JP2006172165A - Process mapping device, process mapping management device, process mapping method, process mapping program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently extract a process that can be reused between a plurality of systems without depending on the process granularity.
A process mapping server 3 includes software components c _1j of a system SY1 whose input message set MI _1j is included in an input message set MI _2i of a software component c _2i of the system SY2. A sub-component whose origin is c _1j and whose destination is a software component in which a part of the output message set MO _1l of the software component c _1l of the system SY1 matches the output message set MO _2k of the software component c _2k of the system SY2 Processes are extracted, and these sub-processes are ranked and presented to the user terminal 6.
[Selection] Figure 1

Description

  The present invention relates to a process mapping apparatus, a process mapping management apparatus, a process mapping method, and a process mapping program, and in particular, a method for automatically mapping a process which is an execution procedure of a software component which is a system component between different systems. It is suitable for application to.

  As technologies disclosed in Non-Patent Documents 1 and 2, for example, CORBA (Common Object Request Broker Architecture), which is a representative of distributed object technology, is used as a technology for making software components that are system components available through a network. And a Web service using a message standard called SOAP (Simple Object Access Protocol). For example, a Web service has a WSDL (Web Service Description Language) that describes a message format exchanged between software components in XML, and has a BPEL that describes a process that is an execution procedure of the software component in XML. In the Web service, the code of the module that calls the Web service can be automatically generated by causing the development tool to read the WSDL document or the BPEL document. Further, in the protocol called SOAP, distributed objects on the Web can be used across the platform walls by using Internet standard technologies such as HTTP and XML (extensible markup language).

  Non-Patent Document 3 discloses a method for automatically matching message formats between systems by using a common knowledge base in which common concepts specific to domains are hierarchically arranged. . Here, a common knowledge base is constructed by using a language that hierarchically describes conceptual relationships such as XML and XML-based ontology description language OWL.

In Non-Patent Document 4, the meaning of the execution right, the scope of influence, and the resource locking method is described in the process description, and when creating a similar process, the process description in which these meanings are described is reused. Thus, a method for reducing the design cost for system design is disclosed.
Non-Patent Document 5 discloses a system design approach called SOA (Service Oriented Architecture) that speeds up design by reusing and combining existing processes. In this SOA, a process call procedure and granularity are appropriately designed, and an interface is disclosed using a Web service or the like, thereby aiming at flexible system cooperation by reusing services.

Dark Slama, Jason Garvis, Perry Russell: Enterprise CORBA, Pearson (1999) Satoshi Yoshida, Hideki Aoki: Web Service Kihon, atmarkit, http: // www. atmarkit. co. jp / fxml / tanpatsu / 21websvc / websvc01. html (2002). Paulucci, M.C. , Kawamura, T .; Payne, R.A. and Sycara, K .; : Semantic Matching of Web Services Capabilities, Int'l Semantic Web Conf, pp. 333-347 (2002) Benatallah, B.M. Casati, F .; Toumani, F .; and Hamadi, R .; : Conceptual Modeling of Web Service Conversions, Int'l conf. Advanced Information Systems Engineering, pp. 449-467 (2003) Masayoshi Yasuda: Considering the future of service-oriented architecture, atmarkkit, http: // www. atmarkit. co. jp / fxml / tanpatsu / 33soa / soa01. html (2004).

  However, in the methods disclosed in Patent Documents 1 and 2, in order to realize system linkage, a designer examines a system specification and determines a message format and software component distributed between software components. Since it is necessary to manually match the process description describing the execution procedure, there is a problem that the burden on the manpower is large. For example, in Web services, system cooperation could not be realized unless the message format used for software components defined in WSDL was manually redefined as a common format among the systems to be linked. . Furthermore, in order to link and integrate processes between systems, it is necessary to search for a reusable part of the process description. For this reason, in Web services, system designers must closely examine system specifications, manually discover reusable parts of processes between systems, and redefine new processes in BEPL, enabling system linkage I couldn't.

  Further, in the method disclosed in Patent Document 3, a common concept specialized for a domain is simply arranged hierarchically, and therefore, a reusable process cannot be automatically searched. was there. For this reason, in order to realize system linkage, designers carefully examine multiple specifications of linked systems, manually discover reusable parts of processes between systems, and process linked systems I had to redefine the description.

  Further, the method disclosed in Patent Document 4 has a problem that a reusable part cannot be automatically searched between processes of different systems only by using the meaning described in the process description. For this reason, in order to realize system linkage, the designer has to scrutinize the system specifications and manually match the process between different systems.

In addition, in the method disclosed in Patent Document 5, in order to realize flexible system cooperation by reusing services, it is necessary to appropriately set the process call procedure and granularity, and the reusable process. Since it is necessary to overcome the search problem, there has been a problem that SOA has not been stably established as a technology.
Therefore, an object of the present invention is to provide a process mapping apparatus, a process mapping management apparatus, a process mapping method, and a process that can efficiently extract a process that can be reused between a plurality of systems without depending on the process granularity. To provide a mapping program.

  In order to solve the above-described problem, according to the process mapping apparatus according to claim 1, an input message for extracting a software component having a correspondence relationship between processes based on an input message of the software component constituting the process. Mapping means; output message mapping means for extracting software components in correspondence between processes based on output messages of software components constituting the process; input message mapping means; and output message mapping means Sub-process extraction means for extracting sub-processes having the software component extracted in step 1 as a starting point and an ending point, respectively.

  As a result, when extracting sub-processes, the input and output messages at the start and end points of the software component are matched, and the various software components that make up the process can be diverted to other processes. . This eliminates the need for designers to scrutinize system specifications and manually align process descriptions that describe software component execution procedures to extract processes that can be reused between systems. It is possible to cope with such process granularity, and it is possible to efficiently extract a process that can be reused between a plurality of systems without depending on the process granularity.

  According to another aspect of the process mapping apparatus of the present invention, the sub-process extraction means includes a software program in which the input message set of the software component of the first system is included in the input message set of the software component of the second system. Extracting a sub-process starting from a component and ending with a software component in which a part of the output message set of the software component of the first system matches a part of the output message set of the software component of the second system It is characterized by.

As a result, the input of the first system can be used as it is as the input of the second system, the system cooperation can be easily realized, and the output of the first system that is not in the output of the second system can be realized. And various functions can be added while cooperating with the system.
Further, according to the process mapping apparatus of claim 3, the sub-process extraction unit includes components of an input message set and an output message set of software components between the first system and the second system, respectively. The apparatus further comprises a combination analysis unit that ranks the subprocesses extracted by the subprocess extraction unit based on the matching ratio.

  As a result, it is possible to identify a sub-process including a software component having a high ratio of items included in the input message or the output message matching between systems. For this reason, it is possible to narrow down the sub-processes that are highly reusable from the sub-processes extracted by the sub-process extraction means, and reduce the time and labor when determining reusable processes among multiple systems. Can do.

  Further, according to the process mapping apparatus according to claim 4, the presenting means for presenting the subprocesses ranked by the combination analysis means, and the selection of the subprocesses determined to be reusable among the presented subprocesses It further comprises receiving means for receiving a result, and a process common knowledge base for registering a correspondence relationship of the sub-processes determined to be reusable as a reusable process.

  As a result, it is possible to manually determine whether the sub-process extracted by the sub-process extraction means has a reuse value. This eliminates the need for the process mapping device to determine whether the sub-process extracted by the sub-process extraction means is reusable, and can be reused without the need to build a value judgment necessary for system linkage. Stable process mapping technology can be established. The subprocesses extracted by the subprocess extraction means can be presented to the user after narrowing down the subprocesses with high reusability, and the subprocesses extracted by the subprocess extraction means can be reused. Even in the case where it is manually determined whether or not there is value, it is possible to reduce the labor required for the human labor.

  The process mapping apparatus according to claim 5 further includes a message common knowledge base for storing a correspondence relationship between messages distributed between processes, and the input message mapping means is configured to refer to the reference result of the message common knowledge base. Based on the above, the input message format distributed between the software components is matched, and the software components having a correspondence relationship between the processes are extracted, and the output message mapping means is configured to refer to the message common knowledge base. Based on the result, the output message format distributed between the software components is matched, and the software components having a correspondence relationship between the processes are extracted.

  As a result, even if the message contents match between systems having different message formats, information can be distributed between these systems. This eliminates the need for designers to scrutinize system specifications and manually align the message format distributed between software components, and the application of reusable processes while reducing human labor. The range can be expanded.

  According to the process mapping management device of claim 6, the process common knowledge base in which the correspondence relationship of the reusable process is registered, and the input / output for receiving the input message and the output message of the software component constituting the process A message receiving means, a process searching means for searching a process having a software component corresponding to the input message and the output message as a starting point and an ending point from the process common knowledge base, and the process searching means And a process search means for searching for a reusable process by tracing the correspondence between the processes on the process common knowledge base.

  Thereby, a reusable process can be found by referring to the process common knowledge base. For this reason, in order to extract processes that can be reused between systems, it is no longer necessary for designers to scrutinize system specifications and manually align process descriptions that describe software component execution procedures. Multiple possible processes can be obtained quickly.

  In addition, according to the process mapping method of claim 7, the step of inputting the mapping process description and the mapped process description to the process mapping server, and the software component described in the mapping process description and the mapped process description Extracting a software component having an input message having a correspondence relationship between the mapping process description and the mapped process description, and matching the input message format distributed between the mapping process description and the mapping process description Correspondence between the mapping process description and the mapped process description after matching the output message format distributed between the software components described in the mapped process description A step of extracting a software component having an output message at a subordinate, and a sub-part having a software component having an input message and an output message in a correspondence relationship between the mapping process description and the mapped process description as a starting point and an ending point, respectively. The component of the input message set of the source software component and the component of the output message set of the destination software component match between the mapping process description and the mapped process description, respectively. Ranking and presenting the extracted sub-processes based on the ratio.

  As a result, even if the message contents match even between systems with different message formats, it is possible to distribute information between these systems, and if the process gives the input / output result intended by the designer, A sub-process with high reusability can be presented to the user regardless of the granularity. For this reason, in order to extract processes that can be reused between systems, the designer reviews the system specifications and describes the message format distributed between the software components and the execution procedure of the software components. There is no need to manually match the descriptions, and it is possible to reduce the labor involved in system linkage.

According to the process mapping method of claim 8, the step of presenting the ranked subprocess and the subprocess determined to be reusable among the presented subprocesses are selected as reusable processes. And further comprising the step of:
This eliminates the need for the process mapping device to determine whether or not the subprocess extracted by the subprocess extraction means is reusable, and can establish a reusable process mapping technique stably. it can.

The process mapping method according to claim 9 further comprises the step of creating a database of correspondence relationships of the sub-processes determined to be reusable as reusable processes.
Thus, by referring to the process common knowledge base, a reusable process can be found, and a plurality of reusable processes can be quickly obtained.

  Further, according to the process mapping program according to claim 10, the step of extracting the software component having a correspondence relationship between the processes based on the input message of the software component constituting the process, and the software constituting the process A step of extracting a software component having a correspondence relationship between the processes based on an output message of the component, and a software component extracted by the input message mapping unit and the output message mapping unit as a starting point and an ending point, respectively And a step of causing a computer to execute a step of extracting a sub-process.

  This eliminates the need for designers to scrutinize system specifications and manually align process descriptions describing software component execution procedures to extract processes that can be reused between systems. It is possible to cope with such process granularity, and it is possible to efficiently extract a process that can be reused between a plurality of systems without depending on the process granularity.

  As described above, according to the present invention, when sub-processes are extracted, a process that gives input / output results intended by a designer by matching input messages and output messages at the start and end points of a software component. If so, sub-processes with high reusability can be presented to the user regardless of the granularity, and the effort required for system cooperation can be reduced.

Hereinafter, a process mapping apparatus and a process mapping method according to an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a schematic configuration of a process mapping apparatus according to an embodiment of the present invention.
In FIG. 1, the process mapping apparatus includes a process mapping server 3, a message knowledge base server 4, and a process knowledge base server 5. Here, the process mapping server 3, the message knowledge base server 4, and the process knowledge base server 5 can be connected via a communication network. In addition, the process mapping server 3 can be connected to the user terminal 6 via a communication network. As a communication network connecting the process mapping server 3, the message knowledge base server 4 and the process knowledge base server 5, for example, a public communication network for performing IP communication can be used, and the Internet may be used. Moreover, although it may be a private communication network between companies or a public communication network, it is like an IP-VAN (Internet Protocol-Virtual Private Network) that can provide dedicated communication with high reliability and security. The Internet may be good. The process mapping server 3, the message knowledge base server 4, the process knowledge base server 5 and the user terminal 6 may be a notebook personal computer or a desktop personal computer, such as a mobile phone terminal or a PDA (Personal Data Assistant). Also good.

  The process mapping server 3 is provided with an input message mapping device 31, an output message mapping device 32, a combination analysis device 33, and a user presentation device 34. The mapping process description 1 of the system SY1 and the mapping process description 2 of the system SY2 Are input to the process mapping server 3. Here, the process description refers to, for example, BPEL in a Web service or IDL in CORBA, and can be described in an ontology description language based on XML or XML. In the process description, an arrangement of software components and input / output messages of the software components can be described.

  The process description to be mapped is called mapped process description 1, and the process description to be mapped is called mapping process description 2. Then, by performing the process mapping process of the mapping process description 2 in which the process SY2 of a certain system is described and the mapped process description 1 in which the process of the other system SY1 is described, it can be reused for the process of a certain system SY2. This process can be extracted from the processes of the other system SY1, and a certain system SY2 can be linked or integrated with the processes of the other system SY1.

FIG. 2 is a block diagram showing the relationship between a system and a process according to an embodiment of the present invention.
In FIG. 2, the software component provision server V1 includes software components C1 and C2, the software component provision server V2 includes software components C3 and C4, and the software component provision server V3 includes Software components C5 and C6 are installed, and software component C7 is installed in the software component providing server V4. Here, input messages MI1 to MI7 and output messages MO1 to MO7 are given to the software components C1 to C7, respectively. Then, the process P can be constructed by describing the execution procedure as a process description in the distributed software components C1 to C7. The system is constructed by distributed software components C1 to C7, and provides a target function.

  Further, in FIG. 1, the input message mapping device 31 matches the input message format distributed between the software components described in the mapping process description 1 and the mapped process description 2 and then maps the mapping process description 1 and A software component having an input message in a correspondence relationship between the mapped process descriptions 2 is extracted. The input message mapping device 31 can refer to the message common knowledge base 41 stored in the message knowledge base server 4 when matching the input message format distributed between the software components.

  The output message mapping device 32 matches the output message format distributed between the software components described in the mapping process description 1 and the mapped process description 2 and then maps between the mapping process description 1 and the mapped process description 2. Extract software components with output messages that have a corresponding relationship. The output message mapping device 32 can refer to the message common knowledge base 41 stored in the message knowledge base server 4 when matching the input message format distributed between the software components.

  The combination analysis device 33 starts from a software component in which the input message set of the software component of the system SY1 is included in the input message set of the software component of the system SY2, and is one of the output message sets of the software component of the system SY1. A sub-process having a software component whose end point matches the part of the output message set of the software component of the system SY2 is extracted. Then, based on the ratio between the components of the input message set of the software component as the starting point and the components of the output message set of the software component as the end point between the system SY1 and the system SY2, Subprocesses can be ranked.

  The user presentation device 34 presents the subprocesses ranked by the combination analysis device 33 to the user terminal 6 and receives from the user terminal 6 the selection result of the subprocesses determined to be reusable among the presented subprocesses. . Then, the correspondence relationship of the subprocesses determined to be reusable by the user terminal 6 is registered in the process common knowledge base 51 as a reusable process.

In addition, the message knowledge base server 4 is provided with a message common knowledge base 41 that accumulates correspondences of messages distributed between processes. The process knowledge base server 5 is provided with a process common knowledge base 51 for registering a correspondence relationship of sub-processes determined to be reusable as a reusable process.
Further, the user terminal 6 verifies the reuse value of the sub-processes ranked in the process mapping server 3 and returns the verification result to the process mapping server 3.

FIG. 3 is a block diagram showing a processing sequence of the process mapping apparatus of FIG. In the following description, a process mapping procedure for realizing system cooperation between the system SY2 and the system SY1 having a different administrator is taken as an example. In this example, since the system SY2 performs process mapping on the system SY1, the process description of the system SY1 is the mapped process description 1, and the process description of the system SY2 is the mapping process description 2. The system SY1 includes software components c ₁₁ , c ₁₂ , c ₁₃ , c ₁₄ , and c ₁₅ , and the process (software component execution procedure) is c ₁₁ → c ₁₂ → c ₁₃ → c ₁₄ → c. It shall be ₁₅ . The system SY2 is composed of software components c ₂₁ , c ₂₂ , and c ₂₃ , and the process is assumed to be c ₂₁ → c ₂₂ → c ₂₃ .

  In FIG. 3, when the mapped process description 1 of the system SY1 and the mapping process description 2 of the system SY2 are input to the input message mapping device 31 (S1a, S1c), the input message mapping device 31 stores the message common knowledge base 41. By referencing, the input message format distributed between the software components described in the mapping process description 1 and the mapped process description 2 is matched (S2a). Here, by matching the input message format distributed between the systems SY1 and SY2, even if the message contents match between the systems SY1 and SY2 having different message formats, the information is distributed between these systems SY1 and SY2. Can be made. This eliminates the need for the designer to scrutinize the specifications of the systems SY1 and SY2 and manually match the message format distributed between the software components. The scope of the process can be expanded.

Then, the input message mapping device 31 matches the input message format distributed between the systems SY1 and SY2, and the software component having the input message set having the correspondence relationship between the mapping process description 1 and the mapping process description 2 Extract. That is, among the software components _{c 1j (1 ≦ j ≦ 5} ) with the system SY1, the input message set MI _2i of the input message set MI _1j is a system SY2 software component _{c 2i (1 ≦ i ≦ 3} ) Search for software component c _1j as _contained in. The union set U and the intersection set N between the input message sets MI _1j and MI _2i are calculated for all combinations of the software component c _1j of the system SY1 and the software component c _2i of the system SY2 that satisfy such conditions. Then, the match score S _2i1j = | N | / | U | of the software component c _1j viewed from the software component c _2i is calculated. Note that even if the message formats differ between the systems SY1 and SY2, if the input message contents match, the input messages match and the match score S _2i1j is calculated.

For example, an input message set MI _1j software component c _1j system SY1, against the input message set MI _2i software components c _2i system SY2, if the [MI ₁₁ ∈MI _21], the system SY1 The software component c ₁₁ is selected as a search result of the input message mapping device 31. When the software component c ₁₁ as viewed from the software component c ₂₁ is retrieved to calculate the sum set U and intersection N between input message set MI _11, MI _21, viewed from the software component c ₂₁ calculating a match score S ₂₁₁₁ software components c _11.

Here, by searching for the software component c _1j whose input message set MI _1j is included in the input message set MI _2i , the input message of the system SY 1 can be used as it is as the input message of the system SY 2. It is possible to easily realize system cooperation. When the mapped process description 1 of the system SY1 and the mapping process description 2 of the system SY2 are input to the output message mapping device 32 (S1b, S1d), output message mapping is performed. The device 32 refers to the message common knowledge base 41 to match the input message format distributed between the software components described in the mapping process description 1 and the mapped process description 2 (S). 2b).

Then, the output message mapping device 32 matches the output message format distributed between the systems SY1 and SY2, and the software component having the output message set having the correspondence relationship between the mapping process description 1 and the mapping process description 2 Extract. That is, the sum of the output message set MO _1l of an output message set MO _2k for a system SY2 software component _{c 2k (1 ≦ K ≦ 3} ), a system SY1 software component _{c 1l (1 ≦ l ≦ 5} ) The set U and the intersection set N are calculated, and the coincidence score S _2k1l = | N | / | U | between the software components c _2k and c _1l is calculated. This calculation is performed between all the software components c _2k and c _{1l of} both systems SY1 and SY2. Even if the message formats are different between the systems SY1 and SY2, if the output message contents match, the output message matches and the match score _S2k1l is calculated.

Here, by calculating the coincidence score S _2k1l by _dragging between the software components c _2k and c _{1l of} both systems SY1 and SY2, it is possible to capture what is not in the output message of the system SY2 from the output message of the system SY1. It becomes possible, and various functions can be added while trying to cooperate with the system.
Next, when the input message mapping apparatus 31 extracts a software component having an input message set in a correspondence relationship between the mapping process description 1 and the mapped process description 2, the input message set having the input message set is mapped to the mapping result. Is output to the combination analyzer 33 (S3a). Further, when the output message mapping device 32 extracts a software component having an output message set in a correspondence relationship between the mapping process description 1 and the mapped process description 2, the software component having the output message set is used as a mapping result. It outputs to the combination analyzer 33 (S3b).

Then, when receiving the mapping results from the input message mapping device 31 and the output message mapping device 32, the combination analysis device 33 extracts subprocesses based on these mapping results and outputs them to the user presentation device 34 (S4). .
That is, when the combination analysis apparatus 33 receives the mapping results from the input message mapping apparatus 31 and the output message mapping apparatus 32, among the software components c _1j of the system SY1, the input message set MI _1j is software with the system SY2. component c _2i of the starting point of the software components c _1j subsumed input message set MI _2i, software components of the system SY1 c _1l output message set MO _1l some software components of the system SY2 c _2k A sub-process having a software component c _1l that matches a part of the output message set MO _2k as the end point is extracted. Here, the sub-process means at least a part of a process in a certain system.

  As a result, it is possible to search for software components that have a correspondence relationship with all of the software components that make up the two processes and extract subprocesses, so that they can be reused regardless of the granularity of the process. Various processes according to the system designer's application, ranging from processes with rich functions but heavy processing to processes with low functions but light processing. Can be extracted.

Then, when the subprocesses that are the starting point and the ending point of the software component are extracted, the software components are combined in descending order of the matching score of both the starting point and the ending point, and the subprocesses are ranked. Then, the user presentation device 34 presents the ranked subprocesses to the user terminal 6 (S5). For example, a software component c ₁₁ as a starting point, the software components c ₁₅ and subprocess p ₁ to the end point, the subprocess p ₂ which as a starting point the software components c _21, and ending the software components c ₂₂ is It can be presented to the administrator of the system SY2 as a candidate for a process that can be reused.

  Then, the user terminal 6 verifies the reuse value of the sub-processes ranked by the process mapping server 3 and selects a reusable process from the processes that can be reused. When the user terminal 6 selects a reusable process from the reusable processes, the user terminal 6 returns the selection result to the process mapping server 3 (S6). Here, when a reusable process is selected from reusable processes, the reusability value of the subprocess can be manually verified on the user terminal 6. This eliminates the need for the process mapping server 3 to determine whether or not the sub-process extracted by the process mapping server 3 has a reuse value, and makes it unnecessary to construct a value judgment necessary for system cooperation. Possible process mapping techniques can be established stably. Further, among the subprocesses extracted by the process mapping server 3, the subprocesses having high reusability can be narrowed down and presented to the user terminal 6, and the subprocesses extracted by the process mapping server 3 can be displayed. Even in the case where it is manually determined whether or not there is a reuse value, it is possible to reduce the labor required for the manual operation.

  In addition, processes that can be reused are extracted by automatic mapping of processes using input / output messages of software components between different systems, and can be presented to the system administrator. It is possible to support system design appropriate for the designer while reducing the amount. In other words, by mapping processes of multiple systems, automatically searching for reusable processes between systems, and presenting the search results to the designer, multiple distributed operation systems and business systems are integrated. When building an operation system or business system, it is possible to quickly determine how to design these integrated systems. In addition, by using the process mapping results, it is possible to quickly find detailed processes and lighter processes than the currently designed process, and to integrate and link the processes as required by the designer. It becomes possible.

  Further, in the user terminal 6, for example, if a system design is performed using a Web service, the process description selected by the user terminal 6 is converted into a BPEL that is a process description in the Web service, and is read into a Web service development tool. By doing so, it is possible to call software components described in the process description selected by the user terminal 6 one after another and automatically generate a program for providing a system that realizes a target function. For this reason, by using the mapping result of the process description output from the process mapping server 3, a system that realizes the target function can be automatically constructed. Note that the mapping result of the process description output from the process mapping server 3 can be applied to an inter-system interface other than the Web service or CORBA.

Next, when the user presentation device 34 receives the selection result of the subprocess determined to be reusable from the user terminal 6, the correspondence relationship of the subprocess determined to be reusable at the user terminal 6 can be reused. Is registered in the process common knowledge base 51 as a simple process (S7).
For example, the administrator of the system SY2 on the user terminal 6, a software component c ₁₁ as a starting point, the software components c ₁₅ and subprocess p ₁ to the end point, starting from the software components c _21, software components When it is determined that the sub-process p ₂ whose end point is c ₂₂ is reusable, the sub-process p ₁ and p ₂ is given a meaning (p ₁ = p ₂ ) that can be reused, and the user presentation device 34 can be registered in the process common knowledge base 51.

  As a result, processes that can be reused with each other can be stored in the process common knowledge base 51, and the reusable processes can be created while using the semantic description that is given by past process mapping. It becomes possible to search. Therefore, when a new system design is performed, a plurality of reusable process candidates can be obtained as search results of the process common knowledge base 51 by issuing a certain input message and output message as an inquiry. , Can speed up the system design. In addition, by using the process common knowledge base 51, an appropriate process can be found regardless of the granularity as long as it is a process that gives an input / output result intended by the designer, and is necessary for realizing system linkage. Time and effort can be reduced.

  Further, when the reusable process is accumulated in the process common knowledge base 51, the user terminal 6 can inquire the process knowledge base server 5 about what function system is constructed (S8). For example, the user terminal 6 can send an input message and an output message as an inquiry to the process knowledge base server 5. When the process knowledge base server 5 receives an inquiry from the user terminal 6, the process knowledge base server 5 returns a process having a function that can be reused as a response to the user terminal 6 (S9). For example, when the process knowledge base server 5 receives the input message and the output message from the user terminal 6, the process knowledge base server 5 searches the process common knowledge base 51 for a reusable process candidate corresponding to the input message and the output message, and sends an inquiry response. As a result, rankings of a plurality of reusable processes can be sent to the user terminal 6.

  When the user terminal 6 receives a response from the process knowledge base server 5, the user terminal 6 designs a system using a reusable process obtained by the response. At this time, for example, if the system design is performed using the Web service, the process description selected by the user terminal 6 is converted into BPEL which is the process description in the Web service, and is read by the Web service development tool. Software components described in the process description selected by the user terminal 6 can be called one after another, and a program for providing a system that realizes a target function can be automatically generated. For this reason, by using the mapping result of the process description output from the process mapping server 3, a system that realizes the target function can be automatically constructed.

Note that the input message mapping device 31, the output message mapping device 32, the combination analysis device 33, and the user presentation device 34 are realized by causing a computer to execute a program in which a command for performing processing performed by these devices is described. be able to.
If this program is stored in a storage medium such as a CD-ROM, the input message mapping device 31, the output message mapping device 32, and the like are installed in the computer and the program is installed in the computer. Processing performed by the combination analyzer 33 and the user presentation device 34 can be realized. Moreover, this program can be easily spread by downloading this program via the Internet.

  Further, when the computer executes a program in which an instruction for executing the processing performed by the input message mapping device 31, the output message mapping device 32, the combination analysis device 33, and the user presentation device 34 is executed, the program is executed by a stand-alone computer. Alternatively, it may be distributed to a plurality of computers connected to the network.

Hereinafter, as an embodiment of the present invention, a mapping process between processes of user login systems (system SY1 and system SY2) in two online stores will be described as an example.
FIG. 4 is a block diagram illustrating a specific example of the process of the system SY1 according to the embodiment of the present invention. FIG. 5 is a block diagram illustrating a specific example of the process of the system SY2 according to the embodiment of the present invention. 4 and 5 are simplified processes derived by actually observing the login procedure in the online store.

In FIG. 4, the system SY1 includes software components C ₁₁ , C ₁₂ , C ₁₃ , C ₁₄ , and C ₁₅ . Then, input message sets MI ₁₁ , MI ₁₂ , MI ₁₃ , MI ₁₄ , and MI ₁₅ are assigned to the software components C ₁₁ , C ₁₂ , C ₁₃ , C ₁₄ , and C ₁₅ , respectively, and an output message set MO _{11 is} set. , MO ₁₂ , MO ₁₃ , MO ₁₄ , and MO ₁₅ are respectively assigned. Here, for example, in the input message set MI ₁₁ , “login command” is set as a message type, and “customer name” and “password” are set as data. For example, in the output message set MO ₁₁ , “login data request command” is set as a message type, and “customer name” and “password” are set as data.

In FIG. 5, the system SY2 includes software components C ₂₁ , C ₂₂ , and C ₂₃ . Then, input message sets MI ₂₁ , MI ₂₂ , MI ₂₃ are assigned to the software components C ₂₁ , C ₂₂ , C ₂₃ , respectively, and output message sets MO ₂₁ , MO ₂₂ , MO ₂₃ are assigned respectively. . Here, for example, in the input message set MI ₂₁ , “login command” is set as the message type, and “name” and “password” are set as the data. For example, in the output message set MO ₂₁ , “login data request command” is set as a message type, and “name” and “password” are set as data. Note that the data here refers to data that is actually exchanged between systems.

Here, for example, in the input message set MI ₁₁ of the system SY1, the data is “customer name” and “password”, whereas in the input message set MI ₂₁ of the system SY2, the data is “name” and “password”. The password format is different, and the message format is different even though the input message contents match between the input message sets MI ₁₁ and MI ₂₁ . For this reason, by referring to the message knowledge base server 4, the input message format distributed between the systems SY1 and SY2 is matched. As a result, even if the message contents match between the systems SY1 and SY2 having different message formats, information can be distributed between the systems SY1 and SY2.

FIG. 6 is a diagram showing an equivalence relation of data included in a message between the systems SY1 and SY2 according to the embodiment of the present invention.
6A, the message knowledge base server 4 includes, for example, an equivalence relation between data “customer name” in the system SY1 and data “name” in the system SY2, and data “system” in the system SY1 and the system SY2. The equivalence relation between the data “user” and the data “password” in the system SY1 and the data “password” in the system SY2 are registered.

  In FIG. 6B, the message knowledge base server 4 includes, for example, an equivalence relationship between data “product” in the system SY1 and data “product” in the system SY2, and data and system “system” in the system SY1. An equivalence relation of data “author” of SY2 and an equivalence relation of data “genre” of system SY1 and data of “genre” of system SY2 are registered.

Therefore, by referring to the message knowledge base server 4, the data “customer name” and “password” of the input message set MI ₁₁ of the system SY1, and the “name” and “password” of the input message set MI ₂₁ of the system SY2 Can be matched, and information can be distributed between the systems SY1 and SY2 having different message formats.

If the message types are different, the messages cannot be distributed between the systems SY1 and SY2 unless the message types are matched between the systems SY1 and SY2. For this reason, the message type distributed between the systems SY1 and SY2 is matched by referring to the message knowledge base server 4.
FIG. 7 is a diagram showing a message type equivalence relationship between the systems SY1 and SY2 according to the embodiment of the present invention.

  In FIG. 7, the message knowledge base server 4 includes, for example, an equivalence relation between a message type “login command” of the system SY1 and a message type “login command” of the system SY2, and a message “user data request command” of the system SY1. Equivalent relationship of the message type “user data request command” of the system SY2, the equivalence relationship of the message type “user data response command” of the system SY1 and the message type “user data response command” of the system SY2, The equivalence relation between the message type “search IF display command” and the message type “search IF display command” of the system SY2 is registered. Thus, by referring to the message knowledge base server 4, the message types distributed between the systems SY1 and SY2 can be matched.

FIG. 8 is a diagram illustrating an example of a message type according to an embodiment of the present invention. In the embodiment of FIG. 8, the message in the CORBA IDL is taken as an example.
In FIG. 8, the message type defines the type and arrangement of data for exchanging messages between systems. For example, declare the message name distributed to other systems in the interface type, declare the purpose of the message “AddFilter” in the ResultType type, declare the message to be exchanged between the systems in the in type and out type, and raise an exception in the rises clause An item can be described.

Note that data included in the message is the main subject of the message mapping. In other words, what is distributed between systems is not the message type but the data included in the message, so if the contents of the data included in the message match, by changing the message type between the systems, Messages can be distributed between systems.
If the system SY2 in FIG. 4 performs process mapping on the system SY1 in FIG. 5, the process description of the system SY1 in FIG. 5 is input as the mapped process description 1 to the input message mapping apparatus 31 in FIG. The process description of the system SY2 is input to the input message mapping device 31 as the mapping process description 2.

Then, the input message mapping device 31 refers to the message common knowledge base 41, and among the software components C _1j (1 ≦ j ≦ 5) of the system SY1, the input message set MI _1j is a software program having the system SY2. The software component C _1j that is included in the input message set MI _2i of the component C _2i (1 ≦ i ≦ 3) is searched. The union set U and the intersection set N between the input message sets MI _1j and MI _2i are calculated for all combinations of the software component C _1j of the system SY1 and the software component C _2i of the system SY2 that satisfy such conditions. Then, the match score S _2i1j = | N | / | U | of the software component C _1j viewed from the software component C _2i is calculated. Note that even if the message formats differ between the systems SY1 and SY2, if the input message contents match, the input messages match and the match score S _2i1j is calculated.

For example, the input message set MI ₁₁ software component C ₁₁ is {(customer name), (password)}, the input message set MI ₂₁ software component C ₂₁ is {(name), (password)}, Figure 6, it is understood that (customer name) and (name) are equivalent to each other, and (password) and (password) are equivalent to each other. Therefore, {MI ₁₁ ∈MI _21}, and the input message set MI ₁₁ software component C ₁₁ from being included in the input message set MI ₂₁ software components C _21, the software component C ₂₁ software component C ₁₁ is selected. The union U between the input message sets MI ₁₁ and MI ₂₁ is {(customer name = name), (password = password)}, and the product set N is {(customer name = name), (password = password)}. _Therefore , the coincidence score S ₂₁₁₁ is S ₂₁₁₁ = | N | / | U | = 1.

Similarly, the software components C _21, since a {MI ₁₂ ∈MI _21}, software component C ₁₂ is selected. The union U between the input message sets MI ₁₂ and MI ₂₁ is {(customer name = name), (password = password)}, and the product set N is {(customer name = name), (password = password)}. _Therefore , the match score S ₂₁₁₂ is S ₂₁₁₂ = | N | / | U | = 1.

On the other hand, the software components _{C 21, {MI 13 ∋MI 21} }, because it is _{{MI 14 ∋MI 21}, {} MI 15 ∋MI 21}, software components C _13, C _14, C ₁₅ is Not selected, the coincidence scores S ₂₁₁₃ , S ₂₁₁₄ , S ₂₁₁₅ are not calculated.
Moreover, the software component C _22, since a {MI ₁₁ ∈MI _22}, software component C ₁₁ is selected. The union U between the input message sets MI ₁₁ and MI ₂₂ is {(customer name = name), (password = password)}, and the product set N is {(customer name = name), (password = password)}. _Therefore , the matching score S ₂₂₁₁ is S ₂₂₁₁ = | N | / | U | = 1.

Moreover, the software component C _22, since a {MI ₁₂ ∈MI _22}, software component C ₁₂ is selected. The union U between the input message sets MI ₁₂ and MI ₂₂ is {(customer name = name), (password = password)}, and the product set N is {(customer name = name), (password = password)}. _Therefore , the match score S ₂₂₁₂ is S ₂₂₁₂ = | N | / | U | = 1.

On the other hand, the software components _{C 22, {MI 13 ∋MI 22} }, because it is _{{MI 14 ∋MI 22}, {} MI 15 ∋MI 22}, software components C _13, C _14, C ₁₅ is Not selected, the match scores S ₂₂₁₃ , S ₂₂₁₄ , S ₂₂₁₅ are not calculated.
For the software component C ₂₃ , the input message sets MI ₁₁ , MI ₁₂ , MI ₁₃ , MI ₁₄ , MI ₁₅ are included in the input message set MI ₂₃ , and the software components C ₁₁ , C ₁₂ , C _{13 are included.} , C ₁₄ and C ₁₅ are not selected, so the software components C ₁₁ , C ₁₂ , C ₁₃ , C ₁₄ and C ₁₅ are not selected.

Further, the process description of the system SY1 of FIG. 5 is input to the output message mapping apparatus 32 of FIG. 1 as the mapped process description 1, and the process description of the system SY2 is input to the output message mapping apparatus 32 as the mapping process description 2.
Then, the output message mapping device 32 refers to the message common knowledge base 41, and sets the output message set MO _2k of the software component C _2k (1 ≦ K ≦ 3) with the system SY2 and the software component C with the system SY1. The union set U and the intersection set N of the output message set MO _1l of _1l (1 ≦ l ≦ 5) are calculated, and the coincidence score S _2k1l = | N | / | U | between the software components C _2k and C _1l is calculated To do. This calculation is performed between all the software components C _2k and C _{1l of} both systems SY1 and SY2. Even if the message formats are different between the systems SY1 and SY2, if the output message contents match, the output message matches and the match score _S2k1l is calculated.

For example, the output message set MO ₁₁ software component C ₁₁ is {(customer name), (password)}, the output message set MO ₂₁ software component C ₂₁ is {(name), (password)}, Figure 6, it is understood that (customer name) and (name) are equivalent to each other, and (password) and (password) are equivalent to each other. Therefore, the union U between the output message sets MO ₁₁ and MO ₂₁ is {(customer name = name), (password = password)}, and the product set N is {(customer name = name), (password = password)}. Thus, the coincidence score S ₂₁₁₁ of the software component C _{11 with} respect to the software component C ₂₁ is S ₂₁₁₁ = | N | / | U | = 1.

Similarly, the union U between the output message sets MO ₁₂ and MO ₂₁ is {(customer name = name), (product purchase history), (product access history)}, and the product set N is {(customer name = name)}. _Therefore , the matching score S ₂₁₁₂ of the software component C _{12 with} respect to the software component C ₂₁ is S ₂₁₁₂ = | N | / | U | = 1/3.
Similarly, the union U between the output message sets MO ₁₃ and MO ₂₁ is {(customer name = name), (product purchase history), (product access history), (recommended product)}, and the product set N is {(customer name = name)}. Therefore, match score S ₂₁₁₃ software components C ₁₃ to the software component C ₂₁ is S ₂₁₁₃ = | a = 1/4 | N | / | U.

Similarly, the union U between the output message sets MO ₁₄ and MO ₂₁ is {(customer name = name), (recommended product), (other customer recommended product)}, and the product set N is {(customer name = name)}. since the match score S ₂₁₁₄ software components C ₁₄ to the software component C ₂₁ is S ₂₁₁₄ = | a = 1/3 | N | / | U.
Similarly, the union U between the output message sets MO ₁₅ and MO ₂₁ is {(customer name = name), (recommended product), (other customer recommended product), (search product)}, and the product set N is {(customer Name = name)}, the coincidence score S ₂₁₁₅ of the software component C _{15 with} respect to the software component C ₂₁ is S ₂₁₁₅ = | N | / | U | = 1/4.

Similarly, the union U between the output message sets MO ₁₁ and MO ₂₂ is {(customer name = name), (password = password)}, the product set N is {(customer name = name)}, and the software component C The coincidence score S ₂₂₁₁ of the software component C _{11 with} respect to ₂₂ is S ₂₂₁₁ = | N | / | U | = 1/2.
Similarly, the union U between the output message sets MO ₁₂ and MO ₂₂ is {(customer name = name), (product purchase history), (product access history)}, and the product set N is {(customer name = name)}. _Therefore , the matching score S ₂₂₁₂ of the software component C _{12 with} respect to the software component C ₂₂ is S ₂₂₁₂ = | N | / | U | = 1/3.

Similarly, the union U between the output message sets MO ₁₃ and MO ₂₂ is {(customer name = name), (product purchase history), (product access history), (recommended product)}, and the product set N is {(customer Name = name)}, the matching score S ₂₂₁₃ of the software component C _{13 with} respect to the software component C ₂₂ is S ₂₂₁₃ = | N | / | U | = 1/4.
Similarly, the union U between the output message sets MO ₁₄ and MO ₂₂ is {(customer name = name), (recommended product), (other customer recommended product)}, and the product set N is {(customer name = name)}. _Therefore , the matching score S ₂₂₁₄ of the software component C _{14 with} respect to the software component C ₂₂ is S ₂₂₁₄ = | N | / | U | = 1/3.

Similarly, the union U between the output message sets MO ₁₅ and MO ₂₂ is {(customer name = name), (recommended product), (other customer recommended product), (search product)}, and the product set N is {(customer Name = name)}, the matching score S ₂₂₁₅ of the software component C _{15 with} respect to the software component C ₂₂ is S ₂₂₁₅ = | N | / | U | = 1/4.
Similarly, the union U between the output message sets MO ₁₁ and MO ₂₃ is {(customer name = name), (password = password)}, the product set N is {(customer name = name)}, and the software component C The match score S ₂₃₁₁ of the software component C _{11 with} respect to ₂₃ is S ₂₃₁₁ = | N | / | U | = 1/2.

Similarly, the union U between the output message sets MO ₁₂ and MO ₂₃ is {(customer name = name), (product purchase history), (product access history)}, and the product set N is {(customer name = name)}. _Therefore , the matching score S ₂₃₁₂ of the software component C _{12 with} respect to the software component C ₂₃ is S ₂₃₁₂ = | N | / | U | = 1/3.
Similarly, the union U between the output message sets MO ₁₃ and MO ₂₃ is {(customer name = name), (product purchase history), (product access history), (recommended product)}, and the product set N is {(customer Name = name)}, the matching score S ₂₃₁₃ of the software component C _{13 with} respect to the software component C ₂₃ is S ₂₃₁₃ = | N | / | U | = 1/4.

Similarly, the union U between the output message sets MO ₁₄ and MO ₂₃ is {(customer name = name), (recommended product), (other customer recommended product)}, and the product set N is {(customer name = name)}. _Therefore , the match score S ₂₃₁₄ of the software component C _{14 with} respect to the software component C ₂₃ is S ₂₃₁₄ = | N | / | U | = 1/3.
Similarly, the union U between the output message sets MO ₁₅ and MO ₂₃ is {(customer name = name), (recommended product), (other customer recommended product), (search product)}, and the product set N is {(customer (Name = name), (search product)}, the matching score S ₂₃₁₅ of the software component C _{15 with} respect to the software component C ₂₃ is S ₂₃₁₅ = | N | / | U | = 1/2.

Next, when the input message mapping device 31 extracts a software component having an input message set in a correspondence relationship between the systems SY1 and SY2, the input message mapping device 31 outputs the extraction result to the combination analysis device 33. Further, when the output message mapping device 32 extracts a software component having an output message set in a correspondence relationship between the systems SY1 and SY2, the output message mapping device 32 outputs the extraction result to the combination analysis device 33. The combined analysis device 33, among the software components C _1j system SY1, its input message set MI _1j software component C _1j subsumed input message set MI _2i software component C _2i where the system SY2 was a starting point, the end point of the software component C _1l which part of the output message set MO _1l software component C _1l system SY1 coincides with part of the output message set MO _2k software component C _2k with system SY2 Are extracted as reusable processes.
Then, the user presentation device 34 combines the extracted sub-processes in descending order of the matching score of both the starting and ending software components, ranks these sub-processes, and presents them to the user terminal 6.

9 to 13 are block diagrams showing examples of reusable processes extracted by the process mapping apparatus of FIG.
In FIG. 9, between the input message sets MI _1j and MI _2i of the systems SY1 and SY2, the matching scores S ₂₁₁₁ , S ₂₁₁₂ , S ₂₂₁₁ , S of the software components C ₁₁ , C ₁₂ with respect to the software components C ₂₁ , C _{22 2212} is 1, but between the output message sets MO _1l and MO _{2k of} the systems SY1 and SY2, the matching score S ₂₁₁₁ of the software component C _{11 with} respect to the software component C ₂₁ is 1, which is the largest. ing. For this reason, the subprocess (C ₁₁ ) whose start and end points are composed of the software component C ₁₁ and the sub process (C ₂₁ ) whose start and end points are composed of the software component C ₂₁ are extracted as the first ranking.

In FIG. 10, between the input message sets MI _1j and MI _2i of the systems SY1 and SY2, the matching scores S ₂₁₁₁ , S ₂₁₁₂ , and S ₂₂₁₁ of the software components C ₁₁ and C ₁₂ with respect to the software components C ₂₁ and C ₂₂ are obtained. , S ₂₂₁₂ is 1, but the matching score S ₂₂₁₁ of the software component C _{11 with} respect to the software component C ₂₂ is ½ between the output message sets MO _1l , MO _{2k of} the systems SY1, SY2. Second largest. Therefore, as a ranking position 2, a sub-process start point and the end point consists of software components C ₁₁ (C _11), subprocess (C ₂₁ to start and end point consists of software components C _21, C ₂₂ respectively, C ₂₂ ) Is extracted.

In FIG. 11, between the input message sets MI _1j and MI _2i of the systems SY1 and SY2, the matching scores S ₂₁₁₁ , S ₂₁₁₂ , and S ₂₂₁₁ of the software components C ₁₁ and C ₁₂ with respect to the software components C ₂₁ and C ₂₂ are obtained. S ₂₂₁₂ is 1, but the matching score S ₂₃₁₁ of the software component C _{11 with} respect to the software component C ₂₃ is ½ between the output message sets MO _1l and MO _{2k of} the systems SY 1 and SY ₂ . Second largest. For this reason, as a ranking second place, a subprocess (C ₁₁ ) whose start and end points are software components C ₁₁ and a sub-process (C ₂₁ and C ₂₂ whose start and end points are software components C ₂₁ and C ₂₃ , respectively). , C ₂₃ ) are extracted.

In FIG. 12 and FIG. 13, between the input message sets MI _1j and MI _2i of the systems SY1 and SY2, the matching scores S ₂₁₁₁ and S ₂₁₁₂ of the software components C ₁₁ and C ₁₂ with respect to the software components C ₂₁ and C ₂₂ are obtained. , S ₂₂₁₁ , S ₂₂₁₂ are both 1, but between the output message sets MO _1l , MO _{2k of} the systems SY1, SY2, the match score S ₂₃₁₁ of the software component C _{15 with} respect to the software component C ₂₃ is 1 / 2, which is the second largest. Therefore, as the second ranking, as shown in FIG. 12, the sub-process (C ₁₁ , C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ) whose starting and ending points are respectively composed of software components C ₁₁ and C ₁₅ , Sub-processes (C ₂₁ , C ₂₂ , C ₂₃ ) whose starting and ending points are software components C ₂₁ and C ₂₃ are extracted. Further, as shown in FIG. 13, the ranking is second, and the starting point and the ending point are sub-processes (C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ) each consisting of software components C ₁₂ and C ₁₅ , and the starting point and ending point are Sub-processes (C ₂₁ , C ₂₂ , C ₂₃ ) each consisting of software components C ₂₁ , C ₂₃ are extracted.

Then, when the sub-process of FIGS. 9 to 13 is obtained as a higher candidate of a process that can be reused, the user presentation device 34 presents the sub-process of FIGS. 9 to 13 to the user terminal 6.
When the sub-process ranked by the process mapping server 3 is indicated, the user terminal 6 verifies the reuse value of the sub-process and returns the verification result to the process mapping server 3.

For example, the administrator of the system SY2 determines that these functions are the same for the subprocess (C ₁₁ ) of the system SY1 and the subprocess (C ₂₁ ) of the system SY2 in FIG. 9 and can be reused. can do. As for the sub-process of the sub-process (C ₁₁₎ and system SY2 system SY1 in FIG. _{10 (C 21, C 22)} , features a subprocess system SY2 (C _21, C ₂₂₎ remain the same system SY1 is regarded as possible to shorten the subprocess (C _11), it can be determined that is reusable. As for the sub-process of the system SY1 in FIG. 11 subprocesses (C ₁₁₎ and system _{SY2 (C 21, C 22,} C 23) , the system sub-processes of the system _{SY2 (C 21, C 22,} C 23) Although it can be shortened to the sub-process (C ₁₁ ) of SY1, since the search function is lost, it can be determined that it cannot be reused. Further, the sub-process of the system SY2 in FIG. 12 is the sub-process of the system SY2 (C ₁₁ , C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ) and the sub-process of the system SY 2 (C ₂₁ , C ₂₂ , C ₂₃ ). _{(C 21, C 22, C} 23) can be shortened to a sub-process of the system _{SY1 (C 11, C 12,} C 13, C 14, C 15), since the additional attenuation to the commodity recommendation function, the reuse value is greater Judgment can be made. The sub-process of the system SY1 in FIG. _{13 (C 12, C 13,} C 14, C 15) and the sub-processes of the system _{SY2 (C 21, C 22,} C 23) , the sub-process of the system SY2 (C ₂₁ , C ₂₂ , C ₂₃ ) can be shortened to the sub-processes (C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ) of the system SY1, and the product recommendation function is added, so that it can be determined that the reuse value is great.

In this way, the designer uses the sub-processes (C ₁₁ , C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ) of the system SY 1 without examining the specifications of the systems SY 1 and SY 1. It becomes possible to reconstruct the processes (C ₂₁ , C ₂₂ , C ₂₃ ), and it is possible to realize flexible system linkage by reusing services while reducing the burden on manpower. In addition, since the software components having a correspondence relationship are searched for all the software components constituting the two processes, process mapping can be realized regardless of the granularity of the processes. For this reason, even for a process including branching and synchronization, process mapping can be applied as long as it can be expressed as a simple continuous process by increasing or decreasing the granularity.

Further, sub-process (C ₁₁₎ and of the system SY2 subprocess system SY1 in FIG. 9, which is determined to be reused in the user terminal 6 (C _21), sub-process of the system SY1 in FIG. 10 (C ₁₁ ) And sub-processes (C ₂₁ , C ₂₂ ) of the system SY 2, sub-processes (C ₁₁ , C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ) and sub-processes (C ₂₁ , C ₁₅ ) of the system SY 2 in FIG. ₂₂ , C ₂₃ ) and the sub-process (C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ) of the system SY 1 and the sub-process (C ₂₁ , C ₂₂ , C ₂₃ ) of the system SY 2 in FIG. To the process common knowledge base 51.

FIG. 14 is a block diagram showing a specific example of a reusable process between the systems SY1 and SY2 registered in the process common knowledge base 51 of FIG.
In FIG. 14, the process common knowledge base 51 includes a subprocess (C ₁₁ ) of the system SY 1, a sub process (C ₂₁ ) of the system SY 2, a sub process (C ₁₁ ) of the system SY 1, and a sub process (C ₁₁ ) of the system SY 2. ₂₁ , C ₂₂ ), system SY1 sub-process (C ₁₁ , C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ), system SY 2 sub-process (C ₂₁ , C ₂₂ , C ₂₃ ) and system SY 1 sub-process ( C ₁₂ , C ₁₃ , C ₁₄ , C ₁₅ ) and the sub-processes (C ₂₁ , C ₂₂ , C ₂₃ ) of the system SY 2 are registered to be reusable. Note that the process common knowledge base 51 can be composed of, for example, a text file using a language that hierarchically describes conceptual relationships such as XML and XML-based ontology description language OWL.

  Then, by using the process common knowledge base 51 in which the reusable processes between the systems SY1 and SY2 are registered, the reuse can be performed while using the semantic description given by the past process mapping. It is possible to search for possible processes. Therefore, when a new system design is performed, a plurality of reusable process candidates can be obtained as search results of the process common knowledge base 51 by issuing a certain input message and output message as an inquiry. , Can speed up the system design.

FIG. 15 is a block diagram illustrating a method for acquiring a reusable process between the systems SY1 and SY2 using the process common knowledge base 51 of FIG.
In FIG. 15, it is assumed that the system designer submits an inquiry of {input message (name, password), output message (product search)} from the user terminal 6 to the process knowledge base server 5. When the process knowledge base server 5 receives the inquiry from the user terminal 6, the process knowledge base server 5 searches the process common knowledge base 51 for a reusable process corresponding to the inquiry. Note that the process mapping method described above can be used as a method for searching the process common knowledge base 51 for a reusable process corresponding to the query from the user terminal 6. Then, when only one reusable process is retrieved from the process common knowledge base 51, the reusable relationship of the first retrieved process is traced using the semantic description that the process is reusable. The process common knowledge base 51 is searched for other reusable processes.

  The present invention reuses an existing business process without building a new business process from scratch, with a service that implements a business process such as credit inquiry, order processing, and inventory check as a unit. By connecting a series of business processes such as product development, procurement, production, sales and logistics, it is possible to quickly and flexibly build a value chain according to a business model designed based on management strategy. .

It is a block diagram which shows schematic structure of the process mapping apparatus which concerns on one Embodiment of this invention. It is a block diagram which shows the relationship between the system and process which concern on one Embodiment of this invention. It is a block diagram which shows the process sequence of the process mapping apparatus of FIG. It is a block diagram which shows the specific example of the process of system SY1 which concerns on one Embodiment of this invention. It is a block diagram which shows the specific example of the process of system SY2 which concerns on one Embodiment of this invention. It is a figure which shows the equivalence relationship of the data contained in the message between the systems SY1 and SY2 which concern on one Embodiment of this invention. It is a figure which shows the equivalence relationship of the message type between the systems SY1 and SY2 which concern on one Embodiment of this invention. It is a figure which shows an example of the message type which concerns on one Embodiment of this invention. It is a block diagram which shows an example of the process with the possibility of reuse extracted by the process mapping apparatus of FIG. It is a block diagram which shows the other example of the process with the possibility of reuse extracted in the process mapping apparatus of FIG. It is a block diagram which shows the other example of the process with the possibility of reuse extracted in the process mapping apparatus of FIG. It is a block diagram which shows the other example of the process with the possibility of reuse extracted in the process mapping apparatus of FIG. It is a block diagram which shows the other example of the process with the possibility of reuse extracted in the process mapping apparatus of FIG. It is a block diagram which shows the specific example of the process which can be reused between the systems SY1 and SY2 registered into the process common knowledge base 51 of FIG. It is a block diagram which shows the acquisition method of the reusable process between systems SY1 and SY2 using the process common knowledge base 51 of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Mapping process description 2 Mapping process description 3 Process mapping server 4 Message knowledge base server 5 Process knowledge base server 6 User terminal 31 Input message mapping apparatus 32 Output message mapping apparatus 33 Combination analysis apparatus 34 User presentation apparatus 41 Message common knowledge base 51 Process common knowledge base V1-V4 Software component providing server C1-C7 Software component MI1-MI7, MI11-MI15 Input message MO1-MO7, MO11-MO15 Output message P Process SY1, SY2 System

Claims (10)

An input message mapping means for extracting software components having a correspondence relationship between processes based on input messages of software components constituting the process;
Output message mapping means for extracting software components in a correspondence relationship between the processes based on output messages of software components constituting the process;
A process mapping apparatus comprising: subprocess extraction means for extracting subprocesses having a software component extracted by the input message mapping means and the output message mapping means as a starting point and an ending point, respectively.   The sub-process extraction means starts from a software component whose input message set of the software component of the first system is included in the input message set of the software component of the second system, and outputs the software component of the first system 2. The process mapping apparatus according to claim 1, wherein a sub-process whose destination is a software component whose part of the message set matches a part of the output message set of the software component of the second system is extracted. The sub-process extraction means includes
Ranking the sub-processes extracted by the sub-process extracting means based on the proportion of the components of the input message set and the output message set of the software component that match between the first system and the second system. The process mapping apparatus according to claim 2, further comprising a combination analysis unit. Presenting means for presenting sub-processes ranked by the combination analysis means;
Receiving means for receiving a selection result of the subprocesses determined to be reusable among the presented subprocesses;
The process mapping apparatus according to claim 3, further comprising a process common knowledge base for registering a correspondence relationship of the sub-processes determined to be reusable as a reusable process. A message common knowledge base for accumulating correspondences of messages distributed between processes is further provided.
The input message mapping means extracts a software component having a correspondence relationship between processes after matching the input message format distributed between the software components based on the reference result of the message common knowledge base. And
The output message mapping means matches the output message format distributed between the software components based on the reference result of the message common knowledge base, and extracts software components having a correspondence relationship between processes. The process mapping apparatus according to claim 1, wherein the process mapping apparatus includes: Process common knowledge base with registered reusable process correspondence,
Input / output message receiving means for receiving input messages and output messages of software components constituting the process;
Process search means for searching the process common knowledge base for processes starting and ending with software components having a corresponding relationship with the input message and the output message, respectively;
A process mapping management apparatus comprising: a process search unit that searches for a reusable process by tracing a correspondence relationship of processes searched by the process search unit on the process common knowledge base. Inputting a mapping process description and a mapped process description into a process mapping server;
An input message having a correspondence relationship between the mapping process description and the mapped process description after matching the input message format distributed between the software components described in the mapping process description and the mapped process description Extracting a software component having
An output message having a correspondence relationship between the mapping process description and the mapped process description after matching the output message format distributed between the software component described in the mapping process description and the mapped process description Extracting a software component having
Extracting sub-processes starting and ending with software components having an input message and an output message corresponding to each other between the mapping process description and the mapped process description;
The extraction is performed based on a ratio in which the component of the input message set of the source software component and the component of the output message set of the destination software component match between the mapping process description and the mapped process description, respectively. And a step of ranking and presenting the processed sub-processes. Presenting the ranked sub-processes;
The process mapping method according to claim 7, further comprising: selecting a subprocess determined to be reusable among the presented subprocesses as a reusable process.   9. The process mapping method according to claim 8, further comprising the step of creating a database of correspondence relationships of the sub-processes determined to be reusable as reusable processes. Extracting a software component having a correspondence relationship between processes based on input messages of software components constituting the process;
Extracting a software component having a correspondence relationship between the processes based on an output message of the software component constituting the process;
A process mapping program for causing a computer to execute a step of extracting subprocesses having a software component extracted by the input message mapping means and the output message mapping means as a starting point and an ending point, respectively.

Images