JP6297919B2 - Method for supporting work planning in software development, and work planning support device - Google Patents

Description

  The present invention relates to a method for supporting work planning in software development, and a work plan planning support apparatus.

  In Patent Document 1, in order to prevent over-detection of clone code in clone code detection processing and source code evaluation between files and to evaluate and analyze clone code efficiently and correctly, a predetermined code in a clone code detection target file is disclosed. Registers section information for each unit data in the detection table, registers section information for each predetermined unit data in the file to be compared with the clone code detection target file in the detection table, and stores the predetermined unit in the detection table. If the section information between the two data satisfies the predetermined condition, the corresponding section information in the detection table is combined for each file, the section information in the detection table is extracted, registered in the clone code table, and registered in the clone code table The degree of influence on file quality from the recorded section information It has been described.

  Patent Document 2 relates to a source code management system for managing source code of software under development, manages the source code of software under development in the repository, registers only the common source code in the repository, and the source code The common code implementation method is presented, the code clone included in the source code is detected, it is determined whether the source code is standardized, and the source code that the user standardized is obtained from the metrics. It is described that evaluation is performed using the calculated improvement rate of the source code and the best common implementation result is selected.

JP 2011-48639 A JP 2009-86814 A

  In general, in software development, it is not easy to create an appropriate work plan so that work can be completed within man-hours that can be input and the effect is maximized. One of the reasons is the existence of work restrictions. As the above-mentioned restriction, there is an exclusive relationship (competition relationship) in which different tasks compete for some reason and both cannot be selected at the same time. An exclusive relationship is a relationship in which a plurality of work policies exist and cannot be made compatible. The relationship that cannot be made compatible is, for example, the case where one plan cannot be implemented due to time or resource constraints when the other plan cannot be established if work is performed according to one plan. If an appropriate work plan is planned in consideration of the exclusive relationship between these tasks, it is necessary to calculate the effect after selecting a combination of tasks that can be performed at the same time. It is necessary to verify the effect while confirming the exclusive relationship each time, and it is necessary to perform complicated and enormous calculation.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a method for supporting work planning in software development and a work plan planning support apparatus that can efficiently and appropriately create a work plan.

  One aspect of the present invention for achieving the above object is a method for supporting the creation of a work plan in software development, in which an information processing apparatus uses a plurality of work relationships in software development as a constraint, A subset of the tasks is determined by solving a combinatorial optimization problem with the effect of the task group obtained by combining the tasks as an objective function.

  In addition, the subject which this application discloses, and its solution method are clarified by the column of the form for inventing, and drawing.

  According to the present invention, it is possible to efficiently and appropriately formulate a work plan related to software development.

1 is a diagram showing a schematic configuration of a work plan planning support apparatus 10. FIG. It is a data flow figure explaining work expected value determination processing S200. It is an example of the work information 251. It is an example of a work expected value policy 252. 4 is an example of work priority information 253; It is an example of work expected value information 255. It is an example of work competition information 254. 4 is an example of work subset information 256; It is a flowchart explaining work subset determination processing S900. It is a table | surface which shows the relationship of whether the combination of work | work and a constraint condition are satisfied. It is the figure which extracted a part of line of FIG. It is a figure which shows schematic structure of the work plan planning assistance apparatus 10 of 2nd Embodiment. It is a figure explaining the structure of object software. It is a data flow figure explaining code clone distribution analysis processing S1400. It is an example of the component element information 262. It is an example of a code clone detection result 264. It is an example of the code clone distribution information 265. It is a flowchart explaining work analysis processing S1800. It is a flowchart explaining a code clone distribution analysis process 1813. It is a data flow figure explaining work competition information generation processing S1815. It is an example of a table 2000. It is an example of work competition information (side list format) 254A. It is an example of work competition information (matrix format) 254B. It is a flowchart explaining work competition information generation processing S1815. It is a data flow figure explaining work expected value determination processing S2500. It is an example of a quantitative constraint policy 258. It is the table | surface which enumerated the relationship between a work subset, required time, and an expected value.

  Hereinafter, embodiments will be described with reference to the drawings.

First embodiment

  FIG. 1 shows a schematic configuration of a work plan planning support apparatus 10 that is an information processing apparatus (computer) described in the first embodiment. The work plan planning support apparatus 10 generates information for supporting the planning of a work plan in software development. The work planning support apparatus 10 solves a combinatorial optimization problem (combination planning problem) that uses a competitive relationship between a plurality of works as a constraint for an objective function based on a work expectation value that is an index representing work effects. Thus, a work group (hereinafter also referred to as a work subset) having the maximum objective function is determined from a plurality of work. In the determination of the work subset, the work planning support apparatus 10 includes information related to work (work information), work expectation value policy that defines work expectation values, work conflict information describing constraint conditions, and work priority. Information about the degree is given. In addition, the work planning support device 10 pays attention to the exclusive relationship between the work when determining the work subset, and applies the dynamic programming method, the branch and bound method, the genetic algorithm, etc. as needed to solve the problem once. The complexity of processing and the load are reduced by avoiding solving the partial problems that have been duplicated or not solving the unnecessary partial problems.

  As illustrated in FIG. 1, the work planning support device 10 includes a processor 11, a storage device 12, an input device 13, a display device 14, and a communication device 15. These are connected to be communicable via a communication means such as a bus.

  The processor 11 is configured using, for example, a CPU (Central Processing Unit) and an MPU (Micro Processing Unit). The processor 11 reads out and executes the program stored in the storage device 12, thereby realizing various functions of the work planning support device 10. The storage device 12 (including a main storage device, an auxiliary storage device, and an external storage device) is a device that stores programs and data. For example, a ROM (Read Only Memory), a RAM (Random Access Memory), and an NVRAM (Non Volatile). RAM), hard disk drive, SSD (Solid State Drive), optical storage device (CD-ROM, DVD-ROM, etc.).

  The input device 13 is a user interface that receives input of information and instructions from the user, and is, for example, a keyboard, a mouse, or a touch panel. The display device 14 is a user interface that provides information to the user, and is, for example, a liquid crystal monitor or an LCD (Liquid Crystal Display). The communication device 15 is a communication interface that communicates with the external device 2 via the communication network 5, and is, for example, a NIC (Network Interface Card). The storage device 12 stores, for example, programs and data downloaded from the external device 2 via the communication network 5.

  As shown in the figure, the storage device 12 stores an expected work value determination PG 211 and a work subset determination PG 212. In the following description, the functions realized by these programs are also referred to as a work expected value determination unit 271 and a work subset determination unit 272, respectively. The work expected value determination PG 211 and the work subset determination PG 212 may or may not operate under the OS (Operating System). The work expected value determination PG 211 and the work subset determination PG 212 may be executed on a virtual machine realized by virtualization software.

  As shown in the figure, the storage device 12 stores work information 251, work expected value policy 252, work priority information 253, work conflict information 254, work expected value information 255, and work subset information 256.

  FIG. 2 is a data flow diagram for explaining the flow of processing (hereinafter also referred to as expected work value determination processing S200) performed in the work planning support device 10. Hereinafter, the functions of the work planning support apparatus 10 will be described with reference to FIG.

  Based on the work information 251, the work expected value policy 252, and the work priority information 253, the work expected value determination unit 271 expects each work described in the work information 251 (hereinafter, also referred to as work expected value). And expected work value information 255 describing the expected work value of each work is generated.

  FIG. 3 shows an example of the work information 251. As shown in the figure, the work information 251 is composed of a plurality of records including items of a work ID 2511, a target file 2512, a target file number 2513, and a target step number 2514. Among these, the work ID 2511 is set with an identifier assigned to each work (hereinafter also referred to as work ID). In the target file 2512, an identifier (hereinafter also referred to as a file ID) of a file to be worked (hereinafter also referred to as a target file) is set. In the target file number 2513, the number of target files of the work is set. In the target step number 2514, the number of steps of the program described in the target file is set. In this embodiment, the case where the work target is a file is illustrated, but the work target may be another type such as a module.

FIG. 4 shows an example of the work expected value policy 252. This work expectation value policy 252 is an example in the case where the effect of work is judged from the viewpoint of how much of a product can be processed as a work target, and the amount of the product is determined by the number of steps in the source code. Etc. The work expected value policy 252 includes information for obtaining the work expected value of each work described in the work information 251. In the expected work value policy 252 shown in FIG. 10, “c” is a work ID, “F” is the number of files, “S” is the number of steps (for example, the number of steps per file), and “P” is the priority. It is stipulated that the value “value (c)” obtained from the following equation when a count that takes a corresponding value (hereinafter also referred to as priority count) is used as a function for obtaining the expected work value of each work Has been.
value (c) = P × F × S Equation 1

  The work expectation value policy 252 includes information indicating the correspondence between the priorities (ranks A to D) and the priority counts P (priority counts P corresponding to the priority ranks A to D in order of “3. 0 ”,“ 2.0 ”,“ 1.0 ”, and“ 0.5 ”are also described.

  FIG. 5 shows an example of the work priority information 253. As shown in the figure, the work priority information 253 includes information indicating the work priority 2532 set for each work (work ID 2531). The user or the like can appropriately reflect the priority of each work in the objective function by appropriately setting the priority of each work in the work priority information 253, and each work subset can be determined by the work planning support apparatus 10. The priority of work can be reflected.

FIG. 6 shows an example of the expected work value information 255. As shown in the figure, the expected work value information 255 describes the expected work value 2552 of each work (work ID 2551). Note that the expected work value information 255 shown in the figure is generated based on the work information 251 in FIG. 3, the expected work value policy 252 in FIG. 4, and the work priority information 253 in FIG. For example, for the work whose work ID is “1” in the work information 251 of FIG. 3, the work priority 2532 is identified as “B” from the work priority information 253 of FIG. 5, and the work priority value policy 252 of FIG. The priority coefficient P is specified as 2.0. From the work information 251 in FIG. 3, the number of files of the work whose work ID is “1” is “3” and the number of steps is “12”, so the work whose work ID is “1” from the work expected value policy 252. The expected working value of
value (c) = P × F × S = 2.0 × 3.0 × 12 = 72
In the expected work value information 255 of FIG. 6, the expected work value “72” is set for the work with the work ID “1”.

  Returning to FIG. 2, the work subset determination unit 272 performs combination optimization calculation based on the expected work value information 255 and the work conflict information 254 generated by the expected work value determination unit 271, and performs an objective function (for example, an operation function to be performed). A work subset that maximizes the sum of value (c) in the combination (= Σvalue (c))) is determined, and work subset information 256 describing the determined work subset is generated.

  FIG. 7 shows an example of the work conflict information 254. The work competition information 254 describes a combination of works (work IDs) having a competitive relationship. In the figure, for example, “(1, 2)” indicates that the work with the work ID “1” and the work with the work ID “2” are in a competitive relationship. For example, when the work is a code clone correction (removal) work, the work ID “1” is the code clone A correction work, and the work ID “2” is the code clone B correction work, The code clone A correction work and the code clone B correction work are in a competitive relationship (the code clone A and the code clone B are included in the same file (or module) and are shared at the same time. If it is difficult to do). Note that the mode of expression of competing work is not limited to that shown in FIG. For example, combinations of module identifiers that are difficult to verify simultaneously (competitive) or identifiers of documents that are difficult to verify simultaneously (competitive) (for example, chapter and section numbers, page numbers) Or a combination of these.

  FIG. 8 shows an example of the work subset information 256. The work subset information 256 describes information indicating the components (work) of the work subset. The work subset information 256 shown in the drawing indicates that a work having a work ID “2, 3, 6, 10, 13, 24, 27, 35” that is not in a competitive relationship is a work subset.

  FIG. 9 is a flowchart for explaining a process (hereinafter also referred to as a work subset determination process S900) performed by the work planning support apparatus 10 when determining a work subset.

  First, the work planning support apparatus 10 acquires work information 251, work expected value policy 252, work priority information 253, and work conflict information 254, and stores them in the storage device 12 (S911). Such information may be preset by a user or the like and input to the work planning support apparatus 10, or may be acquired from the external apparatus 2 by the work planning support apparatus 10 via the communication network 5. .

  Subsequently, the work planning support apparatus 10 determines the work expected value of each work in the work information 251 based on the work information 251, the work expected value policy 252, and the work priority information 253, and the work in which the result is described. Expected value information 255 is generated (S912).

  Subsequently, the work planning support apparatus 10 performs combination optimization calculation based on the work competition information 254 and the expected work value information 255 generated in S912 (S913). Details of this processing will be described later.

  Subsequently, the work planning support apparatus 10 determines work subset information that maximizes the objective function based on the work expected value of the work described in the work expected value information 255 (S914).

  Subsequently, the work planning support apparatus 10 displays the determined work subset information 256 on the display device 14 (S915).

  Next, the combination optimization calculation S913 and the work subset determination S914 in FIG. 9 will be described in detail. The combinatorial optimization calculation is an optimization of the method in the combinatorial planning problem.

The work plan planning support apparatus 10 performs the combination optimization calculation S913 based on the following definition, for example.
“Definition:
The whole set of tasks is abstracted as Tasks, and the effect (gain) obtained by each task c∈Tasks is abstracted as value (c). At this time, the following subset ChosenTasks is selected from the task set Tasks. Then, in the range that satisfies the constraints, the one that maximizes the sum (= Σvalue (c)) of value (c) in the subset ChosenTasks is extracted as a work subset.
maximize Σvalue (c)
c ∈ ChosenTasks⊆Tasks
However, ∀c1, c2 (c1 ≠ c2) ∈ChosenTasks; the work of c1 and c2 does not compete. Note that ∀x ∈ S; P (x) is a proposition P (x) for any element x of set S Is established. "

Here, this definition can be converted into the following equivalent calculation formula.
First, the operation of whether or not the work ci (i = 1, 2,..., N) is selected is expressed as follows.
ci∈ {0,1} (i = 1,2, ..., n) ... Equation 2

Further, from the expected work value information 255 (here, expected work value information 255 shown in FIG. 6 as an example), the objective function is
Σvalue (c) = 72 × c1 + 50 × c2 + 48 × c3 + 14 × c4 + 20 × c5 + 36 × c6 Equation 3
It becomes.

Here, “ci = 1” is selected when the work ci is selected, and “ci = 0” is selected when the work ci is not selected. For example, when the work 1 to 4 is selected, c1 to c4 are “1” and c5. c6 is “0” and Equation 3 is
Σvalue (c) = 72 * 1 + 50 * 1 + 48 * 1 + 14 * 1 + 20 * 0 + 36 * 0
It becomes. For example, when only operations 1 and 4 are selected, c1 and c4 are “1”, and others are “0”.
Σvalue (c) = 72 * 1 + 50 * 0 + 48 * 0 + 14 * 1 + 20 * 0 + 36 * 0 = 72 * 1 + 14 * 1
It becomes.

Among the above definitions, “the work of c1 and c2 does not compete for ∀c1, c2 (c1 ≠ c2) ∈ ChosenTasks” can be expressed as follows from Equation 2:
ci + cj ∈ {0,1} (i ≒ j) ・ ・ ・ Equation 5
Expression 5 means that both ci and cj are selected (ci = 1 and cj = 1), that is, ci + cj = 2 is excluded. For example, the work conflict information 254 shown in FIG. Can be expressed as:
c1 + c2∈ {0, 1}
c2 + c3∈ {0, 1}
c2 + c4∈ {0, 1}
c3 + c4∈ {0, 1}
c1 + c5∈ {0, 1}

  FIG. 10 is a diagram illustrating the operation of each of the six operations (hereinafter referred to as C1 to C6) having the operation IDs “1” to “6” in the operation expected value information 255 illustrated in FIG. Not}, that is, all combinations that select r (r = 1, 2,... 6) from six different ones. In the figure, the number “1” in the table indicates that the work (any of C1, C2,... C6) is selected, and “0” means that the work is not selected. For example, the No. 40 row is “(C1, C2, C3, C4, C5, C6) = (0, 0, 1, 1, 0, 1)”, but this is “C1 = 0 ( And C2 = 0 (not selected), C3 = 1 (selected), C4 = 1 (selected), C5 = 0 (not selected), and C6 = 1 (selected) ”.

  Here, when the work conflict information 254 is the content of FIG. 7, the two tasks c3 and c4 do not compete with “に 対 し c3, c4 ∈ ChosenTasks”, that is, “c3 + c4∈ {0, 1}”. However, in the case of No.40, because “c3 + c4 = 2”, this condition is not satisfied, that is, the combination of the No.40 row cannot be selected. “X” indicating that a combination cannot be selected is set. On the other hand, “O” is set in “Availability” for combinations where work does not compete. In the case of the figure, “O” is set in “Availability” of 22 rows, and the work planning support apparatus 10 determines a combination of operations corresponding to these rows as a work subset.

  FIG. 11 is an excerpt of a part of the rows in FIG. 10 and an “expected value” column is added. In the “expected value” column, the value of Σvalue (c) calculated according to Equation 3 is set.

  Of the rows where “Yes” is set in “Yes / No”, the row with the maximum Σvalue (c) (maximize Σvalue (c)) is the No. 31 row ((C1, C2, C3, C4, C5 , C6) = (1, 0, 0, 1, 0, 1), Σvalue (c) = 158), and the work planning support apparatus 10 performs the work subset {c1, c4, c6} in S914 of FIG. Is generated.

  In order to improve the optimization accuracy in consideration of the exclusive relationship (competitive relationship) between tasks in the general combination planning problem method, a combination of tasks that can be performed simultaneously is selected appropriately. Therefore, it is necessary to check the exclusive relationship every time an operation is added or canceled, and it is necessary to perform a complicated and enormous calculation. For example, if the number of code clones is 1000, the number of all combinations of {correct and not correct} is 2 ^ 1000, making it difficult to complete the search within a realistic time. .

  Therefore, when the number of combinations is enormous, for example, dynamic programming, branch and bound methods, genetic algorithms, and the like may be applied to reduce the complexity of processing and the load. In this case, an exclusive relationship between operations is used as a method for omitting the search for the branched subproblem, so that the subproblems solved once are not solved redundantly or unnecessary subproblems are not solved. For example, among the 64 combinations shown in FIG. 10, the processing can be omitted for 42 combinations in which “x” is given to “possibility”.

As an example, a case where the branch and bound method is applied will be described. The branch and bound method is a search method that ignores a partial problem (performs a limited operation) when it is determined that the partial problem cannot be executed or an optimal solution cannot be obtained. Specifically:
(1) The total expected value is evaluated by dividing into a subset of pairs including and not including work ci.
(2) A subset that obtains the maximum evaluation value is adopted.

In (1), branching and evaluation are repeated as in the following (1-a) and (1-b).
(1-a) Find the pair with the largest expected value among the combinations including the work ci.
→ Hereafter, the files to be recursively changed sequentially and divided into “including combinations” and “not including combinations” and evaluated in the same manner.
(1-b) Find the pair with the largest expected value among the combinations not including the work ci.
→ Hereinafter, the files to be recursively changed sequentially, and divided into “combination including work ci” and “combination not including work ci” and evaluated in the same manner.

  In the branch and bound method, unnecessary problems are not solved by solving partial problems that have been solved once or unnecessary partial problems, thereby reducing the amount of calculation. Specifically, in the evaluation of the above (1-a) and (1-b), an exclusive relationship between operations is used as a constraint. For example, there is a competitive relationship (exclusive relationship) between the correction work of code clone A and the correction work of code clone B, and if the work for correcting code clone A is selected, the work for correcting code clone B cannot be selected. For example, the combination of the work of correcting the code clone A and the work of correcting the code clone B is excluded from the search target.

  As described above, in the present embodiment, the work selected so that the objective function is maximized from a plurality of works under the constraint based on the exclusive relationship between the works is determined as a work subset. Therefore, it is possible to provide useful information to the user or the like when making a work plan related to software development. Therefore, a work plan can be made efficiently and appropriately in software development. In addition, paying attention to the exclusive relationship between tasks, the dynamic programming method, the branch and bound method, the genetic algorithm, etc. are applied as needed, so that the complexity of processing and the load can be reduced.

  In the above description, a work group having the maximum objective function is determined as a work subset from a plurality of works. For example, a work group whose objective function exceeds a predetermined threshold is determined as a work subset. Also good.

Second embodiment

  In the first embodiment, the work information 251, work priority information 253, and work conflict information 254 are given to the work planning support apparatus 10 from the outside. In the second embodiment, these are provided by the work planning support apparatus 10. Generate. In the second embodiment, work related to code clone removal is taken up as work in software development. For example, when a large-scale source code is input to a known code clone analysis tool, thousands to tens of thousands of code clones may be output. However, the number of man-hours required for correction work is limited. It is necessary to determine (select) which code clone is to be modified. In the second embodiment, a configuration in which the work planning support device 10 generates information for optimizing the code clone correction work (maximizing the effect under constraint conditions) will be described as an example. Hereinafter, the description will focus on differences from the first embodiment.

  FIG. 12 shows a schematic configuration of the work plan planning support apparatus 10 of the second embodiment. The work plan planning support apparatus 10 of the second embodiment has the same configuration as the work plan planning support apparatus 10 of the first embodiment. The work planning support device 10 of the second embodiment includes a work analysis PG 220 (code clone detection PG 221, code clone distribution analysis PG 222, code clone modification priority determination PG 223, and work conflict information generation PG 224) in the storage device 12. .), The target software 261, the component element information 262, the modification priority policy 263, the code clone detection result 264, and the code clone distribution information 265 are stored and configured with the work plan planning support apparatus 10 of the first embodiment. Is different.

  In the following description, the functions realized by the work analysis PG 220 include the work analysis unit 280, the code clone detection PG 221, the code clone distribution analysis PG 222, the code clone modification priority determination PG 223, and the work conflict information generation PG 224. The functions realized by each of them are also referred to as a code clone detection unit 281, a code clone distribution analysis unit 282, a code clone modification priority determination unit 283, and a work competition information generation unit 284 in order.

In the second embodiment, the “work does not compete” situation in the combination planning problem in the first embodiment is expressed as follows.
“Clones is the complete set of work related to code clone modification (such as standardization of code clones) as if each work is associated with one of the code clones to be modified. A set of files containing each code clone in ∈ Clones is Files (c).
に 対 し Files (c1) ∩ Files (c2) = φ for c1, c2 ∈ ChosenTasks ''
In the above expression, for example, code clone A and code clone B exist in the same module (file), and by modifying code clone A, the contents of the file are changed and code clone B of the file is lost. It is assumed that the code clone B cannot be corrected due to reasons such as. For example, the code clone A includes the 10th to 20th lines of the source code (file), and the code clone B includes the 15th to 30th lines of the same source code (file). The work conflict information 254 in the second embodiment cannot select the code clone correction work related to the module (file) selected once based on such an assumption (the two correction works conflict). Are expressed as a pair of code clone identifiers (hereinafter also referred to as clone IDs).

  FIG. 13 is a diagram illustrating a configuration of software (hereinafter, also referred to as target software 261) that is a target of work analysis performed by the work analysis unit 280, which is used in the following description. The target software 261 includes a plurality of components (modules). Here, a case where the component of the target software 261 is a module will be described as an example. However, the component may be, for example, a file or a function.

  As shown in the figure, the target software 261 includes modules m1, m2,. The target software 261 includes a plurality of code clones (five code clones having clone IDs “1”, “2”, “3”, “4”, and “5”, respectively). For example, in the module m1, there are two code clones with clone IDs “1” and “2”, respectively.

  FIG. 14 is a data flow diagram for explaining the flow of processing (hereinafter also referred to as code clone distribution analysis processing S1400) performed by the work plan planning support apparatus 10 when performing code clone distribution analysis.

  The code clone detection unit 281 detects a code clone in the source code of the target software 261 by executing a known code clone detection tool (for example, “CCFinderX”) on the target software 261, for example.

  The code clone distribution analysis unit 282 performs code clone distribution analysis in the target software 261 based on the component information 262 of the target software 261 and the code clone detection result 264, and the code clone distribution in the target software 261 is described. Code clone distribution information 265 is generated.

  FIG. 15 shows an example of the component element information 262. As shown in the figure, the component element information 262 describes information indicating the correspondence between a file (file name) and a module (module name).

  FIG. 16 shows an example of the code clone detection result 264. As shown in the figure, the code clone detection result 264 describes information indicating the correspondence between the code clone (clone ID) detected by the code clone detection tool and the file (file name) including the code clone. Yes.

  FIG. 17 shows an example of the code clone distribution information 265. As shown in the figure, the code clone distribution information 265 includes information indicating the correspondence between the code clone (clone ID) and the module (module name).

  Returning to FIG. 14, the code clone modification priority determining unit 283 determines the modification priority of the code clone based on the code clone distribution information 265 and the modification priority policy 263, and the work priority information describing the determined modification priority. 253 is generated. Details of the modification priority policy 263 and the work priority information 253 will be described later.

  The work conflict information generation unit 284 generates work conflict information 254 (work conflict information (side list format) 254A and work conflict information (matrix format) 254B) based on the code clone distribution information 265 and the component information 262.

  As described above, in the second embodiment, the work planning support device 10 automatically generates the work information 251, the work priority information 253, and the work conflict information 254, so that the user or the like can support the work planning from the outside. The information for supporting the planning of the work plan can be obtained without giving such information to the apparatus 10.

  FIG. 18 illustrates a process (hereinafter also referred to as a work conflict information generation process S1800) performed when the work analysis unit 280 of the work plan planning support apparatus 10 analyzes the target software 261 and generates the work conflict information 254. It is a flowchart to do.

  First, the work planning support apparatus 10 acquires the target software 261 and the component information 262 and stores them in the storage device 12 (S1811).

  Subsequently, the work planning support device 10 executes a code clone detection tool, detects a code clone included in the target software 261, and generates a code clone detection result 264 (S1812).

  Subsequently, the work planning support device 10 analyzes the code clone distribution based on the component information 262 and the code clone detection result 264 to generate the code clone distribution information 265 (S1813). Details of this processing (hereinafter also referred to as code clone distribution analysis processing S1813) will be described later.

  Subsequently, the work planning support apparatus 10 determines the modification priority of each code clone included in the code clone distribution information 265 according to the modification priority policy 263 (S1814).

  Subsequently, the work planning support apparatus 10 generates work competition information 254 based on the component information 262 and the code clone distribution information 265 (S1815). This process (hereinafter also referred to as work competition information generation process S1815) will be described later.

  Subsequently, the work planning support apparatus 10 displays the work conflict information 254 generated in S1815 on the display device 14 (S1816).

  FIG. 19 is a flowchart for explaining the details of the code clone distribution analysis processing S1813 of FIG.

  First, the work planning support apparatus 10 sets 1 to a variable N (S18131).

  Next, the work planning support apparatus 10 extracts the contents of the Nth row of the code clone detection result 264, and substitutes the value of the clone ID column into the variable CID and the value of the file name column into the variable FILE (S18132).

  Next, the work planning support apparatus 10 searches the component information 262 for a line whose file name column matches the value of the variable FILE, and substitutes the value of the module name column of the found line into the variable MODULE ( S18133).

  Next, the work planning support apparatus 10 inputs the value of the variable CID in the clone ID column of the Nth row of the code clone distribution information 265 and the value of the variable MODULE in the module name column (S18134).

  Next, the work planning support apparatus 10 determines whether the Nth row is the last row of the code clone detection result 264 (S18135). If the Nth row is the last row of the code clone detection result 264 (S18135: YES), the work planning support apparatus 10 ends the code clone distribution analysis process S1813. If the Nth line is not the last line of the code clone detection result 264 (S18135: NO), the work plan planning support apparatus 10 sets N = N + 1 (S18136), and then the process returns to S18132.

  FIG. 20 is a data flow diagram illustrating the work conflict information generation process S1815 of FIG. As shown in the figure, the work conflict information generation unit 284 includes a table generation unit 2841, a conflict determination unit 2842, and a matrix element input unit 2843.

  Based on the code clone distribution information 265, the table generation unit 2841 generates a table 2000 having code clone clone IDs as rows and columns.

  FIG. 21 shows an example of the table 2000. As shown in the figure, the rows and columns of the table 2000 list the clone IDs (in this example, “1, 2, 3, 4, 5”) of the code clones detected by the code clone detection process. Yes.

  Returning to FIG. 20, the competition determination unit 2842 determines whether or not corrections compete for a combination of code clones based on the code clone distribution information 265, and outputs the determined result as work competition information (side list format) 254 </ b> A. . On the other hand, the matrix element input unit 2843 inputs data to the table 2000 generated by the table generation unit 2841 based on the work conflict information (side list format) 254A, and generates work conflict information (matrix format) 254B.

  FIG. 22 shows an example of work conflict information (side list format) 254A. In the figure, for example, “(1, 2)” indicates that there is a competitive relationship between the modification of the code clone with the clone ID “1” and the modification of the code clone with the clone ID “2”.

  FIG. 23 shows an example of work conflict information (matrix format) 254B. The work conflict information (matrix format) 254B represents the competitive relationship of correction work between code clones in a matrix format. If there is a competitive relationship between the code clones, “x” is entered in the corresponding cell. In the figure, for example, “x” is input to a cell in one row and two columns. This indicates that there is a competitive relationship between the code clone with the clone ID “1” and the code clone with the clone ID “2”.

  FIG. 24 is a flowchart for explaining details of the work conflict information generation processing S1815 in FIG. Hereinafter, it will be described with reference to FIG.

  First, the work planning support apparatus 10 generates a table 2000 based on the code clone distribution information 265 (S18151).

  Subsequently, the work planning support apparatus 10 determines whether or not correction work competes with each set of code clones based on the code clone distribution information 265, and outputs the result as work conflict information 254 (S18152). ). For example, the code clone distribution information 265 shown in FIG. 17 includes two rows with a module name “m1” (a row with a clone ID “1” and a row with a clone ID “2”). In this case, the work planning support apparatus 10 determines that the code clone set (1, 2) is “conflicting”.

  Subsequently, the work planning support apparatus 10 refers to the work conflict information 254 and inputs “x” to a corresponding location in the table 2000. When the input to the table 2000 is completed for all the contents of the work conflict information 254, the work plan planning support apparatus 10 outputs the table 2000 as work conflict information (matrix) 254B (S18153).

  Next, a process in which the code clone priority determination unit 283 of the work plan planning support apparatus 10 generates the work priority information 253 based on the modification priority policy 263 and the code clone distribution information 265 will be described with reference to FIG.

  The work planning support apparatus 10 determines the priority of the correction work based on, for example, the size of the code clone, the ease of correction, and whether or not to cross the team charge range. For example, since the code clone having a larger scale has a larger improvement effect due to the correction, the work planning support apparatus 10 sets the priority of the correction work higher for the code clone having a larger scale. Further, for example, since a code clone that can be easily corrected can be corrected with a smaller number of man-hours, the work planning support apparatus 10 sets a higher priority for the correction operation for a code clone that can be easily corrected. Also, for example, code clones that straddle the team's area of responsibility cause maintenance difficulties, so the work planning support apparatus 10 sets higher priority for such code clones and correction work.

  In the modification priority policy 263 shown in FIG. 14, information serving as a guideline for determining such priority is set. For example, in the case of a guideline that “a code clone with a larger scale has a higher priority”, for example, “priority A for code clones of 30 lines or more, priority B for code clones of 10 lines or more and 29 lines or less, 9 lines” The content of “priority C for the following code clone” is set in the modification priority policy 263.

  The priority may be determined by a plurality of guidelines. For example, the priority may be determined based on two guidelines of code clone size and ease of correction. As an example, the first priorities A, B, and C are determined based on the size of the code clone, and then the second priorities a, b, and c are determined based on the ease of correction, and these first and second priorities are determined. Combination of degrees ((A, a), (A, b), (A, c), (B, a), (B, b), (B, c), (C, a), (C, b ), (C, c), etc.). As a priority, for example, a method of setting a priority lexicographically for each combination, a high (or low) priority is set when both the first and second priorities are high (or low), and other cases For example, a method of determining that the composite priorities are equal can be considered. The code clone modification priority determining unit 283 determines the modification priority of each code clone included in the code clone distribution information 265 according to the modification priority policy 263 set in this way.

Third embodiment

  In the first embodiment and the second embodiment, the work plan planning support apparatus 10 imposes work conflict information 254 (competitive relationship between a plurality of tasks) as a constraint condition when determining a work subset. In the third embodiment, a condition (time constraint or the like) related to the input amount of work is further imposed as a constraint condition. For example, in agile software development, planning, requirements analysis, design, implementation (coding), testing, and documentation with a predetermined short period (for example, 1 to 4 weeks) for the purpose of minimizing risk. Such operations occur repeatedly. In such a development method, it is desirable that the same amount of work (work time or man-hours) is distributed within a cyclic repetition time unit (iteration, etc.) when making a work plan. Therefore, the work planning support apparatus 10 of the third embodiment selects a work subset so that the maximum effect can be obtained in a predetermined time.

Here, when the work selection problem of the first embodiment is redefined by imposing a condition ([constraint condition 2]) regarding the work input amount, it is as follows.
“Definition:
maximize Σvalue (c)
c ∈ ChosenTasks⊆Tasks
However,
c1 and c2 (c1 ≠ c2) ∈ChosenTasks does not compete with c1 and c2
... [Constraint 1]
Σcost (c) ≤ available man-hours and work time
... [Constraint 2]

Further, the problem of selecting a correction work based on the detection result of the code clone of the second embodiment is redefined by imposing a condition ([constraint condition 2]) on the input amount of work.
“Definition:
maximize Σvalue (c)
c ∈ ChosenClones⊆Clones
However,
Files (c1) ∩Files (c2) = φ for ∀c1, c2 ∈ ChosenClones
... [Constraint 1]
Σcost (c) ≤ available man-hours and work time
... [Constraint 2]

  In each definition above, Clones is the entire set of clones output by the clone analysis tool, Files (c) is the set of files belonging to the clone c∈Clones, and the value (gain) of the common software obtained by sharing them ) Is value (c), and its cost (man-hours for sharing, etc.) is cost (c).

  FIG. 25 is a data flow diagram for explaining the outline of the process (hereinafter also referred to as an expected work value determination process S2500) performed by the work plan planning support apparatus 10 of the third embodiment. The difference from the expected work value determination process S200 of FIG. 2 is that the work subset determination unit 272 of the work planning support apparatus 10 generates the work subset information 256 as the [constraint condition 2] and the quantitative constraint policy 258. It is a point to refer to.

  FIG. 26 shows examples of quantitative constraint policies (quantitative constraint policies 258A and 258B). In the quantitative constraint policy 258A, the input resource amount upper limit man-hour 2581 (for example, designation of an iteration period) is designated as the [constraint condition 2]. On the other hand, the quantitative constraint policy 258B has a form in which a quantitative constraint policy is designated by categorizing into a plurality of stages based on criteria such as ease of work (correction).

  Hereinafter, the function of the work plan planning support apparatus 10 according to the third embodiment will be described by taking as an example the case where [constraint condition 2] is imposed in the configuration of the work plan planning support apparatus 10 according to the second embodiment. In the second embodiment, the case where the work is a correction work of a code clone included in the source code of the target software 261 has been described as an example. However, in the third embodiment, the work is a test of the target software 261. A case will be described as an example. The basic functions of the work plan planning support apparatus 10 of the third embodiment are the same as the functions of the work plan planning support apparatus 10 of the second embodiment when the clone ID is read as a test item. In the third embodiment, the work competition information 254 in FIG. 7 indicates the competition relationship between test items. For example, in (1, 2) in FIG. 7, the test item 1 and the test item 2 are in an exclusive relationship. It shows that. In the third embodiment, the combination optimization calculation process S913 in FIG. 9 includes a process related to [Restriction condition 2]. Hereinafter, a description will be given centering on differences from the work planning support device 10 of the second embodiment.

  FIG. 27 is obtained by adding a column of “required time” and a column of “total availability” to the table of FIG. In the column “required time”, “2” is added to the total (Σcost (c)) if the total (Σcost (c)) of cost (c) of the quantitative constraint policy 258B in FIG. ) Is less than 50, “1” is set. [Constrained condition 1] success / failure (“O” when satisfied, “×” when not satisfied) is displayed in the “Availability” column, and [Restriction condition 1] and [Restriction condition 2] are displayed in the “Overall availability” field. ] Is established (“○” if both are established, “x” otherwise). FIG. 27 shows the case where “the upper limit of the required time is 3 or less” is imposed as [Constraint 2].

  “Maximize Σvalue (c)” in the above-described definition is an “expected value” of a row whose overall availability is “◯”. In this example, “108” that is the “expected value” of No. 12 is maximized Σvalue (c), and the work planning support apparatus 10 sets {c1, c6} as the work subset information 256 in S914 in FIG. decide. In addition, when the branch and bound method is applied in the combination optimization calculation process S913 of FIG. 9, for example, in addition to [Constraint 1], the upper limit of Σcost (c) related to [Constraint 2] (for example, the total allowable man-hours) ) The following branches will be excluded.

By the way, as the work subset information 256, the work planning support apparatus 10 generates and outputs not only the optimal solution (the solution that maximizes the objective function) but also the suboptimal work subset and approximate solution. May be. For example, as working subset information 256,
(I) A working subset that obtains maximized Σvalue (c) in a state where the upper limit of the required time is increased from “3” to “4” (that is, a state in which the constraints are slightly relaxed), and maximized Σvalue ( c) variation,
(Ii) When the length of the iteration period is specified as the upper limit of the required time, the time corresponding to a multiple of the period (for example, the upper limit of the required time is “6” (= twice the current “3”) ) And the change amount of the maximized Σvalue (c) in that case, together with the optimal solution, may be generated and output. In this case, for example, when the amount of change in maximize Σvalue (c) due to an operation such as minute relaxation of the constraint is large, it is possible to prompt the determination to accept the change of [Constraint 2] in the work plan. For example, (ii) is assumed to be used in a development mode in which a plurality of teams work while synchronizing at the end of iteration. For example, if other teams complete their work at an end timing equivalent to a multiple of the iteration period, work can be done while ensuring work synchronization timing between teams based on the value of maximize Σvalue (c). You can make a plan to proceed.

  In addition, this invention is not limited to above-described embodiment, Various modifications are included. For example, the above-described embodiment has been described in detail for easy understanding of the present invention, and is not necessarily limited to one having all the configurations described. Further, a part of the configuration of an embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of an embodiment. In addition, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

  Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files for realizing each function can be stored in a recording device such as a memory, a hard disk, or an SSD, or a recording medium such as an IC card, an SD card, or a DVD.

  The control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

  In the above embodiment, the case where the working subset is determined for the work related to the management of the source code (removal of the code clone) of the target software 261 has been described as an example. The present invention can also be applied when determining a work subset for work related to management.

  The present invention can be extended and applied to work related to test code management, which is a code for confirming whether software operates correctly. As a restriction condition regarding test code management, for example, when a test is shared by a plurality of teams, the correction work of one team affects other teams and the adjustment work between the teams or the other teams Don't share multiple test items that may be waiting for work across multiple teams (for example, if a certain test item is selected by a team, other tests will be performed within the same iteration period) For example, the item may not be selected by another team. By imposing such restrictions, it is possible to avoid the risk of impairing the speed of development, such as the occurrence of work waiting time.

10 work planning support device, 251 work information, 252 work expected value policy, 253 work priority information, 254 work conflict information, 255 work expected value information, 256 work subset information, 261 target software, 262 component information, 263 Modification priority policy, 264 code clone detection result, 265 code clone distribution information, 280 work analysis section, 281 code clone detection section, 282 code clone distribution analysis section, 283 code clone modification priority determination section, 284 work conflict information generation section 271 Work expected value determination unit, 272 Work subset determination unit, S200 Work expected value determination process, S900 Work subset determination process, S1400 Code clone distribution analysis process, S1800 Work conflict information generation process, S1813 Code clone distribution analysis process, S18 5 work conflict information generating process

Claims (13)

A method of supporting work planning in software development,
Information processing device
Determine a subset of the work by solving a combinatorial optimization problem with the objective function as the effect of the work group obtained by combining multiple works, with the competitive relationship between multiple works in software development as a constraint. And
Analyzing the source code or document in software development to generate work conflict information that is information about the conflict relationship between the tasks,
Using a competition relationship based on the work competition information as the constraint condition,
A method to support the planning of work plans. A method for supporting the drafting of a work plan according to claim 1,
The information processing apparatus determines the work group that maximizes the objective function or exceeds a predetermined threshold as a work subset by solving the combination optimization problem.
A method to support the planning of work plans. A method for supporting the creation of a work plan according to claim 1 or 2,
The objective function is a function for calculating a sum of expected work values, which is an index representing an effect for each work constituting the work group.
A method to support the planning of work plans. A method for supporting the drafting of a work plan according to claim 3,
The information processing device calculates the expected work value based on information indicating the amount of the work and a priority set for each of the work;
A method to support the planning of work plans. A method for supporting the creation of a work plan according to claim 1 or 2,
The information processing apparatus further imposes a constraint on at least one of the number of man-hours for the work, the time required for the work, and the amount of the work;
A method to support the planning of work plans. A method for supporting the drafting of a work plan according to claim 5 ,
The information processing apparatus is a method for supporting the creation of a work plan with the man-hour of the work and the time required for the work as constraints. A method for supporting the drafting of a work plan according to claim 6 ,
A method for supporting the creation of a work plan, wherein the time constraint for the work is such that the work is completed at a timing that is a multiple of the iteration period in iterative development. A method for supporting the drafting of a work plan according to claim 6 ,
The information processing apparatus is a method for supporting the planning of a work plan that solves a combinatorial optimization problem by at least one of a dynamic programming method, a branch and bound method, and a genetic algorithm. An information processing apparatus that supports work planning in software development,
Determine a subset of the work by solving a combinatorial optimization problem with the objective function as the effect of the work group obtained by combining multiple works, with the competitive relationship between multiple works in software development as a constraint. A working subset determination unit, and
A work analysis unit that generates work conflict information, which is information related to a competitive relationship between the tasks, by analyzing a source code or a document in software development;
With
The work subset determination unit uses a competition relationship based on the work competition information as the constraint condition.
Work planning support device. The work planning support device according to claim 9 ,
The working subset determination unit determines the working group that maximizes the objective function or exceeds a predetermined threshold as a working subset by solving the combination optimization problem.
Work planning support device. The work planning support device according to claim 9 ,
The objective function is a function for calculating a sum of expected work values, which is an index representing an effect for each work constituting the work group.
Work planning support device. The work planning support device according to claim 11 ,
A work expected value determination unit that calculates the work expected value based on information indicating the amount of work and a priority set for each of the work;
Work planning support device. The work planning support device according to claim 9 ,
The work subset determination unit further imposes a constraint on at least one of the number of steps of the work, the time required for the work, and the amount of the work.
Work planning support device.

Images