JPH0916392A - Software development support system - Google Patents

Abstract

PURPOSE: To provide the information of the related result in accordance with corrected contents at the time of correcting an arbitrary result by providing an integrated database for storing relating information for mutually relating all the results obtained in respective development phases. CONSTITUTION: A typical source material database 110 stores all the results in the respective development phases of system development to be the objects of template preparation such as system constitution, function design, source code, object code, execution format, unit test specification, unit test result, combination test specification, combination test result, overall test specification and overall test result. Besides, a relating database 112 stores the information of relating between all the data stored in the routine material database 110 and a non routine material database 113. Then, all the results provided in the development phases are related so that the related module can be detected and presented in the case of changing a certain module of a software to be developed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a software development support method for improving the efficiency of computer system construction and facilitating work.

[0002]

2. Description of the Related Art FIG. 27 shows the development of a V-curve model which is an example of a software process model in a CASE (Computer Aided Software Environment) tool which is a typical example of a conventional software development support method for developing a computer system. It is a figure which shows a flow.

Next, the operation will be described. In such a software development support method, software development includes requirement analysis (explanation of problem), system (outline) design, function (basic) design, detailed (module) design, source code generation, object code generation, system combination, single unit. The development phases of test, integration test, comprehensive test, and operational test will proceed in sequence. In each development phase, various specifications and
Source programs, test data, etc. are generated as deliverables. Those deliverables differ depending on the computer system targeted for development. Also, if a problem occurs in a certain development phase, the software development is proceeded by returning to the upstream development phase.

In order to support the above-mentioned development phase, the CASE tool using the conventional software development support method defines the method of software development based on the methodologies such as structured method and object-oriented analysis, and these methodologies. It provides an editor for creating specifications using the document notation adopted in. For example, the one adopting the structured method has an editor for drawing a data flow diagram (hereinafter abbreviated as DFD), the one adopting the object-oriented analysis has an editor for the class hierarchy diagram, etc. is there. These have not only the function of drawing, but also the function of inspecting whether the drawn contents conform to the framework of each design method. For example, the DFD checks whether the data flow is complete, and the class hierarchy diagram checks whether the inheritance relationship is looped. Also, as mentioned above,
It has a function to generate a partial or complete source code according to these design contents. Further, it has a function of generating an inspection item as to whether an abnormal operation occurs in the generated source code.

Further, the CASE tool using the conventional software development support method has the function of checking the consistency of the product between the individual development phases described above and the conversion of the product between successive development phases. It has the function of. The consistency check can be performed when all the components of the product are prepared for each development phase. In the development phase, generally, more detailed and specific descriptions are given to those related to analysis, design, and manufacturing after the downstream process. Some CASE tools have a function required for the computer system being created and an evaluation function of whether or not the performance is satisfied. For this as well, a detailed description of all components has been completed. Need to be. In particular, performance evaluation cannot be done until the detailed design is completed. Further, the deliverables generated in each development phase are managed for each individual system, and even when a plurality of systems are continuously created, integrated management between the systems is not performed.

[0006]

Since the conventional software development support method is configured as described above, the functions and performances are evaluated after the detailed design of all the product modules is completed. In addition to modifying the module of the artifact that has insufficient function, in order to absorb the impact, the module of other artifacts must be modified to achieve overall consistency, and the related artifacts There is a problem that it takes a lot of time and effort to find a module or the like.

Further, even if there is a similar function to a previously developed system when a plurality of systems are continuously created, it is possible to divert it to a system to be developed later for efficient development. There was a problem that it could not be done.

Further, when the system function is changed during system development or maintenance, not only the module directly realizing the function is modified but also the related module absorbs the change contents of the former module. However, in such a case, it is difficult to estimate the man-hours for changing the specifications by using the conventional software development support method, and it is not possible to present appropriate judgment information for changing the specifications. There was a problem.

The present invention has been made in order to solve the above-mentioned problems, and when correcting an arbitrary product,
The purpose of the present invention is to obtain a software development support method that presents information on related artifacts in order to easily achieve modification of related artifacts according to the content of the modification.

It is another object of the present invention to provide a software development support system capable of improving the development efficiency by providing the artifacts of a similar system as a template.

Further, according to the present invention, a software development support system capable of providing an appropriate judgment material for a specification change by estimating an influence range in which a modification is required due to a change of a product and estimating a man-hour of the modification accompanying a specification change. Aim to get.

[0012]

The software development support system according to the invention of claim 1 is provided with an integrated database for storing association information for associating all the products obtained in each development phase with each other.

According to the software development support method of the second aspect of the present invention, at least in the development phase of the functional design and the detailed design, the development is performed using the template including the predetermined value.

In the software development support method according to the third aspect of the invention, the integrated database manages association information that correlates materials other than the direct deliverables of each development phase with the deliverables generated in each development phase. To do.

In the software development support method according to the invention of claim 4, the version for each of the subsystems and modules and the overall version combining them are managed, and the version including the revision history of the association information is managed. It manages.

According to a fifth aspect of the present invention, there is provided a software development support method, which uses a template including a default value to perform a software function for each development phase of software development.
The performance is verified.

In the software development support method according to the invention of claim 6, for each development phase of software development, it is determined whether or not all the requirements for configuring the system are satisfied by the ratio of replacement of the default value of the template with the design value. It is a judgment.

In the software development support system according to the invention of claim 7, a product in charge of the user is called from the integrated database based on the information input by the user when starting the use.

The software development support method according to the invention of claim 8 is directly involved in the functional change of a part or all of the system under development or already developed based on the system configuration relationship between the deliverables of the system. This is to estimate the range affected by the remodeling of the part to be affected.

The software development support system according to the ninth aspect of the present invention retrieves and presents the person in charge of the remodeling portion from the estimation of the range of influence and the information of the portion in charge of each user of software development.

In the software development support system according to the tenth aspect of the present invention, the man-hours for remodeling are estimated based on the estimated influence range and the man-hours of each product.

The software development support method according to the invention of claim 11 selects a development case deeply related to the system to be developed from past development cases, based on the characteristics of the system to be developed inputted by the user. The product of the development case is used as a template.

The software development support method according to the invention of claim 12 presents information related to a development case by referring to the association information of the development case managed in the atypical material association step for past development cases. It is a thing.

[0024]

The software development support system according to the invention of claim 1 is used for requirements analysis, system design, function design, detailed design, source code generation, object code generation, system combination, unit test, combination test, comprehensive test, and operation test. By associating all the deliverables obtained in each development phase, it becomes possible to detect and present a related module when changing, for example, a certain module of software to be developed. The association can also be used by the user to call to identify the associated module.

In the software development support method according to the second aspect of the present invention, at least in the development phase of the functional design and the detailed design, the software development is advanced using the template having the predetermined value. As a result, the user can be provided with a guideline when designing the system, and the efficiency of the design can be improved. Also, the use of templates can prevent omissions during design.

In the software development support method according to the third aspect of the invention, the materials other than the direct deliverables of each development phase, including at least the meeting minutes, the meeting materials, and the references, and the results generated in each development phase. The association information for associating objects with each other is managed in the integrated database. In this way, since the association extends to the meeting minutes and the like, the background of the design content can be easily confirmed, and mistakes due to incorrect interpretation of the design policy can be reduced.

According to a fourth aspect of the present invention, there is provided a software development support method, wherein software is at least a subsystem,
The module is subdivided, and the version for each of them and the overall version that combines them are managed, and the version management including the revision history of the association information is performed. By performing the version management for each minimum unit, the unit of association and the unit of version management match, and it becomes possible to manage the association for each version. By managing the association information separately from the entity, it is possible to manage the association flexibly, and it is possible to associate the data taken in from the outside.

In the software development support system according to the fifth aspect of the present invention, the function and performance of the software is verified for each development phase of the software development using the template including the default value. By using the template with embedded default values, when designing the contents of individual modules, use the newly designed contents for that module and the default values for the rest of the modules to simulate the behavior. It will be possible to do. By estimating the function and design on the template, it is possible to confirm whether the partial design meets the requirements of the entire system.

According to the software development support method of the invention of claim 6, each development phase of software development,
It is determined whether or not all the requirements for configuring the system are complete based on the ratio of the default value of the template replaced by the design value. More specifically, by collecting marks indicating whether or not the design attached to each module has been completed, it is possible to present an index for grasping the progress of system development.

The software development support method according to the invention of claim 7 manages at least information about the project in which the user participates, the team in the project, and the part in charge of the user, and when the user starts the use. Based on the information you enter, call the artifact that the user is in charge of from the integrated database. Therefore, it becomes possible to collate the information input by the user at the start of use of the system with the attributes of the project and the user, search for, collect, and present the information necessary for the work.

The software development support method according to the invention of claim 8 is based on the relationship in the system configuration between the deliverables of the system when part or all of the functions of the system under development or already developed are changed. Estimate the extent to which the modification of the part directly related to the change will affect. Therefore, it is possible to estimate the degree of influence of the change of the specific module on the entire system and present the evaluation standard for the change of the specification.

In the software development support method according to the invention of claim 9, the person in charge of the remodeling portion is searched and presented from the estimation of the range of influence and the information of the portion in charge of each user of the software development. In other words, for example, by tracing the system configuration relationship between modules to determine the range of influence, and searching the department in charge and the person in charge of each module from the integrated database, contact for modification at design time or elimination of problems. It is possible to present the person in charge of the destination, that is, the portion that needs to be modified.

In the software development support method according to the tenth aspect of the present invention, the number of man-hours for remodeling is estimated based on the estimated range of influence and the number of man-hours for each product. By predicting the number of man-hours, it is possible to provide information for grasping the scale of modification, and when there are multiple methods of modification, it can be used as an index for comparative examination.

In the software development support method according to the invention of claim 11, based on the characteristics of the development target system input by the user, a development case closely related to the development target system is selected from past development cases, and the development case is selected. Use case deliverables as templates. By searching the characteristics of the system, more specifically the module, that is, the function and performance,
It is possible to present similar systems and modules developed in the past. By using not only the template prepared in advance but also the function of searching for similar cases, it becomes easy to obtain a design policy at the time of new design.

The software development support method according to the invention of claim 12 presents information relating to a development case by referring to the association information of the development case managed in the atypical material association step for past development cases. When a development case is searched for, it is possible to better understand the points to be changed in the new design by tracing the association information and knowing the basis of the design of the development case.

[0036]

【Example】

Embodiment 1 FIG. An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a software development support system according to an embodiment of the present invention. In the figure, 101 is a database interface, 108 is a project attribute database (integrated data, base) for storing project attributes, 109. Is a user attribute database (integrated data, base) that stores user attributes,
110 is a fixed-form material database (integrated data, base) that stores fixed-form materials, 111 is a case database (integrated data, base) that stores case information, and 112 is an association database (integrated data, base) that stores association information.
Reference numeral 113 is an atypical material database (integrated data, base) that stores atypical materials. The database interface 101 includes a display control unit 102, an attribute determination unit 103, a repository search unit 104, and a case search unit 10.
5, an association determination unit 106, and an external access unit 107.

Next, the operation will be described. The database interface 101 controls the interface between the user and each database. Each subsystem included in the database interface 101, the display control unit 1
02, attribute determination unit 103, repository search unit 104, case search unit 105, relation determination unit 106, external access unit 1
07 accesses each database while communicating with each other, and calls and updates the data requested by the user.

The project attribute database 108 is
Stores information such as the development system and development content for software development projects that are the target of software development support. Specifically, the name of the project, the project manager, the person in charge, the team division of the person in charge, the content of each team, and the information on the development target system of the project are stored. Figure 2 shows the project attribute database 1
It is a figure which shows the format of the project attribute information stored in 08, In the figure, 201 is a project number, 2
Reference numeral 02 is a project name, 203 is a user number of a person in charge, 204 is a plurality of user numbers of persons in charge, 205 is a plurality of team information, and 206 is a system number of a software system. Project number 201, user number 2
03, 204, and system number 206 are for identifying the project, user, and software system, respectively, and are the serial numbers of the project, system, and user registered in the software system by the software development support method according to the present invention. is there. still,
The project attribute database 108 is searched using the project number 201 or the project name 202 as a key.

FIG. 3 is a diagram showing the format of the team information 205 referenced from the project attribute information. In the figure, 301 is a team number, 302 is a team name, and 303.
Is a user number of a person in charge of the team, 304 is a plurality of user numbers of persons in charge, and 305 to 307 are numbers of a plurality of production units in charge, and the team information 205 includes these pieces of information. The team number 301 is for identifying the team division within the project, and is a serial number within the project. The production unit number is used to identify the product within the project, but the production unit varies depending on the applied development method. For example, when object-oriented analysis / design is applied, the program is subdivided in the order of subsystems, modules, and classes. Therefore, the sequence of numbers of production units is the sequence of these numbers. In order to specify in class units, as shown in FIG. 3, it is specified by a combination of the subsystem number 305 to which it belongs, the module number 306 in the subsystem, and the class number 307 in the module. When designating in module units, it is designated by the subsystem number 305 to which it belongs, the module number 306 in the subsystem, and the identifier indicating the module itself. The identifier of the module itself can be realized by, for example, assigning a class number from 1 to 0 and inserting the module itself.

The user attribute database 109 stores information such as the contents of charge of each user. Specifically, the information of the name of the user, the participating project, the team to which the participating project belongs, and the production unit in charge is stored. FIG. 4 is a diagram showing a format of user attribute data in the user attribute database 109.
Reference numeral 401 is a user number, 402 is a user name, and 403 is information on a plurality of participating projects. Further, FIG. 5 is a diagram showing a format of a list 403 of participating project information referenced from the user attribute data of the user attribute database 109, in which 501 is a project number and 50 is a number.
2 is a line of information in charge. In addition, FIG. 6 is a sequence 5 of information in charge referred to by the sequence 403 of participating project information.
02 is a diagram showing the format of 02, in which 601 is a team number, 602 to 604 are numbers of a plurality of production units in charge of a user, that is, a subsystem number, a module number,
The class number. The user attribute database 109
User number 401 and project number 50 used in
1, team number 601, production unit number, that is, subsystem number 602, module number 603, class number 6
04 is common with that used in the project attribute database 108. The user attribute database 10
9 is searched using the user number 401 or the user name 402 as a key.

The standard material database 110 stores standard products such as a system configuration diagram and a data flow chart created in each development phase of software system development. Specifically, template creation of system configuration, functional design, detailed design, source code, object code, execution format, unit test specifications, unit test results, combination test specifications, combination test results, comprehensive test specifications, comprehensive test results, etc. All the deliverables in each development phase of the target system development are stored in the standard material database 110.

For example, in the case of performing object-oriented analysis / design, a subsystem communication model expressing relationships between subsystems, an object communication model expressing relationships between objects, an object state model expressing state transitions of objects, It is stipulated to create a number of deliverables such as an action data flow diagram (hereinafter abbreviated as ADFD) that expresses the exchange of object operations and data, and a control thread diagram that expresses the control of the overall operation. 7 is an explanatory diagram showing an example of an object state model, FIG. 8 is an explanatory diagram showing an example of an ADFD,
FIG. 10 is an explanatory diagram showing an example of a control thread diagram. These figures are translated by Shinichi Tadashi / Kiyoshi Ito, "Sequel: Object-Oriented System Analysis", Keigaku Shuppan.
Quoted from Melter, Prentice Hall (page 41 for state model, page 102 for control thread diagram, 1 for ADFD)
Described on page 22).

The fixed form material information stored in the fixed form material database 110 is for determining the display method according to the version information for managing the revision history and the contents of the information such as text, diagram and block diagram. Display control information is added. FIG. 10 is a diagram showing the format of the fixed form material information in the fixed form material database 110. In FIG.
01 is the system number, 1002 is the system name, 100
3 is a list of product information. FIG. 11 is a diagram showing a format of a product information array 1003 referenced from the fixed form material information shown in FIG. 10, in which 1101 is a product number, 1102 is a product name, 1103 is a development phase number, 1104. Is the production unit number, 1105 is the man-hour, 110
6 is a display identification number, 1107 is a plurality of product entity information,
Reference numeral 1108 is an unedited list. Development phase number 11
03 is defined in the correspondence table with the development phase name. How to divide the development phase is applied V curve model,
It is determined according to the development method such as the spiral model. The array 1107 of product entity information indicates a revision history for each version of the product. The man-hour 1105 includes an estimate before the design is completed and a result after the design is completed. The unedited list 1108 is a list in which the default value of the template of the product is not replaced with the design value. This depends on the format of the product entity information.

For example, in the case of the control thread diagram shown in FIG. 9, since the processing time of each process is a design target, the initial value of the unedited list 1108 is a list of process names. As the design progresses, that is, each time the default value is replaced with the design value, the item is removed from the unedited list 1108. When the design is completed, the unedited list 1108 is empty. In addition, the standard material database 11
0 is the system number 1001 or system name 1002
Is used as a key to search.

FIG. 12 is a sequence 1 of the product entity information of FIG.
107 is a diagram showing 107, in which 1201 is a product entity number, 1202 is a version number, 1203 is a user number,
1204 is a team number, 1205 is a real call number,
Reference numeral 1206 is a sequence of entry information. FIG. 13 is a diagram showing a sequence 1206 of entry information in FIG. 12, in which 1301 is an entry number, 1302 is position information, and 1303 is a relation number.

The display identification number 1106 shown in FIG. 11 is numbered according to the display format of the information used, such as a flock diagram, an arrow diagram, and a text.
Further, the designation method of the position information 1302 shown in FIG. 13 differs depending on the display format. A block number has a meaning in a block unit like a block diagram, and a node and arc type and its number have a meaning in an arrow diagram such as a node and an arc. The entity call number 1205 of FIG. 12 is used as an identifier for calling the entity of the material. The source code, object code, execution format, etc. as a product may be stored as a file in a normal file system, or may be added to a repository of a specific CASE tool.

The case database 111 stores search information for using past development cases as a template for a newly developed system. Specifically, it stores the names of the systems that have been developed in the past and the information about their functions and performance characteristics. In other words, for the purpose of characterizing each system and module based on its function and performance, and searching for past development cases using those as keys,
Those functions and performances are registered in the case database 111. The functions and performances are expressed by presetting the items expressing them and, for example, digitizing and evaluating each item. FIG. 14 shows the format of the case data in the case database 111.
Is a case number, 1402 is a system number, 1403 is a sequence of function evaluation values (or functional characteristics), and 1404 is a sequence of performance evaluation values (or performance characteristics). System number 1402
Is for identifying a system as an example, and is common to that used in the project attribute database 108.

The function evaluation value (or function characteristic) 1403 and the performance evaluation value (or performance characteristic) 1404 are determined by the user in order to characterize each system by the function and the performance at the end of each project. The function evaluation value (or function characteristic) 1403 is a numerical expression of the function of the system. For example, items such as real-time property and versatility are itemized, and an evaluation value is given to each item. Further, the application field may be used as the function evaluation value (or function characteristic) 1403, and the number can be designated by performing classification according to the type of application. For example, in the database search system, it is possible to classify whether the search target is a bank account, material stock, insurance customer list, or the like. Alternatively, it is possible to adopt a method such as a keyword search of PATLIS. The performance evaluation value (or performance characteristic) 1404 is the function evaluation value (or functional characteristic) 140.
Similar to item 3, items that characterize the system in terms of function are set in advance, and evaluation values are given to them. For example, when items such as responsiveness and module size are set, benchmark quantitative data can be used. Also,
As the performance evaluation value or performance characteristic 1404, in addition to the user's subjective evaluation such as maintainability indicating the easiness of change, the stability of the system indicating the degree of repair of the test result can be used. The case database 111 is searched by an evaluation function based on the function evaluation value (or function characteristic) 1403 or the performance evaluation value (or performance characteristic) 1404.

The association database 112 stores information about associations used for searching related information, such as obtaining proof of design contents and checking test results from designs of past development cases. Specifically, the association database 112 stores information on the association between all the data stored in the standard material database 110 and the non-standard material database 113 described in detail below. FIG. 15 is a diagram showing the structure of the association information in the association database 112. In FIG.
Is an association number, 1502 is an association weight value,
1503 and 1507 are system numbers, 1504 and 150
8 is the deliverable number, 1505 and 1509 are version numbers, 150
6, 1510 are entry numbers. Thus, one association number 1501 is given to one association information, and is composed of a system number, an artifact number, a version number, and an entry number of a pair of associated artifacts.

The value of the association weight is given only when both the associated materials are standard materials. If the association destination is a non-standard material, system number 1503
Alternatively, 1507 is a mark indicating that it is an atypical material, and the product number 1508 is an atypical material number. The association number 1501 is a number for identifying association information, and is assigned through all associations. System number 1503, 1507, deliverable number 1504, 15
08, version number 1505, 1509, entry number 150
6, 1510, atypical material number (when the association destination is an atypical material), the project attribute database 108,
It is common with the standard material database 110, the non-standard material database 113, and the like. A mark indicating that the material is atypical is, for example, the system number 1503 or 150.
It can be realized by numbering 7 from 1, and 0 as the mark of non-standard material. The association database 112 is searched by the association number 1501.

The non-standardized material database 113 stores external materials referred to as supporting materials for the artifacts stored in the standardized material database 110. In particular,
Information such as minutes and meeting materials other than those stored in the standard material database 110 is stored. For atypical materials,
Display control information is added. FIG. 16 is a diagram showing a format of an atypical material in the atypical material database 113. In the figure, 1601 is an atypical material number, 1602.
Is an atypical material name, 1604 is a display identification number, 1605
Is the entity call number. Atypical material incense number 1601
Is an identifier for searching the atypical materials associated with the atypical materials, and is a serial number of those atypical materials. The entity call number 1605 is used to access the entity as in the case of the standard material. The display identification number 1604 is also used to identify the display format of the non-standard material as well as the standard material. The atypical material database 113 is searched by the atypical material number 1601 or the atypical material name 1602.

Next, the operation of each part of the database interface 101 will be described. Display control unit 102
Controls the appropriate display on a display device such as a CRT according to the attributes of the information of the standard material and the non-standard material stored in the standard material database 110 and the non-standard material database 113. The display format is determined by the display identification numbers 1106 and 1604 which are display identifiers.

The attribute determining section 103 has a user name (corresponding to the user name 402 in FIG. 4) and a project name (project name 20 in FIG. 2) that the user inputs at the start of use.
2) and the team name (corresponding to the team name 302 in FIG. 3) are searched for the project attribute database 108 having the format shown in FIG. 2 and the user attribute database 109 having the format shown in FIG. Obtain a list of system numbers 206, production unit numbers, etc. in charge of the user. The team name is entered in order to deal with belonging to multiple teams in duplicate. The version of the product is assumed to be the latest, but when specified, the value of the specified version is used.

The repository search unit 104 includes the attribute determination unit 1
03 got the system number 206, module number 30
6. From the version number 1202, the standard material database 110
To obtain the product entity number 1201 which is the target of the work performed by the user.

The case retrieval unit 105 retrieves the case information database 111 from the function and performance information of the system input by the user, and obtains the system number 1402 whose parameters are the closest. Any method may be used for parameter matching. For example, a method of preparing some evaluation formulas using parameters input in advance and the parameters in the case information database 111, allowing selection from them at the time of search, and freely creating an evaluation formula at the time of search There is. In either case, the evaluation formula is maximized or minimized, and some of the higher ranks are output as a result.

The relation determining unit 106 searches the relational database 112 from the related call information selected by the user on the display of the fixed or atypical material, and determines the other product number 1504 or 1508 of the relation, that is, the fixed or non-constant. Get the standard material number. The external access unit 107 accesses the database in which the entity is stored in the external database among the fixed-form or non-fixed-form material and calls the data.

Next, the operation of software development by the software development support system of this embodiment will be described. Prior to the development of the software system, the guidelines for the deliverables to be generated in each development phase are defined by the template. The template includes the average system configuration of similar systems developed in the past and the functions and performances of the individual modules included therein. Logical relationships such as data exchange between classes and modules, function and logic calls are included in the system configuration. Within each development phase, the deliverables are associated with each other based on the information such as the system module configuration, and the association information is stored in the association database 112 described above. Other than this, the association information between the artifacts may be stored in the association database 112 across all development phases so that the designer can freely associate the artifacts of interest. The association covers all documents generated during system development. The meeting minutes of the meeting for determining the specifications, the communication records between the development teams, and the like are also taken into the association database 112 and are to be associated. The association between modules is weighted according to the strength of mutual association. The association information is determined by a logical structure such as a function call between modules, but when those objective indexes for obtaining the weighting cannot be obtained, it is determined by the user. Weighting is limited to deliverables. No weight is given to the association with external materials.

17, FIG. 18, and FIG. 19 are flowcharts showing the flow of processing for obtaining the association information in the association database 112. There are two types of association between deliverables: one that is logically determined, such as the calling relationship of functions between modules, and one that is freely set by the user. In the former case, the association information is automatically registered in the database of the CASE tool in the process of designing the system with the module configuration editor etc. included in the commercially available CASE tool, so the association database is automatically generated from the association information. It can be stored in 112. The weighting can be automatically determined from the dependency relation based on the inter-function call or the like, but can be overwritten by the user.

On the other hand, in the latter case, as shown in FIG. 17, in step ST1, the user first selects two materials to be associated with each other, and sets an entry by designating an arbitrary position of each of them to set both entries. To associate. In step ST2, it is determined whether or not both (or both ends) entries have been set. If both entries have not been set yet, the process returns to step ST1. Next, step S
At T3, check whether both associations are standard materials,
If both associations are standard materials, step ST
At 4, the value of the association weight is set by the user. Then, in step ST5, the created association information is registered and the process ends.

The entry setting in step ST1 in FIG. 17 is executed according to the flowchart in FIG. First, in step ST11, the user sets an arbitrary entry position in each of two materials to be associated. In both cases of the association by the automatic generation and the association by the user, it is determined in step ST12 whether or not there is an entry at the designated position. If there is no entry, a new entry is created in steps ST13 and ST14. register. The entry position setting in step ST11 of FIG. 18 is executed according to the flowchart shown in FIG. In order to specify the position of the entry, first, in step ST21, it is determined whether or not an arbitrary part of the associated data is structural data as shown in the block diagram. Element is designated, and in the case of non-structured data such as text data, it is designated by coordinates such as the position from the beginning of the sentence in step ST23.

The association database 112 stores atypical materials such as meeting minutes, meeting materials, and references that are not directly involved in the software development support method, and deliverables that are routine materials generated via the software development support method. You may make it possible to manage the association information between them. In this case, it is realized by the flowchart of the entry position setting shown in FIG. 20 in which the position designation of the entry shown in FIG. 19 is modified so as to correspond to the graphic image or the like. Atypical materials such as meeting minutes, meeting materials, and references that are not directly related to the software development support method need to be specified in two dimensions because they include images such as photographs and FAX output. First, in step ST31, it is determined whether one of the related materials is a standard material. If it is not the standard material, in step ST32, the position of the information to be related is designated two-dimensionally. The process (steps ST33 to ST35) in the case of the standard material is shown in FIG.
Therefore, duplicate description will be omitted.

As described above, when all the deliverables are stored in the integrated database, the association information with other production units or other development phases is stored in the association database 112.
Will be recorded. There are two types of association information, one determined by the configuration of the system and one freely set by the user, and all the deliverables are stored in association with each other by the first type of the former. When any one of the artifacts is modified, it is necessary to modify the related artifacts according to the content of the modification so as to obtain the overall consistency, and the association information stored in the association database 112 helps. That is, by associating the artifacts, it becomes possible to detect and present the related module when changing a certain module. At this time, a more accurate judgment can be made by tracing the association information set by the user of the latter second type and knowing the background of the background design. The association information can also be used by the user to call to identify the associated module. Furthermore, as the association extends to atypical materials such as minutes,
It is easy to check the background of the design contents, and it is possible to reduce mistakes due to incorrect interpretation of the design policy.

Next, the template that defines the guideline of the product to be generated in each development phase will be described. The template for the functional design development phase includes a description of the function of each module, and the template for the detailed design development phase includes a description of the performance of each module. The function and performance of each module are set to the average contents of the conventional system as default values. The software system is developed by replacing the default value embedded in the template with the newly designed content. As an example of the template, the template has a typical structure and contents for the fixed form material as shown in FIGS. 7, 8 and 9. The contents included in these are used as prescribed values. The software system will be designed by replacing the specified values with new design contents.

In verifying the function and performance of the software for each development phase of the software development, the simulated operation is executed by using the default value of the template described above, and the partial design contents are required for the entire system. , Verify that the performance is met. In each development phase,
The operation is simulated according to the contents of the function and performance described for each development module. In the functional design development phase,
The operation contents of the system are simulated, and at the detailed design level, the simulation of performance such as responsiveness is performed. The verification of such functions and performances is performed using the template of the standard material as shown in FIGS. 7, 8 and 9. The deliverables that have been partially designed are in a state where the default values have been replaced with some of the design contents.

The ADFD as shown in FIG. 8 describes what kind of conversion is performed for data exchange. Since the input / output with the outside of the system is also described in the ADFD, by verifying all paths from the input from the outside of the system to the output to the outside of the system on the ADFD, it can be confirmed whether the system is processing the data correctly. You can judge. This allows verification of system functions. Since verification of this function is performed by tracing the flow of data, it is necessary to describe the transfer of all data for the module to be verified.

The performance verification is performed using a diagram describing the operation control for each process as in the control thread diagram shown in FIG. The process mentioned here is a meaningful unit of processing such as data rewriting and formula calculation. Each process in the control thread diagram is accompanied by the time required for its processing. Each process cannot start processing until all the processes preceding it have completed. The response performance of the target module can be evaluated by collecting the required time while tracing the control thread diagram from top to bottom. Similar to the ADFD, the control thread diagram needs to describe the processing time and context of all processes.

As described above, in the system under design, default values are embedded in the undesigned parts of the template. By tracing the part where the design is not completed by using the default value, the function and performance can be verified even for the module in the middle of design. It is possible to determine whether or not all the requirements that make up the system are complete for each development phase of software development using the specified values of the template. The design work is performed by replacing the default value on the template with a new design value or adding a new element. At the time of calling the template, all the elements of each standard material are registered in the unedited list 1108 included in the deliverable information 1003 shown in FIG. Each time the default value is replaced, the corresponding element is added to the unedited list 110
Remove from 8. Whether or not the element remains in the unedited list 1108 is the determination of the system design completion. More specifically, the progress of software development is grasped by the rate at which the default value is replaced with the design value or the design content. When the operation to replace the default value with the design value is performed, it is marked that the module has been designed. Depending on the module, the default value may be used as it is, but in this case as well, a mark indicating that the design is completed is made. By collecting the marks of whether or not the design attached to each module has been completed, it becomes possible to present an index for grasping the progress of system development.

As described above, by using the template having the default value, it is possible to give the user a guideline for system design and improve the design efficiency. The template also has the effect of preventing leakage during design. Furthermore, by using the template with embedded default values, when considering the contents of an individual module, the newly designed contents are used for that module, and the default values are used for the rest of the modules. By performing the simulation, it is possible to verify the function and performance on the template for the system under design, and it is possible to confirm whether the partial design matches the requirements of the entire system. Therefore, it is possible to make corrections earlier than when the functional performance is estimated after the entire design is finished, and the design efficiency can be increased.

Next, version management will be described. When the software is subdivided into subsystems, modules, etc., version management for each of them and overall version management combining them, as well as version management including revision history of associations are performed. As shown in the structure of the product entity information 1107 of the standard material database 110 of FIG. 12, the product entity and the version number 1202 are managed as a set. When the product entity is modified and registered as a new version, the product entity is combined with the designated version number 1202 and registered as a new standard material. Register as a new version when the association is changed. The version update due to the change of the association is realized by adding the version update process as shown in FIG. 21 to the flowchart of the association registration shown in FIG. As shown in FIG. 21, in step ST41, the user first selects two materials to be associated with each other, sets an entry by designating an arbitrary position of each of them, and associates them. In step ST42, it is determined whether both (or both ends) entries have been set, and if both entries have not been set yet, the process returns to step ST41. Next, in step ST43, the created association information is registered, and in step ST44
At, the version of the deliverable or the like related to the association information is updated and the process ends. The version is called by searching the project attribute database 108, the user attribute database 109, and the fixed form material database 110, drawing out a product list, and calling the latest version of each product.
For calling a specific version, call the version with the highest version number with a value less than the specified version.

As described above, by managing the version for each minimum unit, the unit of association and the unit of version management match, and it becomes possible to manage the association for each version.
In other words, by managing the version information separately from the entity, it is possible to realize flexible version management for each minimum production unit, and to associate it with data such as atypical materials imported from the outside. .

Next, referring to the flow chart of the process of calling the related information of the user shown in FIG. 22, from the integrated database including the user attribute database 109 based on the information input when the user starts the use. The operation of calling the necessary information will be described. As described above, the user attribute database 109 manages information about the project in which the user participates, the team in the project, the part in charge, and the like. As shown in FIG. 22, when the user name is input in step ST51, in step ST52, the attribute determination unit 103
From the user information in the user attribute database 109 shown in FIG. 4, information on the project in which the user participates is obtained to create a project list, and in step ST53, the project list is displayed on a display device such as a CRT display. Next, in step ST54, when the user selects a specific project from the project list, in step ST55,
Determine the content of the person in charge and call the product entity of the production unit.

Therefore, it is possible to collate the information input by the user at the start of using the software development support system with the attributes of the project and the user, retrieve and collect the information necessary for the work, and present it to the user. It will be possible. Next, when changing part or all of the functions of the system under development or already developed, based on the system configuration relationship between the deliverables of the system, the modification of the part directly related to the change will affect. The estimation process of the range of influence will be described.
FIG. 23 is a flowchart showing the flow of the influence range estimation process, which will be described below with reference to this figure. First,
When the user selects the change target and determines the modification target in step ST61, in step ST62, the association information between the artifacts in the system configuration is searched from the association database 112 and the production unit that may be affected by the modification is selected. Ask. Then, in step ST63, the degree of influence of the modification is calculated for each production unit,
In step ST64, the influence range is determined from the calculated influence degree.

The expression method of the association information on the system configuration differs depending on the system design method and the implementation language, but here, the case where the implementation is performed in an object-oriented language such as C ++ based on the object-oriented analysis / design will be described. In the case of C ++, the smallest unit of implementation is a class. The association information between classes includes function calls, data member access, and inheritance relationships. These association information can be represented by a directed graph having classes as nodes, function calls, data member access, and inheritance as arcs. As described above, the association information is added with the weighting value of the strength of the dependency determined by the logical structure or determined by the user. The weight is labeled on the arc. The calculation of the degree of influence in an object-oriented language will be described below using this directed graph.

A set of nodes corresponding to a class group that is a target of direct modification upon remodeling is a primitive set. The directed graph is searched in consideration of the direction of the arc, and the nodes that can be traced from the nodes of the primitive set are the range of influence. However, if such a search is performed on a directed graph that simply represents the system configuration, the elements in the system have some relation to each other, so that all nodes are likely to be included in the influence range. Therefore, it is determined whether each node is within the range of influence based on the length of the route from the primitive set and the number of routes from the primitive set. An evaluation function expressed by the following equation is used for the judgment.

[0075]

(Equation 1)

Here, V is a set of nodes and E is a set of arcs. t is an association type such as inheritance or reference, and f is a function of weighting by type. P is the influence ripple determined based on the relationship between v and u. This value is the strength of the dependency relationship set by the user described above. This evaluation function is recursively calculated by tracing an arc from the primitive set. When the evaluation function of each node becomes smaller than the threshold value, the value of the evaluation function of that node is set to 0. As a result, a node having a value of the evaluation function other than 0 becomes a node group predicted to be affected.

It is determined that the set of product units corresponding to the set of nodes determined by the above method is likely to be affected as a result of the modification. When the class and module that are predicted to be affected are determined, the person in charge of the affected module and the team in charge can search, and can present the details of the modification to the user as the other party to contact.
FIG. 24 is a flow chart showing the flow of such a person-in-charge search processing. This flowchart is obtained by adding the person-in-charge search to the flowchart showing the flow of the estimation process of the influence range of the modification in FIG. As shown in FIG. 24,
When the influence range of remodeling is determined in steps ST71 to ST74 which are the same as steps ST61 to ST64 of FIG. 23 described above, in step ST75, it is shown in FIG. 12 for each product unit included in the range of remodeling obtained. Information on the user in charge is searched and acquired from the product entity information of the fixed form material database 110.

As described above, it is highly possible that all the deliverables will be the target of the correction only by tracing the related information by the system configuration, but the degree of influence is determined based on the evaluation function that reflects the strength of the association. By evaluating, it is possible to present an evaluation standard for estimating the degree of influence of the change of the specific module on the entire system. As a result, a guideline as to whether or not each work product needs to be corrected can be obtained, and the correction target can be determined efficiently and without omission as compared with individual judgment. Further, by tracing the associations between the modules and searching the departments and persons in charge for each module from the fixed form material database 110, it is possible to present a contact or a negotiation partner for design change, problem solving. It will be possible.

Next, a description will be given of a modification man-hour estimation process based on the modification influence estimation and the man-hours of each product. The estimation of the man-hours due to the remodeling is performed using the influence range estimation process shown in FIG. 23 and the man-hour estimation or the actual result for each product. FIG. 25 is a flowchart showing the flow of the manhour calculation process. As shown in FIG. 25, as shown in FIG.
Step ST8 which is the same as steps ST61 to ST64 of
When the influence range of the modification is determined in 1 to ST84, the man-hours of the modification are calculated for each product unit included in the calculated influence range of the modification in step ST85. Use the man-hours c (v) of each class for the man-hours of modification accompanying a certain change.

(Equation 2) Can be expressed as U (v) is an evaluation function for each class, that is, an influence ripple factor obtained by the equation (1). The estimated man-hours for modification are presented to the user as an index for evaluation of multiple modification plans.

In the process of determining the range of influence of the above modification, the description was made for the class, but the module may be the target, and when considering mainly the module,
The same can be done using a directed graph obtained by contracting a directed graph based on the relation between classes based on the inclusion relation between modules and classes.

As described above, by predicting the number of man-hours for modification, it is possible to provide information for making an appropriate decision on the specification change and grasping the scale of modification, and it is possible to consider multiple modification methods. In such cases, it can be used as an index for comparative examination.

Next, based on the characteristics of the system input by the user, a case with a close relationship is selected from the past development cases stored in the integrated database, presented, and used as a template for new development. The function that enables the above will be described. FIG. 26 is a flowchart showing the flow of processing for obtaining a similar system from such past development examples. As shown in FIG. 26, first, in step ST91, when the user inputs the characteristic parameters of the system, in step ST92, the case search unit 105
In the past cases stored in the case database 111 of FIG. 14, the function evaluation value 1403 and the performance evaluation value 14 added as the parameters of the function and performance evaluation of the system.
04, the correlation between the characteristic parameters input by the user and these parameters is calculated by the evaluation function. An arbitrary evaluation function may be used according to the application, and the average similarity can be evaluated by using, for example, the least square error evaluated by the sum of squared differences of each item. If you want to attach importance to a particular property, you can adjust it by applying a weighting factor to each parameter. Then, in step ST93, the similar system is determined based on whether or not the evaluation function is larger than a certain threshold.

As described above, not only the template but also the system, the function of the module, the parameter of the evaluation of the performance are used to search for the similar past development cases, and the similar system and the module developed in the past are presented. This makes it easier for the user to obtain a design policy even at the time of new design. It has the function of presenting the design basis for past development cases.
The presentation of the basis of the past development case can be realized by calling the atypical material associated with it from the retrieved past cases. This is because the association determining unit 106 searches the association database 112 to obtain the atypical material number of the associated past case, and the external access unit 107 is currently editing the artifact without processing the atypical material. It is possible by calling in the same format as. Therefore, invoking an atypical material associated with a past case copies all the information of the past case, and FIG.
This is executed by listing all edit targets in the unedited list 1108 showing the progress of editing included in the deliverable information shown in FIG.

As described above, when a development case is searched for, the user can freely change the association between the deliverables and know the design basis of the past development case to change the new case. You can better understand what you should do.

[0085]

As described above, according to the first aspect of the invention, the integrated database for storing the association information for associating all the products obtained in each development phase with each other is provided. So, the association has the effect that the user can also call it to identify the associated module.

According to the second aspect of the invention, at least in the development phase of the functional design and the detailed design, the template is used to perform the development, so that the system design can be performed. The guideline can be given to the user, and the efficiency of design can be improved. Also, the use of the template has the effect of preventing leakage at the time of design.

According to the third aspect of the invention, the integrated database manages the association information that correlates the materials other than the direct deliverables in each development phase with the deliverables generated in each development phase. Since it is configured as described above, since the association extends to the meeting minutes etc. in this way, the background of the design contents can be easily confirmed, and there is an effect that errors due to incorrect interpretation of the design policy can be reduced. .

According to the invention of claim 4, the subsystem,
Since it is configured to manage the version for each module and the overall version that combines them, and to manage the version including the revision history of the association information, version management is performed for each minimum unit. By doing so, the unit of association and the unit of version management match, and it becomes possible to manage the association for each version. By managing the association information separately from the entity, flexible association management is possible,
There is an effect that the data can be associated with the data taken from the outside.

According to the fifth aspect of the invention, the template including the default value is used to verify the function and performance of the software for each development phase of the software development. By using the embedded template, when considering the contents of individual modules, it is possible to simulate the operation using newly designed contents for that module and default values for the rest of the modules. Become. By estimating the function and design on the template, it is possible to confirm whether the partial design meets the requirements of the entire system.

According to the sixth aspect of the present invention, for each development phase of software development, it is determined whether or not all the requirements for configuring the system are complete based on the rate at which the default value of the template is replaced by the design value. Since it is configured as described above, more specifically, it is possible to present an index for grasping the progress of system development by aggregating marks indicating whether or not the design attached to each module has been completed. There is an effect that can be done.

According to the invention of claim 7, since the product handled by the user is called from the integrated database based on the information input by the user when starting the use, , The information input by the user at the start of use of the system is compared with the project and user attributes, and the information required for the work can be searched, collected, and presented.

According to the invention of claim 8, based on the system configuration relationship between the deliverables of the system, modification of the part which is directly involved in the functional change of part or all of the system under development or already developed. Since it is configured to estimate the range of influence of, the effect of being able to estimate the degree of influence of the change of the specific module on the entire system and present the evaluation standard for the change of the specification is effective.

According to the invention of claim 9, the person in charge of the remodeling portion is searched and presented based on the estimation of the influence range and the information of the portion in charge of each user of software development. In other words, for example, by tracing the system configuration relationship between modules to find the range of influence, and searching the department in charge and the person in charge of each module from the integrated database, it is possible to change at design time and solve problems. There is an effect that the contact person, that is, the person in charge of the part that needs to be modified can be presented.

According to the tenth aspect of the invention, the man-hours for modification are estimated on the basis of the estimation of the range of influence and the man-hours of each product. Therefore, by estimating the man-hours, It is possible to provide information for grasping the scale of remodeling, and when there are multiple remodeling methods, there is an effect that it can be used as an index for comparative examination.

According to the eleventh aspect of the invention, based on the characteristics of the development target system input by the user, a development case closely related to the development target system is selected from past development cases, and the result of the development case is selected. Since it is configured to use objects as templates, it is possible to present similar systems and modules developed in the past by searching the characteristics of the system, more specifically the module, that is, function and performance. . By using not only the template prepared in advance but also the function of searching for similar cases, there is an effect that the design policy can be easily obtained even at the time of new design.

According to the twelfth aspect of the invention, the information relating to the development case is presented by referring to the association information of the development case managed in the atypical material relating step with respect to the past development cases. With this configuration, when a development case is searched for, it is possible to better understand the points to be changed in the new design by tracing the association information and knowing the basis of the design of the development case.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a software development support system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a data structure of project attribute information stored in a project attribute database 108 in the above embodiment.

FIG. 3 is a diagram showing a data structure of a team information portion of the project attribute information shown in FIG.

FIG. 4 is a diagram showing a data structure of user information stored in a user attribute database 109 in the above embodiment.

5 is a diagram showing a data structure of a participating project information portion of the user attribute information of FIG.

6 is a diagram showing a data structure of a responsible information portion of the participating project information of FIG.

FIG. 7 is a diagram showing an example of an object state model describing an object state change, which is an example of a product when software development by object-oriented design is performed.

FIG. 8 is a diagram showing an example of an action data flow diagram that describes processing contents for each process and data delivery, which is an example of a product when software development is performed by object-oriented design.

FIG. 9 is a control thread diagram describing control of processing for each process, which is an example of a product in the case of performing software development by object-oriented design.

FIG. 10 is a diagram showing a data structure of fixed form material information stored in the fixed form material database 110 in the embodiment.

11 is a diagram showing a data structure of deliverable information 1003 in the fixed form information database 110 of FIG.

12 is a diagram showing a data structure of entity information 1107 of the product of FIG.

FIG. 13 is a diagram showing a data structure of entry information 1206 in the entity information of the product of FIG.

FIG. 14 is a diagram showing a data structure of case data stored in a case database 111.

FIG. 15 is a diagram showing a data structure of association data stored in an association database 112.

16 is a diagram showing a data structure of atypical material data stored in an atypical material database 113. FIG.

FIG. 17 is a flowchart showing the flow of a process of creating association information between deliverables stored in the association database 112.

18 is a flowchart showing the flow of entry setting processing in the flowchart shown in FIG.

19 is a flowchart showing the flow of entry position setting processing in the flowchart shown in FIG.

FIG. 20 is a flowchart showing the flow of entry position setting processing in the processing of associating with a material not directly involved in the software development support method.

FIG. 21 is a flowchart showing the flow of version management associated with the process of creating association information between deliverables stored in the association database 112.

FIG. 22 is a flowchart showing a flow of user-related information calling processing for calling information required by the user.

FIG. 23 is a flowchart showing a flow of influence range estimation processing for estimating the influence range of modification when changing software.

FIG. 24 is a flow chart showing a flow of a person-in-charge contact process for searching a person in charge of a remodeled portion at the time of remodeling.

FIG. 25 is a flowchart showing the flow of man-hour estimation processing for estimating man-hours for modification.

FIG. 26 is a flowchart showing the flow of case use development processing for calling a past development case related to software development.

FIG. 27 is a diagram showing a flow of development of a V-curve model which is an example of a software process model in a conventional software development support method.

[Explanation of symbols]

108 Project Attribute Database (Integrated Database), 109 User Attribute Database (Integrated Database), 110 Fixed Material Database (Integrated Database), 111 Case Database (Integrated Database), 112 Association Database (Integrated Database), 113 Atypical Material Database ( Integrated database).

Claims (12)

[Claims] 1. Requirements analysis, system design, functional design, detailed design, source code generation, object code generation,
Software development that provides an integrated database that stores information about all deliverables obtained in each development phase of system integration, unit test, integration test, comprehensive test, and operational test, and association information that correlates the deliverables with each other. Support method. 2. The software development support method according to claim 1, further comprising a template use development step of performing development using a template including a default value at least in the development phase of the functional design and the detailed design. 3. Association information for correlating materials other than the direct deliverables of each of the development phases and the deliverables generated in each of the development phases, which includes at least the meeting minutes, the meeting materials, and the references. 2. The software development support method according to claim 1, further comprising a step of associating an atypical material with the integrated database. 4. The software is subdivided into at least subsystems and modules, and a version for each of them is managed and an overall version combining them is managed, and
4. The software development support method according to claim 1, further comprising a version management step of performing version management including a revision history of the association information. 5. The software development support method according to claim 2, further comprising a verification step of verifying the function and performance of the software for each development phase of the software development using the template including the default value. . 6. A development progress judging step for judging whether or not all the requirements for constructing a system are complete based on a ratio of replacement of a default value of the template with a design value for each of the development phases of software development. The software development support method according to claim 2, wherein 7. The user manages at least information regarding a project in which the user participates, a team in the project, and a portion in which the user is in charge, and based on the information input when the user starts the use, 2. The software development support method according to claim 1, further comprising a user-related information calling step of calling a product in charge of the user from an integrated database. 8. When modifying a part or all of the functions of a system under development or already developed, based on the system configuration relationship between the deliverables of the system, the modification of the part directly involved in the modification. 2. The software development support method according to claim 1, further comprising an influence range estimation step of estimating a range affected by. 9. A person-in-charge search step of searching and presenting a person in charge of the remodeling portion based on the estimation of the range of influence and information of the portion in charge of each user of software development. 8. Software development support method described in 8. 10. The software development support method according to claim 8, further comprising a remodeling man-hour estimating step for estimating the man-hours of the remodeling based on the estimated influence range and the man-hours of each product. 11. A development case closely related to the system to be developed is selected from past development cases based on the characteristics of the system to be developed, which is input by the user, and the deliverable of the development case is used as a template. The software development support method according to claim 1, further comprising a case use step. 12. A development case design basis presenting step of presenting information related to the development case by referring to association information of the development case managed in the atypical material association step for past development cases. 4. The software development support method according to claim 3, wherein