
C, variance analysis/regression analysis：According to our unit's data characteristics, variance analysis is used for qualitatively variable, it is right
Regression analysis is used in quantitative variable.Variance analysis and regression analysis are divided to y and by all x after correlation analysis
Analysis, forms characteristic equation, i.e., it is as follows to extract equation prototype with matlab：
1. defect concentration object module formula
For qualitatively variable, the performance model formula (2) obtained by variance analysis is：
Y=CLx* (1+u*a* ∑s X1i*c)(1-v*b*∑X2i*c) (2)
Wherein, X1It is the controllable fluctuation factor, X2It is the uncontrollable fluctuation factor；I.e. process performance baseline median value, takes
From historical data；U, v are weight coefficient, u=0.04, v=0.1；A, b, c are regulation coefficient, and the value of regulation coefficient passes through this list
History item the data Y and X of position accumulation1i、X2iIt is brought into formula (2) and draws.
For quantitative variable, the performance model formula (3) obtained by regression analysis is：
Y=CLx*M* (1+a*X11+b*X12+c*X13+d*X14+e*BX15+f*X21+g*X22) (3)
Wherein, X1It is the controllable fluctuation factor, X2It is the uncontrollable fluctuation factor；I.e. process performance baseline median value, takes
From historical data；M is baseline regulation coefficient, a, b, c, d, e, f, g：Respectively correspond to the experience of controllable factor and the uncontrollable factor
History item data Y and X that coefficient, baseline regulation coefficient value and empirical coefficient value are accumulated by our unit1i、X2iIt is brought into public affairs
Formula draws in (3).
2. throughput objectives model formation
The throughput objectives model formation with above-mentioned defect concentration object module formula, due to productivity ratio and defect concentration
Factor of influence is identical, and by statistical simulation, the formula prototype for finally obtaining is consistent；But because the Y of the two is different with Xi, therefore
It is different with the regulation coefficient that Xi is obtained to bring historical data Y into.
3. defect injection rate model formation
Wherein, CLXi is each phase process performance baseline median value, i.e.,Xi represents each stage；Pi is
Anderson-Darling assays；∑ S=∑s (CLXi* ∑ Pi).
4. Defect removal rate model formation
Wherein, Xij is stage problem number, Xi1 12J is X1J to X12J, represents 12 problem numbers in stage.
Above-mentioned steps iteration is performed, and fluctuation factor variable is applied into productivity ratio and the generation of defect concentration predictor formula, will
Defective effect factor variable is applied to defect injection rate and the generation of Defect removal rate predictor formula, determines respectively.With reference to our unit
Actual items data, final the result is notable and available for model.
(4) model testing carries out following inspection according to our unit's data characteristics to above-mentioned model：
Whether A, individual values type variable are close to normal distribution.
Whether each other strong correlation between B, each independent variable.
In C, model mistake whether be it is random, normal distribution.
D, check whether the situation that there is too strong influence on result in the presence of certain project or certain several project；The inspection needs this
The actual items data inputting of unit is to being verified in matlab.
The regression accuracy r of E, inspection equation2, i.e. degree of fitting inspection, regression accuracy illustrates that fitting effect is got over closer to 1
It is good.
F, inspection F distributions and t distribution significance probability, i.e. significance test.When usual probable value is less than or equal to 0.05,
Think that model result is significant.
(5) model refinement
Adaptive ability is improved in order to adapt to the increase of scale of the project increase, data volume and data type, need to be to above-mentioned
Project management process performance model formula is improved and corrects, to adapt to more situations, such as new technology, new tool, new
The application of process approach and abandoning for aging method.Its improved method is：When project data is accumulated to a certain extent, illiteracy is used
Special Carlow simulation tool Crystal Ball are simulated.Using the grown form of formula and the data volume of accumulation as input,
That is Y, it is known that number realization set more than 10,000 times, i.e., x values have more than 10,000 samples, can be more accurate predict coefficient
Value, so as to reach the correction to formula.
Be applied to above-mentioned project management process performance model in project performance model management system by the present invention.The project achievement
The WEB platform that effect model management system is based on .NET realizes that the system includes web front-end, service logic end and data persistence
End three-tier architecture, specifically adds JAVASCRIPT, CSS etc. to write by MASTER templates, and backstage is write by C#, SQL etc..Wherein, should
Project management process performance model formula is used in the project performance model management system by two ways, and one kind is due to formula
The coefficient of middle X will not change with data accumulation, then using the method for built-in formula, flexible defining factor X parameter, formula Y
It is placed in program, such as defect injection rate and Defect removal rate model formation are to be directly located journey by the way of built-in formula
In sequence；Another kind can be constantly updated due to the coefficient of X in formula with data accumulation, then using the side of flexibly input formula
Method, such as flexible defining factor parameter, throughput objectives model formation and defect concentration object module formula are public using in non-built
The mode of formula is applied in the system.
Referring to the drawings shown in 1, the construction method of the project performance model management system is as follows：(1) safeguard fluctuation the factor and
The defective effect factor；(2) rank of the fluctuation factor and the defective effect factor is safeguarded；(3) safeguard that throughput objectives model, defect are close
Degree object module and defect injection rate and Defect removal rate model attributes；(4) selection the fluctuation factor or the defective effect factor；
(5) judge it is built-in formula model or in non-built formula model, if built-in formula model, it is to realize item directly to set formula
The structure of mesh performance model management system；If in non-built formula model, need to safeguard that performance model formula realizes project performance
The structure of model management system.
Wherein, as shown in accompanying drawing 4 to 9, the maintaining method of the fluctuation factor and the defective effect factor is mainly each factor of maintenance
Mark, title, rank, the scope of each class value；The submodels such as throughput objectives model be mainly safeguard submodel mark,
Title, the factor of selection influence target；Model formation is safeguarded if using in non-built formula；Defect injection rate and Defect removal
Rate model is classified according to the standard procedure stage, one model of a stage action；Model issue is mainly according to item
Be combined for each submodel by mesh type, field, ultimately forms a model set and carries out issue and uses；Project process data is adopted
Collection is that the basic data of finished item is safeguarded, including the actual scale of project label, title, project, real work
Amount, item types etc..
Referring to the drawings shown in 2, project performance model management system of the present invention includes performance model management module and performance mould
Type application module.The performance model management module includes fluctuation factor maintenance module, defective effect factor maintenance module, productivity ratio
Object module maintenance module, defect concentration object module maintenance module, defect injection rate model maintenance module and Defect removal rate
Model maintenance module.The performance model application module includes project objective prediction module and defect distribution estimation module.
The project performance model application module is based on the basis of performance model management module, and the application in non-built formula is adopted
With the mode called, logical calculated is then directly carried out in the algorithm to built-in formula.It is first when formula is calculated
The factor values in formula are first given, are then brought into by mark and end value is calculated in formula.
Referring to the drawings shown in 3, the project performance model application module is applied in project construction rate target prediction, defect concentration
Target prediction, in defect injection rate and Defect removal rate, its concrete application flow is as follows：(1) selected according to item types, field
Performance model；(2) factor values of the performance model of selection are safeguarded；(3) corresponding performance model formula is called, if built-in formula,
Defect injection rate and/or Defect removal rate are directly calculated using algorithm；If in non-built formula, then formula is called, then carried out
The calculating of productivity ratio and/or defect concentration；Then complete the application of the project performance model application module.
Project performance model management system practical application interface of the present invention is as shown in accompanying drawing 10 to 13.
Project performance model management system of the present invention is by its performance model management module and performance model application module
Application, be capable of achieving the automatic project objective for calculating project construction rate, defect concentration, defect injection rate and Defect removal rate pre-
Survey, can also realize showing stage by stage that the defect distribution of the defect injection rate, Defect removal rate and defect number in each stage is estimated.
It is close that project performance model management system of the present invention may be implemented in productivity ratio before project starts to the project, defect
Degree, defect injection rate and Defect removal rate carry out target prediction, and predicting the outcome for these targets can be applied in project matter
In amount control and quality performance tracking, supported for the whole project control provides necessary data.
Project management process performance model of the present invention constantly gathers item base data, can be according to season, time etc.
Chronomere is corrected to performance model, and the continuous correction to project performance model management system of the present invention is realized with this, is obtained
To more accurately predicting the outcome.
The above, is only presently preferred embodiments of the present invention, and any formal limitation is not made to the present invention, this
Art personnel make a little simple modification, equivalent variations or modification using the technology contents of the disclosure above, all fall within this hair
In bright protection domain.
Claims (10)
1. a kind of construction method of project management process performance model, it is characterised in that comprise the following steps：
(1) defined variable：Multiple variables are drawn by analyzing our unit's feature, unit project data situation, and it is special according to project
Point, the fluctuation factor and the defective effect factor are divided into by the variable；
(2) variable analysis：The variable data is analyzed using statistical method, confirms that can the variable data be included
In model construction, retain if it can include in model construction, otherwise remove；
(3) model is built：Specifically include：
A, Normality Analysis：The relation of each two variable x is analyzed using statistical analysis technique, is confirmed whether to belong to normal state
Distribution, if belonging to normal distribution, into step B；
B, correlation analysis：Correlation analysis are carried out to each variable x and y using statistical method, the correlation of x and y is confirmed
Power, if correlation by force if include model construction, otherwise reject；
C, variance analysis/regression analysis：According to our unit's data characteristics, variance analysis is used for qualitatively variable, for fixed
The variable of amount uses the regression analysis, the variance analysis to be carried out to y and by all x after the correlation analysis with regression analysis
Analysis, and the project management process performance model formula prototype is extracted using matlab.
2. the construction method of project management process performance model according to claim 1, it is characterised in that the project pipe
Reason process performance model formation prototype includes：
1. defect concentration object module formula
For qualitatively variable, obtaining defect concentration object module formula by the variance analysis is：
Y=CLx* (1+u*a* ∑s X1i*c)(1-v*b*∑X2i*c)
Wherein, X1It is the controllable fluctuation factor, X2It is the uncontrollable fluctuation factor；I.e. process performance baseline median value, takes from and goes through
History data；U, v are weight coefficient, u=0.04, v=0.1；A, b, c are regulation coefficient；
For quantitative variable, obtaining defect concentration object module formula by the regression analysis is：
Y=CLx*M* (1+a*X11+b*X12+c*X13+d*X14+e*BX15+f*X21+g*X22)
Wherein, X1It is the controllable fluctuation factor, X2It is the uncontrollable fluctuation factor；I.e. process performance baseline median value, takes from and goes through
History data；M is baseline regulation coefficient, is calculated by bringing history item data Y and Xi into.a、b、c、d、e、f、g：Respectively
The empirical coefficient of correspondence controllable factor and the uncontrollable factor；
2. throughput objectives model formation
The throughput objectives model formation with the defect concentration object module formula, the throughput objectives model formation with
Regulation coefficient in the defect concentration object module formula is different；
3. defect injection rate model formation
Wherein, CLXi is each phase process performance baseline median value, takes from historical data；Xi represents each stage；Pi is Anderson-
Darling assays；∑ S=∑s (CLXi* ∑ Pi).
4. Defect removal rate model formation
Wherein, Xij is stage problem number, Xi1 12J is X1J to X12J, represents 12 problem numbers in stage.
Fluctuation factor variable is applied to the throughput objectives model formation and defect concentration object module public affairs in the step
In formula, defective effect factor variable is applied in the defect injection rate and Defect removal rate model formation.
3. the construction method of project management process performance model according to claim 2, it is characterised in that the step
(1) variable includes personnel, environment, standard procedure cutting, complexity and work product scale in.
4. the construction method of project management process performance model according to claim 3, it is characterised in that the step
(1) medium wave reason and the defective effect factor are also respectively divided into controllable factor and the uncontrollable factor.
5. the construction method of project management process performance model according to claim 2, it is characterised in that the step B
The amount number of degree of correlation and change direction between description variable, i.e. Pearson correlation coefficient r, its calculating are also calculated in correlation analysis
Formula is：
Wherein, X, Y represent two independent variables, and X is the correlation factor of object module, and Y is target, and Xi and Yi represents two changes
I-th value of amount,It is the average value of all Xi,It is the average value of all Yi, n is the number of the X and Y of statistics.R's is exhausted
Show that value x is more related to y more greatly, it is contemplated that the model for obtaining is more accurate.
6. the construction method of project management process performance model according to claim 1, it is characterised in that the structure side
Method also includes the inspection to the project management process performance model, and the checking procedure includes：
Whether A, individual values type variable are close to normal distribution；
Whether each other strong correlation between B, each independent variable；
Whether the mistake in C, model is random, and belong to normal distribution；
D, check whether the situation that there is too strong influence on result in the presence of certain project or certain several project；
The regression accuracy r of E, inspection equation2, i.e. degree of fitting inspection, regression accuracy illustrates that fitting effect is better closer to 1；
F, inspection F distributions and t distribution significance probabilities, when probable value is less than or equal to 0.05, it is believed that model result is significant.
7. the construction method of project management process performance model according to claim 1, it is characterised in that the structure side
Method also includes the improvement step to the project management process performance model, and the improvement step is：When project data runs up to
To a certain degree, it is simulated using Monte Carlo simulation instrument Crystal Ball, by the grown form and product of model formation
Tired data volume is used as input quantity, the more accurate correction for predicting coefficient value, realizing to model formation.
8. the project pipe that a kind of project management process performance model construction method as described in any one of claim 1 to 7 builds
Reason process performance model.
9. the project performance model management system of the project management process performance model described in a kind of application claim 8, it is special
Levy and be, the model formation of the project management process performance model is used for the project performance model management by two ways
In system,
First way, the coefficient for X in formula will not change with data accumulation, using the method for built-in formula, spirit
The parameter of defining factor X living；
The second way, the coefficient for X in formula is clever using the method for in non-built formula with data accumulation continuous renewal
The parameter of defining factor X living.
10. project performance model management system according to claim 9, it is characterised in that the project performance model pipe
Reason system includes performance model management module and performance model application module,
The performance model management module includes fluctuation factor maintenance module, defective effect factor maintenance module, throughput objectives
Model maintenance module, defect concentration object module maintenance module, defect injection rate model maintenance module and Defect removal rate model
Maintenance module；
The performance model application module includes project objective prediction module and defect distribution estimation module, and the project objective is pre-
Module is surveyed, for calculating project construction rate, defect concentration, defect injection rate and Defect removal rate；The defect distribution is estimated
Module, defect injection rate, Defect removal rate and defect number for calculating each stage stage by stage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611097082.1A CN106779366A (en) | 2016-12-02 | 2016-12-02 | project management process performance model and its construction method and performance model management system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611097082.1A CN106779366A (en) | 2016-12-02 | 2016-12-02 | project management process performance model and its construction method and performance model management system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106779366A true CN106779366A (en) | 2017-05-31 |
Family
ID=58883139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611097082.1A Pending CN106779366A (en) | 2016-12-02 | 2016-12-02 | project management process performance model and its construction method and performance model management system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106779366A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112070415A (en) * | 2020-09-17 | 2020-12-11 | 中国航空无线电电子研究所 | Scenarioized baseline and model implementation method established based on software quantitative management process |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101551750A (en) * | 2009-05-15 | 2009-10-07 | 中国科学院软件研究所 | Method of detecting software process reliability and system of using the same |
| US20110191146A1 (en) * | 2010-02-02 | 2011-08-04 | Bank Of America Corporation | Compliance methodology |
| CN102982416A (en) * | 2012-12-05 | 2013-03-20 | 山东浪潮齐鲁软件产业股份有限公司 | Universal implementation model for performance assessment |
| CN103577882A (en) * | 2013-11-14 | 2014-02-12 | 中国联合网络通信集团有限公司 | Quantification project resource control method based on UML |
-
2016
- 2016-12-02 CN CN201611097082.1A patent/CN106779366A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101551750A (en) * | 2009-05-15 | 2009-10-07 | 中国科学院软件研究所 | Method of detecting software process reliability and system of using the same |
| US20110191146A1 (en) * | 2010-02-02 | 2011-08-04 | Bank Of America Corporation | Compliance methodology |
| CN102982416A (en) * | 2012-12-05 | 2013-03-20 | 山东浪潮齐鲁软件产业股份有限公司 | Universal implementation model for performance assessment |
| CN103577882A (en) * | 2013-11-14 | 2014-02-12 | 中国联合网络通信集团有限公司 | Quantification project resource control method based on UML |
Non-Patent Citations (2)
| Title |
|---|
| 吕英杰: "软件过程性能模型的工程化方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 罗娟等: "基于CMMI4的软件项目量化质量管理研究", 《软件导刊》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112070415A (en) * | 2020-09-17 | 2020-12-11 | 中国航空无线电电子研究所 | Scenarioized baseline and model implementation method established based on software quantitative management process |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bhatia et al. | Critical factors to environment management in a closed loop supply chain | |
| Al Dallal | Object-oriented class maintainability prediction using internal quality attributes | |
| CN103514369B (en) | A kind of Regression Analysis System based on Active Learning and method | |
| CN108877223A (en) | A kind of Short-time Traffic Flow Forecasting Methods based on temporal correlation | |
| CN104835103A (en) | Mobile network health evaluation method based on neural network and fuzzy comprehensive evaluation | |
| CN110738564A (en) | Post-loan risk assessment method and device and storage medium | |
| CN106600464A (en) | Power grid construction new project cost rapid estimation method based on fuzzy mathematics | |
| CN101826090A (en) | WEB public opinion trend forecasting method based on optimal model | |
| CN104699614B (en) | Software defect component predicting method | |
| CN104809066A (en) | Method for evaluating and predicting maintenance work load of open source software (OSS) based on code quality | |
| Ungermann et al. | Data analytics for manufacturing systems–a data-driven approach for process optimization | |
| CN105975797A (en) | Product early-fault root cause recognition method based on fuzzy data processing | |
| CN104021180A (en) | Combined software defect report classification method | |
| Kalair et al. | Dynamic and interpretable hazard-based models of traffic incident durations | |
| CN106779366A (en) | project management process performance model and its construction method and performance model management system | |
| CN104102716A (en) | Imbalance data predicting method based on cluster stratified sampling compensation logic regression | |
| Bhatia et al. | A simulation-based statistical method for planning modular construction manufacturing. | |
| CN110362911A (en) | A kind of agent model selection method of Design-Oriented process | |
| CN109544410A (en) | Cell source of houses value parameter estimation method and device | |
| CN104462215A (en) | Scientific and technical literature quoting number predicting method based on time sequence | |
| CN113255958B (en) | Urban rail transit passenger flow volume change interval prediction method for describing fluctuation characteristics | |
| Bardsiri et al. | Statistical analysis of the most popular software service effort estimation datasets | |
| KR101629178B1 (en) | Apparatus for technology life analysis using multiple patent indicators | |
| Huang et al. | Towards automated model calibration and validation in rail transit simulation | |
| Sidabutar et al. | Comparison of linear regression, neural net, and arima methods for sales prediction of instrumentation and control products in PT. Sarana instrument |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |