JP6382068B2 - Project progress management method and project progress management apparatus - Google Patents

Description

  The present invention relates to a project progress management method and a project progress management apparatus for managing the progress of a project.

  When project progress management is performed using work progress indicating the degree of progress, if the achievement level of the quality of the development product is not accurately reflected in the work progress, the work progress indicates a smooth value. However, there will be a situation in which certain quality and performance cannot be achieved for the development product. As a result, unexpected additional man-hours are required to improve the quality and performance of the developed deliverables, and it is impossible to accurately grasp the remaining man-hours required until the project is completed. Therefore, in order to appropriately manage the progress of the project, it is required that the degree of achievement of the quality of the developed product is accurately reflected in the work progress.

  As a related technique, a technique for performing quality control of development software using a reliability growth curve model is known (see, for example, Patent Document 1).

JP 2006-301870 A

  In the technique described in Patent Document 1, since a fixed reliability growth curve model is used, a divergence between the reliability growth curve and the entity occurs, and the divergence increases as the project progresses. As a result, it is impossible to accurately grasp the number of remaining work required until the project is completed.

  Therefore, the present invention provides a project progress management method capable of deriving an appropriate reliability growth curve model according to the degree of achievement of the quality of a development product in a project and calculating an accurate work remaining time, and a project for implementing the same An object is to provide a progress management device.

A typical example for solving the above problems is as follows. That is, the present invention includes an arithmetic processing unit,
A storage unit for storing information on the project, actual values and actual man-hours regarding the quality of the developed product in the project, and target values regarding the quality of the product;
Comprising
The arithmetic processing unit includes:
Obtain information of the specified target project from the storage unit,
Based on the information of the target project, identify a reliability growth curve model candidate group of development deliverables in the target project,
Obtain the actual value and actual man-hours related to the quality of the development product in the target project from the storage unit,
Based on the actual value related to the quality and the actual man-hours, the reliability growth curve model of the development product in the target project is estimated from the reliability growth curve model candidate group,
Obtaining a target value related to the quality of the development product in the target project from the storage unit;
Based on the target value for the quality and the reliability growth curve model estimated, calculate the number of work remaining in the development product of the target project,
Outputting the number of remaining working hours in association with the information of the target project,
It is characterized by that.

  According to the present invention, it is possible to derive an appropriate reliability growth curve model according to the degree of achievement of quality of a development product in a project, and to calculate an accurate work remaining time.

It is a figure which shows the example of whole structure of the project progress management system 1 which concerns on the Example of this invention. It is a figure which shows the structural example of the project information table 210 which concerns on the Example of this invention. It is a figure which shows the structural example of the quality management table 220 which concerns on the Example of this invention. It is a figure which shows the structural example of the man-hour management table 230 which concerns on the Example of this invention. It is a figure which shows the structural example of the project type management table 240 which concerns on the Example of this invention. It is a flowchart which shows the process which calculates the progress rate of a project which concerns on the Example of this invention, and the number of work remaining hours.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a diagram showing an example of the overall configuration of a project progress management system 1 according to an embodiment of the present invention. As shown in FIG. 1, the project progress management system 1 is a computer network system configured by connecting a project progress management device 100, a database server 200, and a development device 300 to a network 10.

  The network 10 can be realized in various modes as long as it can transmit data. For example, the network 10 can be configured to transmit data using electromagnetic waves, infrared rays, visible light, sound, electricity, electrons, or the like. In addition, the network 10 may be configured to transmit data using a storage medium included in an electronic device such as a reader.

  The project progress management device 100 is a device that manages the progress of a project. As shown in FIG. 1, the project progress management apparatus 100 performs overall control of the project progress management apparatus 100 itself, and also includes an arithmetic processing unit 101 that includes a CPU and a coprocessor that perform various determinations, calculations, and control processes. A storage unit 130, a display unit 140 including a display, and the like, and an input / output unit 150 connected to the network 10 and transmitting / receiving data to / from the database server 200 and the development apparatus 300 are provided. Each unit is connected by a communication path such as a BUS (bus).

  The reliability growth curve estimation unit 110 and the progress calculation unit 120 are realized by the arithmetic processing unit 101 executing a program stored in the storage unit 130. The program may be stored in the storage unit 130 from the outside via the input / output unit 150 or a removable storage medium when necessary.

  The database server 200 includes an arithmetic processing unit 101 composed of a CPU and a coprocessor, an input / output unit 150, and a project information database 202 which is a storage unit connected to these through a communication path such as BUS.

  The project information database 200 includes a project information table 210 (FIG. 2), a quality management table 220 (FIG. 3), a man-hour management table 230 (FIG. 4), and a project type management table 240 (FIG. 5).

  FIG. 2 is a diagram illustrating a configuration example of the project information table 210 according to the embodiment of the present invention. The project information table 210 stores information related to projects.

  As shown in FIG. 2, the project information table 210 includes, as a key, a project ID 211 that uniquely identifies a project, a type 212, a development scale 213, a quality level 214, and at least one work item 215. It is a table stored in association with each other.

  A type 212 indicates a type of a development system that is a development product of the project, and includes a financial system, a public system, and an embedded system. The development scale 213 is information indicating the scale of the development system, and is indicated by the number of development functions, the number of screens, function points, the number of steps of the source code, and the like. The quality level 214 is information indicating the degree of quality required for the development system. The work item 215 indicates the work contents to be executed in the project, and includes requirement definition, basic design, detailed design, coding, unit test, integration test, comprehensive test, and the like. The work content may be assigned WBS (Work Breakdown Structure) which is divided more finely.

  FIG. 3 is a diagram showing a configuration example of the quality management table 220 according to the embodiment of the present invention. The quality management table 220 stores information related to the quality of development products in projects such as development systems and development software.

  As shown in FIG. 3, the quality management table 220 is a table that stores the target defect number 222 and the time-series data 223 of the actual defect number in association with each other using the work item 221 as a key.

  The work item 221 indicates a work item in the project, and includes basic design and detailed design. The target defect number 222 is a predicted value of the total number of defects mixed in a development product such as a document or source code created by the work item 221. The target defect number 222 may be stored in the quality management table 220 at the start of the project, or may be updated while the project is in progress. For example, when the basic design is completed, the target defect count of the detailed design product may be determined in consideration of the quality of the basic design product. The reference table name is stored in the time series data 223 of the number of actual defects. In this example, the work item 221 “PJ001-basic design” is stored in association with “100” as the target defect number 222 and table A224 as time series data 223 of the actual defect number. The number of actual defects is the total number of defects that have been found or removed by a certain point in time.

  The table A224 stores the date 224-1 and the number of actual defects 224-2 corresponding to the date. For example, the actual defect number of “2014/7/1” is “0”, and the actual defect number of “2014/7/2” is “1”.

  FIG. 4 is a diagram illustrating a configuration example of the man-hour management table 230 according to the embodiment of the present invention. The man-hour management table 230 stores information related to man-hours spent on project work items.

  As shown in FIG. 4, the man-hour management table 230 is a table that stores the actual man-hours 234 in association with each other by using the combination of the work item 231, the worker 232, and the date 233 as a key.

  The work item 231 indicates a work item in the project, and includes basic design and detailed design. The worker 232 indicates a member engaged in the project, and in this example, the member is uniquely specified by an ID such as “WK101”.

  The date 224-1 (FIG. 3) and the date 233 (FIG. 4) may be the date and time including the time, or any date as long as the order of time is determined, such as the first review and the second review.

  FIG. 5 is a diagram showing a configuration example of the project type management table 240 according to the embodiment of the present invention. Similar projects can be classified as types. The project type management table 240 stores information on project types.

  As shown in FIG. 5, the project type management table 240 associates the reliability growth curve model 242, the parameter range 243, the project ID 244, the average 245, and the covariance matrix 246 with the project type 241 as a key. It is the table memorized.

  The growth curve model 242 is a model showing the relationship between time and the number of defects, and includes a Gompertz curve y = Aexp {Bexp {−Cx}}, a logistic curve y = A / (1 + Bexp {−Cx}), and an exponential model y. = A {1-exp {-Cx}}. x represents time, work man-hours, costs dropped as work man-hours, and the like. y represents the number of defects found or removed. exp {} is an exponential function. A, B, and C are constants and are given as parameters. The parameter is also expressed as θ. That is, the reliability growth curve model candidate group is given by y = f (x | θ) together with the parameter θ, and the reliability growth curve model is uniquely determined by specifying the value of the parameter θ. In the Gompertz curve and logistic curve, θ = (A, B, C), and in the exponential model, θ = (A, C). The parameter range 243 is a range of values that the parameter θ can take. The project ID 244 is an ID that uniquely identifies a project that belongs to a project type. Multiple projects may belong to the same project type. The mean 245 and covariance matrix 246 are given as follows. Each section of the project type 212 (FIG. 2) is converted into a dummy variable. That is, “1” is assigned to the “financial system” if it is a financial system, and “0” is assigned otherwise. The same applies to “public system” and “embedded system”. Other project information such as the development scale 213 and each of the dummy variables are combined and expressed as a multidimensional vector. The average and covariance matrix for the multidimensional vector for all projects belonging to each project type 241 in the project management table 240 are the average 245 and the covariance matrix 246.

  The project type 241 “PJC001” takes a Gompertz curve as the reliability growth curve model 242, and the parameters of the curve are in the range of “0.2 <B <0.4, 0.5 <C <2”. Each project of ID244 “PJ002,..., PJ003” belongs to the project type. The average 245 for the project type 241 “PJC001” is μ and the covariance matrix 246 is S.

  Returning to FIG.

  The development apparatus 300 is an apparatus for an operator to implement the work content to be implemented in a project. Data generated or input accompanying the work, for example, the number of actual defects, actual man-hours, etc. are transmitted to the database server 200 via the network 10 and stored in the project information database 202.

  Information relating to the project, information relating to the quality of the development product such as the development system, information relating to the work man-hours, information relating to the project type, and the like are generated or input by the project progress management device 100 and the development device 300. The data may be transmitted to the database server 200 and stored in the project information database 202.

  Information about the target project and information stored in the project information database 202 may be stored in the storage unit 130 of the project progress management apparatus 100 as appropriate, or may be read from the project information database 202 when necessary. In this embodiment, it is assumed that the data is appropriately stored in the storage unit 130 of the project progress management apparatus 100.

  FIG. 6 is a flowchart showing a process for calculating the progress rate of the project and the number of remaining working hours according to the embodiment of the present invention.

  The progress rate of the project and the number of remaining work are calculated when the arithmetic processing unit 101 executes programs (reliability growth curve estimation unit 110 and progress calculation unit 120) stored in the storage unit 130.

  First, when receiving the work item for calculating the target project ID and the progress rate, the arithmetic processing unit 101 acquires information on the target project from the project information table 210 (step 301).

  Next, the arithmetic processing unit 101 specifies the project type 241 to which the target project belongs from the acquired information on the target project (step 302).

  Next, the arithmetic processing unit 101 refers to the project type management table 240 based on the identified project type 241 to acquire the corresponding reliability growth curve model 242 y = f (x | θ) and the parameter range 243, A reliability growth curve model group is specified (step 303).

  Next, the arithmetic processing unit 101 refers to the quality management table 220 and the man-hour management table 230, and acquires the actual defect number Y [i] and the actual man-hour W [i], which are indices of the actual results relating to the k pieces of quality. (Step 304). Here, i = 1, 2,..., K, and Y [1] <Y [2] <... <Y [k] is satisfied.

Next, the arithmetic processing unit 101 sets X [0] (θ) = 0, and for each i (i = 1, 2,..., K), X [i] (θ) = f ⁻¹ (Y [ i] | θ) is calculated, and expected man-hours X [i] (θ) −X [i−1] (θ) (i = 1, 2,..., k) are calculated (step 305).

  Next, the arithmetic processing unit 101 calculates a correlation r (θ) between the calculated expected man-hour X [i] (θ) −X [i−1] (θ) and the actual man-hour W [i], and r (θ) The value of θ that maximizes the value of θ) is specified, and that value is defined as θm. The arithmetic processing unit 101 calculates a reliability growth curve model y = f (x | θm) determined by substituting θm for the parameter θ (step 307).

  Next, the arithmetic processing unit 101 refers to the quality management table 220 and obtains a target defect number Y that is a target index regarding the quality of the development product in the target project (step 308).

Next, the arithmetic processing unit 101 calculates an expected total man-hour X = f− ¹ (Y | θm) using the reliability growth curve model y = f (x | θm), and sums the actual man-hours X [k ] Is calculated as X [k] = X [k] (θm) or W [1] + W [2] +... + W [k], and the progress rate pi = X [k] / X and the number of remaining work Xr = X−X [k] are calculated (step 309).

  Next, the arithmetic processing unit 101 outputs the calculated progress rate pi and the remaining work amount Xr to the display unit 140 in association with the project information (step 310).

In order to identify the project type in step 302, for example, the following may be performed. The arithmetic processing unit 101 adds the multidimensional vector z that is the first multidimensional vector of the target project, the average 245 that is the second multidimensional vector in each project type 241 of the project type management table 240, and the covariance matrix 246. Calculate Mahalanobis distance against When the average is μ and the covariance matrix is S, the Mahalanobis distance is sqrt {(z−μ) ^T S ⁻¹ (z−μ)}. However, T represents transposition of a vector, and sqrt {} represents a square root. The arithmetic processing unit 101 obtains the project type 241 having the smallest value as the similarity between the first multidimensional vector and the second multidimensional vector, and identifies this as the project type to which the target project belongs. To do.

  When there is a project type 241 to which the project ID of the target project belongs, the project type 241 may be specified as the project type to which the target project belongs.

  Further, after the project type 241 to which the target project belongs is specified, if the project ID of the target project is not registered in the project ID 244 belonging to the project type 241, the project ID of the target project may be registered. Good. Further, the average 245 and the covariance matrix 246 may be recalculated and updated.

  In order to obtain the actual defect number Y [i] and the actual man-hour W [i] in step 304, for example, the following may be performed. The arithmetic processing unit 101 refers to the quality management table 220 and acquires the time series data 223 of the number of actual defects corresponding to the work item. Next, the arithmetic processing unit 101 refers to the man-hour management table 230 and acquires time-series data of the date 233 and the actual man-hour 234 corresponding to the work item. Here, with respect to the oldest date in the time series data 223 of the actual defect number, the actual defect number corresponding to the date is Y [1], and the total actual man-hours up to the date is W [1]. . Next, the smallest one of the number of actual defects larger than Y [1] is defined as Y [2], and the total number of actual man-hours from the date corresponding to Y [1] to the date corresponding to Y [2] is W [2]. Similarly, Y [3], W [3], Y [4], W [4],..., Y [k], W [k] are determined in the same manner. However, Y [k] is the largest number of actual defects.

  Although the embodiments of the present invention have been described in detail above, according to such embodiments, the reliability growth curve model corresponding to the project type and the range of parameters associated with the reliability growth curve model are specified, and the performance defect Since the reliability growth curve model is estimated using the number and the actual man-hours, the achievement degree such as quality can be reflected more accurately in the work progress rate indicating the progress degree of the project.

  In addition, according to the above embodiment, since an appropriate reliability growth curve model corresponding to the degree of achievement of the quality of the development product in the project can be derived, an accurate work remaining time can be calculated.

  In addition, this invention is not limited to the said Example, A various change is possible in the range which does not deviate from the summary.

  For example, in the above embodiment, the progress rate is calculated using the number of defects as a scale indicating quality, that is, an index, but the progress rate that can be calculated is not limited to quality. The progress rate may be calculated using a scale indicating the degree of achievement of the project deliverables.

  In addition to the quality of the development product described in the above embodiment, the progress rate may be calculated for non-functional requirements such as the performance of the development product such as a development system and the amount of resources. As a result, an appropriate reliability growth curve model can be derived according to the performance of the development product in the project and the degree of achievement of the resource amount, and therefore, an accurate work remaining time can be calculated. For example, in terms of performance, the number of processes per unit time can be used instead of the number of defects. The progress rate calculation process may be performed by assuming that the target defect number 222 is the target processing number and the actual defect number 224-2 is the performance achieved at that time. If it is a resource amount, the used resource amount can be used. Since resource saving is generally superior, the reciprocal of the amount of resources used may be used to reverse the size.

  In addition to the above, non-functional requirements include accuracy such as calculation accuracy and time synchronization, interoperability, security strength, response time, fault tolerance, recovery time from failure, and maintainability. Measures indicating interoperability may be qualitative, such as the applicability of existing systems and applications, but the number of applications that can be applied and milestones in application of existing systems may be used as counts or order measures. . In addition, when the existing application can be applied but the required processing performance is not satisfied, the application non-application is set to “0”, and the processing performance is completely satisfied “1”. A scale may be assigned as the achievement rate of the processing performance when there is not. That is, a quantitative value or the like may be inserted between any two values of the count value or the order scale. By doing so, the scale can be set finely.

  You may apply to said WBS etc. besides a non-functional requirement. The degree of achievement of each WBS sub-work item in the target project is given as the progress rate, the sub-work item is weighted in consideration of the man-hours required for each sub-work item, and the progress rate is added by the weight May be used as a scale. By doing so, the progress rate of the entire project can be calculated.

  In addition to the progress rate pi and the work remaining time Xr, a prediction completion date and a prediction cost may be calculated, and these may be output in association with information of the target project. The predicted completion date is a date obtained by adding Xr / N to the date corresponding to Y [k], where N is the number of workers mobilized to the work item. The predicted cost is a cost obtained by adding Xr * Co to the cost dropped by the date corresponding to Y [k], where Co is the cost per unit man-hour.

  In addition, a value calculated for the target project in any step shown in FIG. 6 may be held and reused when the target project is designated again. By doing so, processing time can be shortened.

  Further, the project progress management device 100 and the development device 300 may be realized by a single computer.

  Further, the program described in the above embodiment may be realized by hardware by setting the integrated circuit (LSI), for example.

DESCRIPTION OF SYMBOLS 10 ... Network, 100 ... Project progress management apparatus, 200 ... Database server, 300 ... Development apparatus.

Claims (10)

A project progress management method in a project progress management device,
Get the information of the specified target project from the project information stored in the storage unit,
Based on the information of the target project, identify a reliability growth curve model candidate group of development deliverables in the target project,
Acquired actual values and actual man-hours related to the quality of the developed deliverables in the target project, stored in the storage unit,
Based on the actual value related to the quality and the actual man-hours, the reliability growth curve model of the development product in the target project is estimated from the reliability growth curve model candidate group,
Obtaining a target value related to the quality of the development product in the target project, stored in the storage unit,
Calculate the expected total man-hours using the estimated reliability growth curve model and the target value,
Calculate the total actual man-hours from the actual man-hours related to the quality of the developed deliverables in the target project,
Calculates the expected by Ri work Gozan man-hours to reduce the total actual man from the total man-hours,
And outputs the work remaining units in association with the target project information,
Project progress management method characterized by this. Converting the target project information into a first multidimensional vector;
Using the second multidimensional vector for the reliability growth curve model candidate group stored in the storage unit and a covariance matrix corresponding to the second multidimensional vector, the first multidimensional vector and the Calculate the similarity to the second multidimensional vector,
Specifying the reliability growth curve model candidate group based on the similarity;
The project progress management method according to claim 1, wherein: Based on the estimated reliability growth curve model, the actual man-hours and target values for the quality of the development product in the target project, calculate the work progress in the development product of the target project,
Outputting the work progress in association with the information of the target project,
3. The project progress management method according to claim 1, wherein the project progress is managed. The actual value related to the quality is the number of defects found or removed among the defects mixed in the development product in the target project,
The target value related to the quality is a predicted value of the total number of defects mixed in the development product in the target project.
The project progress management method according to claim 3, wherein: Instead of the number of defects, the performance of the development product in the target project, the resource amount of the development product in the target project, or a scale indicating the achievement level of the product in the target project, or included in the target project Use a scale that indicates the degree of achievement of
The project progress management method according to claim 4, wherein: An arithmetic processing unit;
A storage unit for storing information on the project, actual values and actual man-hours regarding the quality of the developed product in the project, and target values regarding the quality of the product;
Comprising
The arithmetic processing unit includes:
Obtain information of the specified target project from the storage unit,
Based on the information of the target project, identify a reliability growth curve model candidate group of development deliverables in the target project,
Obtain the actual value and actual man-hours related to the quality of the development product in the target project from the storage unit,
Based on the actual value related to the quality and the actual man-hours, the reliability growth curve model of the development product in the target project is estimated from the reliability growth curve model candidate group,
Obtaining a target value related to the quality of the development product in the target project from the storage unit;
Calculate the expected total man-hours using the estimated reliability growth curve model and the target value,
Calculate the total actual man-hours from the actual man-hours related to the quality of the developed deliverables in the target project,
Calculates the expected by Ri work Gozan man-hours to reduce the total actual man from the total man-hours,
And outputs the work remaining units in association with the target project information,
Project progress management device characterized by that. The storage unit further stores information on a multidimensional vector for the reliability growth curve model and a covariance matrix for the multidimensional vector,
The arithmetic processing unit includes:
Converting the target project information into a first multidimensional vector;
Using the second multidimensional vector and the covariance matrix corresponding to the second multidimensional vector for the reliability growth curve model candidate group from the storage unit, the first multidimensional vector and the second multidimensional vector Calculate the similarity with the dimension vector,
Specifying the reliability growth curve model candidate group based on the similarity;
The project progress management apparatus according to claim 6, wherein: The arithmetic processing unit includes:
Based on the estimated reliability growth curve model, actual man-hours and target values related to the quality of the development product in the target project, calculate the work progress in the development product of the target project,
Outputting the work progress in association with the information of the target project,
The project progress management apparatus according to claim 6 or 7, characterized in that The actual value related to the quality is the number of defects found or removed among the defects mixed in the development product in the target project,
The target value related to the quality is a predicted value of the total number of defects mixed in the development product in the target project.
The project progress management apparatus according to claim 8, wherein: Instead of the number of defects, the performance of the development product in the target project, the resource amount of the development product in the target project, or a scale indicating the achievement level of the product in the target project, or included in the target project Use a scale that indicates the degree of achievement of
The project progress management device according to claim 9, wherein the device is a project progress management device.

Images