JP5471912B2 - Software scale arithmetic unit - Google Patents

Description

  The present invention relates to a software scale arithmetic apparatus that includes a specification description support function that describes the specifications of a software program especially for installation, and that can estimate the function scale of the software program based on the written specifications.

  In general, in the production of an embedded software program, a specification is first created, a software program is designed based on this specification, and finally an execution code is generated and mounted on a product. A plurality of program designers share software program development work based on specifications, and design and produce in parallel. For this reason, there is a need to describe the exact specifications of the development elements necessary to produce high-quality software programs with short delivery times.

  In addition, in order to complete the production of the software program on time, it is necessary to grasp the scale of the software program to be produced at an early stage. This is because by accurately grasping the scale of the software program, the man-hours required for production can be calculated by referring to past productivity data and the like, and a more accurate process plan can be performed.

  As for the estimation of the size of the software program, as described in Patent Document 2, conventionally, there are many cases where a person empirically estimates the software size and performs a process plan. At that time, many methods were used to estimate the software scale based on the file size and the number of lines in the source program.

  However, the method based on experience has a problem that the software scale cannot be estimated objectively and quantitatively. Further, the method based on the file size and the number of lines of the source program has a problem that the number of lines of the source program does not necessarily reflect the software scale. For this reason, Patent Document 2 proposes a method called a function point method in order to estimate an objective and quantitative function scale. In this function point method, since measurement is performed by paying attention to processing and data, elements such as human experience are eliminated, and there is an advantage that an objective and quantitative software scale can be measured. Patent Document 1 discloses an estimation technique based on the COSMIC-FFP method (ISO / IEC 19761) focusing on data movement so as to be suitable for development of software for embedded use. The COSMIC-FFP method is a kind of function point method standardized by ISO.

JP 2008-77449 A Japanese Patent Laid-Open No. 2001-117863

  However, in this function point method, man-hours are required for processing and data extraction that are the basis of estimation, and how to perform this extraction work efficiently becomes a problem.

  For example, in Patent Document 1, an activity diagram is extracted from a use case diagram of a UML diagram, and processing and data that are the basis of estimation are extracted from the structure of the activity diagram. In this method, it is necessary to express the specification in UML. However, especially in the embedded field, there are many cases where it is not easy to apply UML that has evolved in the field of object-oriented development, and there is a problem that it takes more man-hours than when creating specifications with natural sentences (natural language sentences). .

  Further, in Patent Document 2, a predetermined transaction is identified using a keyword in a use case sentence described in a natural sentence, and a function point is calculated as a software scale value based on the identification result. . However, this method has a problem that high accuracy is required for keyword extraction, and there is a problem that the accuracy of the calculated software scale is lowered when keywords are not sufficiently described in the use case.

  The present invention has been made in view of the above-described circumstances, and can efficiently create a specification by natural sentences using data serving as a basis for estimating the function scale. It is an object of the present invention to provide a software scale arithmetic device that enables estimation of scale.

  In order to achieve the above object, in the present invention, an input means for data (hereinafter referred to as “index data”) serving as an index for estimating the function scale of software is provided, a specification description means is provided, and the index data is Use it to describe the specification. After completing the description of the specification, the collation unit confirms that all index data are described in the description of the specification.

  In the present invention, the data movement item name and the corresponding data movement type are adopted as the index data. Also, a specification description means is provided to describe the specification using the data movement item name. After completing the description of the specification, the collation unit confirms that all data movement item names are described in the specification.

  On the other hand, the data movement type includes external data input (entry; E), external data output (exit; X), internal data write (write; W), and internal data read. (Lead: R)

  The function scale of the software program is calculated by counting the number of data movement item names for each data movement type from the data movement item name, data movement type, and specifications input as described above.

  Specifically, the software scale arithmetic device according to the present invention is a software scale arithmetic device that supports the input of the specification data of the software program and calculates the functional scale of the software program based on the input specification data. The software scale arithmetic device inputs index data serving as an index for estimating the function scale of a software program through an input screen and stores it in an index data input means for storing in a storage device, and software for inputting characters in a natural language through the input screen Specification input means for storing program specification data in a storage device, and verification means for comparing the specification data and index data to identify and display characters corresponding to the index data in the specification data displayed on the input screen; And a software program using the index data And estimate calculation unit for calculating the functional size of the ram, characterized by comprising a.

  In the present invention, the specification data can be input efficiently by identifying and displaying characters corresponding to the index data in the specification data, and the function scale of the software program is calculated using the index data.

  In the software scale arithmetic device according to the present invention, the index data includes a data movement item name that is a name assigned to each input of data to the software program and output of data from the software program, and the data movement Reference data representing the function scale weighted for each data movement type is stored in a storage device. The collation means collates the data movement item name and the specification data in the index data, and identifies and displays the character corresponding to the data movement item name in the specification data displayed on the input screen. The estimation calculation unit totals the number of data movement item names for each data movement type, and each data movement type is included in the total result. Multiplied by the reference data to calculate the function scale.

  Software programs are created based on specifications, but the estimated man-hours vary greatly depending on the quality of the specifications. In the present invention, first, a data movement item name for each data movement type that greatly affects the estimation of the function scale is input to clarify the data movement between the external device and the memory. Then, the specification creation is supported so that the data movement item name appears in the specification. This makes it possible to create specifications that enable accurate estimation of the software scale.

  The specification description support unit of the software scale calculation device according to the present invention further includes quality determination means for calculating the quality level of the specification data based on the ratio of the number of data movement item names existing in the specification data. Further, the present invention is characterized by comprising a fluctuation range calculating means for calculating a fluctuation range of the calculated function scale of the software program based on the quality level.

  In the present invention, the variation range of the function scale of the software program calculated from the quality level of the specification data is calculated. This can prevent delays in development processes and excess development costs due to underestimation of the function scale of the software program.

  In addition, the specification description support unit of the software scale arithmetic device according to the present invention further extracts, from among a plurality of specification data stored in the storage device, specification data whose function name matches the specification data to be created. Further, when a plurality of specification data is extracted, similar specification extraction means for displaying in order from the highest quality level based on the quality level is provided.

  In the present invention, specification data that is close to the specification data to be created this time is extracted from the specification data accumulated in the past with a priority level at the quality level.

  According to the present invention, data (index data) that is an index for estimating the function scale of software and specification data are separately input, and characters corresponding to the index data in the specification data are identified and displayed. Good quality specifications can be created. . By doing so, it is possible to generate a specification with no omissions. In addition, since the enumeration of index data that estimates the software function scale such as the data movement item name and data movement type is completed at the same time as the specification description, an objective and quantitative estimation of the software scale can be performed at an early stage of development. Can be obtained. As a result, since the software production work can be shared, high-quality software can be produced in a more accurate process, and a delivery date can be secured.

It is an apparatus block diagram of the software scale calculating apparatus 1 by embodiment of this invention. It is a functional block diagram of the software scale calculating apparatus 1 by the 1st Embodiment of this invention. It is an example of a display of the input screen by embodiment of this invention. It is a flowchart which shows the operation procedure at the time of use of the software scale calculating apparatus 1 of FIG. It is a flowchart which shows the process sequence of the collation means 34 of FIG. It is explanatory drawing of the data movement classification by embodiment of this invention. It is a functional block diagram of the software scale calculating apparatus 1 by the 2nd Embodiment of this invention. It is a flowchart which shows the process sequence of the collation means 34 of FIG. It is explanatory drawing of a mode that the tag was inserted in the specification data by embodiment of this invention, and the data example of specification DB52 of FIG. 7 (FIG. 9 (a)), data movement item name and classification DB51 (FIG. 9). It is an example of data of (b)). It is a flowchart which shows the process sequence of the quality determination means 36 of FIG. It is a flowchart which shows the process sequence of the similar specification extraction means 37 of FIG. It is a flowchart which shows the process sequence of the fluctuation | variation range calculating means 38 of FIG.

  Embodiments of the present invention will be described below. FIG. 1 is a device configuration diagram of a software scale arithmetic device 1 according to the present embodiment. As shown in this figure, the software scale arithmetic device 1 includes an input device 6 for inputting data, an arithmetic device 3 for executing various arithmetic processes, a main storage device 4 used for program execution of the arithmetic device 3, etc. The storage device 5 includes an auxiliary storage device 5 that stores a program and a display device 7 that displays calculation results. Each of the devices 3 to 7 transfers data through the internal bus 2.

  FIG. 2 is a functional block diagram of the software scale arithmetic apparatus 1 according to the first embodiment of the present invention. In this figure, the software scale arithmetic unit 1 inputs a data movement item name which is the name of a data input / output unit via the input device 6 and stores it in the data movement item name / type DB 51. 31. For each data movement item name, input a data movement type, which is information for distinguishing input / output operations with an external device or memory, and associate it with the data movement item name corresponding to the data movement item name / type DB 51. Data movement type input means 32 for saving, specification input means 33 for inputting specification data of a software program and saving it in the specification DB 52, collation for determining whether or not a character corresponding to a data movement item name exists in the specification data Function scale for calculating the function scale of the software program using the means 34 and the data movement item name and the data movement type Has a calculation means 35. Each means 31-35 is realizable by a program as a function of CPU. The means 31 to 34 constitute a specification description support unit 41 that efficiently inputs specification data, and the function scale calculation means 35 constitutes an estimate calculation unit 42 that calculates a software program with high accuracy. The data movement item name input means 31 and the data movement type input means 32 constitute an index data input means 30.

  The function of the software scale arithmetic device 1 shown in FIG. 2 will be described in relation to FIG. 1. A program for realizing each means 31 to 35 is stored in the auxiliary storage device 5, and this program is stored in the main storage device at the start of operation. 4 is executed by the arithmetic unit 3. The input data and calculation results are stored in the auxiliary storage device 5. Note that the input device 6 and the display device 7 may be configured as a single unit, and data may be input through an input screen displayed on the display device 7. In the present embodiment, description will be made on the assumption that the input device 6 and the display device 7 are integrally configured.

  Next, an operation procedure when using the software scale arithmetic apparatus 1 will be described with reference to FIGS. The user inputs a data movement item name, a data movement type, and specification data through an input screen displayed on the display device 7 shown in FIG. On the left side of the screen, a specification item display unit 13 which is a screen (window) for displaying the specification ID, the function name, and the input specification is displayed. FIG. 3 shows a state in which the function name “valve automatic opening function” is selected as the specification item of ID1. This field is initially blank, and the function names sequentially input through the function name input unit 15 are displayed. In the upper right part of the screen, a data movement item name / type display unit 14 which is a window for displaying the data movement item name corresponding to the selected specification item and the corresponding data movement type is displayed. Enter data directly on this screen. Since the data movement type is limited, it may be selected and specified by a pull-down menu or the like. In the example of FIG. 3, up to three items (ID1 to ID3) are displayed, but the screen is scrolled as the number of data increases.

  Below the data movement item name / type display section 14 are a function name input section 15 that is a function name input screen, a specification input section 16 that is a specification data input screen, and a screen that displays a result of collation processing described later. A collation result display unit 17 and a function scale display unit 18 which is a screen for displaying a function scale calculation result are provided.

  The user selects a specification ID to be worked on in the specification item display section 13 of this input screen. If the function name has not been registered yet, the function name is registered (S102). Next, as a step before inputting the specification data, a name is assigned for each data exchange with an external device such as a sensor or a memory, and the name is used as a data movement item name from the data movement item name / type display unit 14. input. The input data is stored in the data movement item name / type DB 51 by the data movement item name input means 31 (S103). Next, the user inputs a data movement type corresponding to each data movement item name. The input data is stored in the data movement item name / type DB 51 by the data movement type input means 32 (S104). Although this database structure is not particularly illustrated, an ID is assigned to each data movement item name as displayed on the data movement item name / type display unit 14, and a data movement type is associated with each data movement item name. ing.

  Next, the user inputs specification data from the specification input unit 16 on the input screen. The input specification data is stored in the specification DB 52 by the specification input means 33 (S105). When describing the specification, an item (sentence, word, etc.) input as a data movement item name is used and input in a natural language. At this time, if the data movement item name is input sufficiently and sufficiently, the specification data is matched and the specification is described by using all the data movement item names in the data movement item name / type display unit 14 in the specification text. (S106).

  On the other hand, when the data movement item name is insufficient and the specification cannot be described as intended, the process returns to step S103 to additionally input the data movement item name and the corresponding data movement type. When the description of the specification is completed using all the data movement item names, the user presses the collation button 19 on the input screen and executes collation processing (S107). In the collation process, it is confirmed that all data movement item names are described in the specification data (S108). As a result of the collation, if there is a data movement item name that does not exist in the specification data, it is deleted as an unnecessary data movement item name (S109). After deletion, collation is performed again (S107), and if there is no data movement item name that does not appear in the specification data, the operation for the specification item is completed (“YES” in S108). The processing from step S102 to step S109 is repeated for necessary specification items (S101a, S101b).

  FIG. 5 is a flowchart showing the processing procedure of the collating means 34. The verification unit 34 secures a verification failure number counter (cnt) for storing the verification failure number and an area for storing the verification result on the main storage device. When the collation process is started, these storage areas are initialized (S201).

  Next, the collating unit 34 selects data movement item names one by one for all data movement item names (S203), and determines whether or not the selected data movement item name appears in the specification data ( S204). If the selected data item does not appear in the specification data, the collating unit 34 increments the collation failure number counter (S205), and stores the name of the data movement item that failed in collation as the collation result (S206). ). The above processing is repeated for all data movement item names (S202a, S202b).

  When the collation is completed for all the data movement item names, the collation unit 34 displays the collation result on the collation result display unit 17 and ends the collation process (S207).

  In FIG. 3, in addition to the screen display unit described above, a collation button 19, a function scale estimation button 20, a save button 21, and an end button 22 are displayed on the lower right of the screen. This button is a pseudo button on the screen, and is operated using a cursor movement button on the touch panel or the keyboard of the input device 6 or a pointing device such as a mouse.

  When the function scale estimation button 20 on the input screen is pressed, the function scale calculation means 35 is activated, counts the number of input data movement item names and data movement types, and calculates the number of function points. As an example of the calculation method, there is a method in which the number of data movement item names is added for each data movement type, and the total is used as a function point. In another embodiment, function points may be calculated by weighting each data movement type (that is, correcting with reference data).

  When the COSMIC-FFP method (ISO / IEC 19761), which is a function point method standardized by ISO, is used, the data movement type is an external data input (entry; E) as shown in FIG. ), Data output to the outside (exit; X), data write to the internal memory (write; W), and data read from the internal memory (read; R). The weighting at the time of addition is 1. In this embodiment, each process is assumed to be a memory transfer, but registers, I / O ports, and the like are also managed by including them in the internal memory.

  In the example of FIG. 3, there is one specification item, and the data movement type corresponding to the data movement item name in the specification item is one entry (E), one exit (X), and one lead (R). Since the number of data and write (W) is 0, when the weight for each data movement item name is added as 1, the total number of data items is 3. If this is added over all the specification items, there is one specification item, and as a result, the total number of data items is three. In the COSMIC-FFP method, the function scale is displayed in units of CFSU, so that the function scale of this software is finally 3CFSU. The calculation results of these function points are displayed on the function scale display unit 18 and are stored in the function scale estimation DB 53. Then, when the save button 21 is pressed, the input data is saved in the auxiliary storage device 5. Further, when the end button 22 is pressed, the input screen is closed, and the functions of the specification description support unit 41 and the estimation calculation unit 42 are ended.

  As described above, according to the present embodiment, when inputting specifications in a natural sentence, a data movement item name and a data movement type, which are indices for estimating the function scale of software, are separately input, while referring to these index data, Check whether all data movement item names appear in the entered specifications. For this reason, unlike the conventional method of extracting keywords and the like as described in Patent Document 2, it is possible not only to prevent an excess or deficiency of the described specification, but at the same time, an excessive amount of data movement item names at the time of specification description. The shortage can also be confirmed. As a result, it is possible to describe a specification with no leakage.

  In addition, since the enumeration of data, which is an index for estimating the software function scale such as the data movement item name and data movement type, is completed at the same time as the specification description, an objective and quantitative estimation method such as the function point method can be used. It can be applied at an early stage, and an objective and quantitative software scale estimate can be calculated at an early stage of development.

  As described above, it becomes possible to share software production work, and it is possible to produce high-quality software based on accurate specifications with no omissions in a short delivery time.

Next, a second embodiment of the present invention will be described.
FIG. 7 is a functional block diagram of the software scale arithmetic apparatus 1 according to this embodiment. The difference from FIG. 2 is that the quality determination means 36 for calculating the quality of the input specification data, the similar specification extraction means 37 for extracting similar specification data from the specification data stored in the specification DB 52, and the function scale This is the addition of the fluctuation range calculation means 38 for calculating the fluctuation range (that is, the error range) of the functional scale calculated by the calculation means 35. The quality determination unit 36 and the similar specification extraction unit 37 are included in the specification description support unit 41. Further, the fluctuation range calculation means 38 is included in the estimation calculation unit 42. Other constituent elements are the same as those in FIG. 2, and the same elements are denoted by the same reference numerals and description thereof is omitted. Each means 31 to 38 can be realized by a program as a function of the CPU.

Hereinafter, the operation of the software scale arithmetic apparatus 1 according to the present embodiment will be described focusing on the differences from the first embodiment.
(Verification process)

First, the operation of the matching unit 34 will be described with reference to FIG. The collating means 34 is activated by an interrupt signal every time a character is input to the input device 6. When the verification unit 34 is activated, first, a character string corresponding to the number of characters of the data movement item name including the final character is extracted from the specification data. Then, it is determined whether or not the extracted character string matches the data movement item name. If they match, a tag is inserted before and after the character string in the specification data. FIG. 9 is an explanatory diagram of a state in which a tag is inserted in the specification data. For example, the character string “<B1> sensor signal reception </ B1>” in the specification data shown in FIG. 9A is the data movement stored in the data movement item name / type DB 51 shown in FIG. 9B. This corresponds to the item name ID and the data movement item name “B1: sensor signal reception”. At this time, the collation unit 34 inserts tags <B1> and </ B1> corresponding to the ID of the data movement item name immediately before and immediately after the character string “sensor signal reception” in the specification data. Generally, for the ID “Bi (i = 1 to n)” of the data movement item name, the corresponding tags <Bi> and </ Bi> (i = 1 to n) are placed before and after the character string of the data movement item name. Inserted into.
When the tag is inserted, the matching unit 34 sets a match flag associated with the data movement item name in the data movement item name / type DB 51. Thereafter, the character strings with tags in the specification data are displayed in different colors.

  On the other hand, if the extracted character string does not match the data movement item name in step S303, it is determined whether or not the end of the data movement item name in the data movement item name / type window 14 has been reached. Otherwise, the name of the next data movement item for which no match flag is set is extracted, and the process proceeds to step S302.

  When the end is reached in step S307, the data movement item name for which the match flag is not set in the data movement item name / type window 14 is displayed in a color-changed manner (S309).

In the present embodiment, collation processing is automatically performed by inputting specification data, so the collation button 19 may be omitted.
(Quality judgment processing)

  The quality judgment means 36 is set in advance corresponding to the data movement type of the data movement item name when the data movement item name in the data movement item name / type window 14 matches the character string in the specification window 16. A matching point counter for adding points and a total point counter for counting the total points of the data movement type corresponding to all the data movement item names in the data movement item name / type window 14 are used.

Hereinafter, the operation of the quality judgment unit 36 will be described with reference to FIG. When the quality determination unit 36 is activated periodically or at the end of the operation of the verification unit 34, the quality determination unit 36 resets the coincidence point counter and the total point counter (S401). Then, the data movement item name / type DB 51 is accessed to read the state of the match flag of the data movement item name (S403) and determine whether or not the match flag is set (S404). If the coincidence flag is set, both the coincidence point counter and the total pointer are counted up by points corresponding to the data movement type (S405). If the coincidence flag is not set, only the total point counter is counted up (S406). After repeating the processing of step S403 to step S406 for all the data movement item names of the specification items to be calculated, the numerical value of the coincidence point counter is divided by the numerical value of the total point counter (S407), and this calculation result is converted to the quality level. To the display device 7 (S408).
(Similar specification extraction process)

  Next, the operation of the similar specification extraction unit 37 will be described with reference to FIG. When the similar specification extraction unit 37 is activated by pressing a separately provided similar specification extraction button, it accesses the specification DB 52 (S601) and extracts the first specification ID (S602). Then, the similar specification extraction unit 37 determines whether or not the function name of the extracted specification ID matches the function name of the specification data currently being input (S603). The specification ID is registered in the temporary buffer (S604). After that, the similar specification extracting unit 37 determines whether or not all the specification IDs have been completed. If not, the similar specification extracting unit 37 extracts the next specification ID (S606), and proceeds to step S603.

When the process is completed for all the specification IDs in step S605, the similar specification extracting unit 37 next determines whether or not a plurality of specification IDs are registered in the temporary buffer (S607), and there is a plurality of specification ID registrations. In this case, sorting is performed in descending order of quality level (S608), and the function name, data movement item name, data movement type, specification data, etc. of the specification are displayed and output to the display device 7 (S609).
(Variation range calculation processing)

Next, the operation of the fluctuation range calculation means 38 will be described with reference to FIG. When the fluctuation range calculation means 38 is activated at the end of the operation of the function scale calculation means 35, the calculation result X of the function scale calculation means 35 is extracted (S501). Further, the fluctuation range calculation means 38 extracts the quality level Y of the specification data calculated by the quality determination means 36, divides the calculation result X by the quality level Y, and calculates the value of the fluctuation range of the functional scale. The maximum value is set (S502).
Then, the fluctuation range calculation means 38 outputs the calculation result X extracted in step S501 as the lower limit value of the functional scale fluctuation range and the maximum value calculated in step S502 as the upper limit value of the functional scale fluctuation range (S503).

  As described above, according to the present embodiment, the following effects are obtained in addition to the effects of the first embodiment. First, during the specification data input, the character string match with the data movement item name is detected, and if it matches, the color change display is performed in the specification data, and the data movement not appearing in the specification data is performed. Since the item names are also displayed in different colors, the accuracy of the specification is improved and the efficiency of the specification data input operation is improved.

  In addition, the quality level is calculated by comparing the data movement item name in the data movement item name / type window with the specification data in the specification window, thereby correcting the estimation of the function scale and determining the fluctuation range. Highly accurate estimation is possible. Furthermore, since the past similar specifications are extracted and the extracted data are displayed in the data movement item name / type window and the specification window, the user only needs to input the corrected data and input the data from the beginning. Because there is no, the efficiency of the specification creation improves.

  The present invention is not limited to the above-described embodiment, and can be implemented with various modifications without departing from the scope of the invention.

  For example, when the data movement item name and the wording in the specification data do not completely match, the data movement item name and the wording in the specification data may be associated with each other manually. If this association is stored in a table and the matching process does not completely match, refer to this table, and if the word corresponding to the data movement item name exists in the specification data, the specification data is considered to match. A tag may be inserted therein, and a match flag for the data movement item name may be set. Thereby, the convenience at the time of preparation of specification data improves more.

DESCRIPTION OF SYMBOLS 1 Software scale arithmetic unit 2 Internal bus 3 Arithmetic unit 4 Main storage unit 5 Auxiliary storage unit 6 Input unit 7 Display unit 13 Specification item display unit 14 Data movement item name / type display unit 15 Function name input unit 16 Specification input unit 17 Verification Result display section 18 Function scale display section 19 Collation button 20 Function scale estimation button 21 Save button 22 End button 30 Index data input means 31 Data movement item name input means 32 Data movement type input means 33 Specification input means 34 Collation means 35 Function scale Calculation means 36 Quality determination means 37 Similar specification extraction means 38 Fluctuation range calculation means 41 Specification description support section 42 Estimate calculation section 51 Data movement item name / type DB
52 Specification DB
53 Function scale estimate DB

Claims (4)

A software scale arithmetic device that supports the input of software program specification data and calculates the functional scale of the software program based on the input specification data,
The software scale computing device
Index data input means for inputting index data as an index for estimating the function scale of the software program through the input screen, and storing it in a storage device;
Specification input means for storing specification data of a software program that is input in natural language through the input screen in the storage device;
Collating means for collating the specification data with the index data and identifying and displaying characters corresponding to the index data in the specification data displayed on the input screen;
A specification description support unit having
A software scale calculation system comprising: an estimation calculation unit that calculates a function scale of the software program using the index data. The software scale arithmetic device according to claim 1, wherein
The index data includes a data movement item name given for each input of data to the software program and output of data from the software program, and the data movement item name associated with the data movement item name between the external device or the memory. It consists of a data movement type for distinguishing input / output operations,
Further, reference data representing the function scale weighted for each data movement type is stored in the storage device,
The collation means collates the data movement item name and the specification data among the index data, and identifies and displays characters corresponding to the data movement item name in the specification data displayed on the input screen,
The estimate calculation unit calculates the function scale by totaling the number of data movement item names for each data movement type and multiplying the total result by the reference data of each data movement type. Software scale arithmetic unit. In the software scale arithmetic device according to claim 2,
The specification description support unit further includes a quality determination unit that calculates a quality level of the specification data according to a ratio in which the number of the data movement item names exists in the specification data,
A software scale computing device, wherein the estimate computing unit comprises a fluctuation range computing means for computing a functional scale fluctuation range of the calculated software program based on the quality level. In the software scale arithmetic device according to claim 3,
The specification description support unit further extracts, from a plurality of specification data stored in the storage device, specification data having a function name that matches the specification data to be created, and the plurality of specification data is extracted. In this case, the software scale calculation device comprises similar specification extraction means for displaying in order from the highest quality level based on the quality level.