JPH03235124A - Rule base constructing system - Google Patents

Abstract

PURPOSE:To simply and accurately execute the rule base constructing work by using a general rule description table of a natural language expression for the rule base construction, and utilizing each rule as a document. CONSTITUTION:The system is provided with a general rule description table 4 for showing the processing contents in which numerical data and a processing object are described as parameters by a natural language expression, and an object classification item table group 5. In such a state, a rule for verifying data is defined by combining freely the designated processing object and each line of the processing contents table, the defined rule is contained in a rule base 3 by a rule base managing part 2, and by executing the designation of the object and the designation of the processing contents by an interactive processing part 1, its rule is contained in the rule base one after another, and the number of rules for verifying the data can be made innumerably theoretically. In such a way, the rule base constructing work can be executed simply and accurately.

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention is used in the CAD field to input verification rules into a computer in a form that can be processed by the computer for automatically verifying whether or not designed data is correct. It is concerned with a rule base construction method for the purpose of this invention, and is particularly aimed at simplifying and clarifying rule definitions.

[Conventional technology]

Figure 3 is, for example, r PIXEL, November 1985, No. 3.
8 P173-P178. This is a system configuration diagram explaining the rule base construction method described in ``Knowledge Engineering Approach of CAD/CAM''. 3) is a rule base management unit that manages generation, modification, and deletion.

In conventional rule base construction methods, for example, when verifying whether certain design data satisfies the design rules, the dialog processing unit (1) is used to construct design rules in a format that can be understood by a computer. The rule is written in a rule description language, and the rule definition contents of the rule base management section (2) are stored in the rule base (3) to construct the rule base.

[Problem to be solved by the invention]

However, in the conventional rule base construction method, the rules are defined using the rule description language as described above, and therefore the operator needs to understand the rule description language in advance. In addition, there is a problem in that writing rules using a rule description language requires a lot of human effort and is prone to definition errors.

This invention was made in order to solve the above-mentioned problems. Instead of using a rule description language for rule definition, it is possible to define rules by combining a table of the rule processing contents and the objects to be processed. The purpose of this invention is to realize a rule base construction method that can create, update, and delete rules, and store rules by registering them as one rule.

[Means to solve the problem]

A rule base construction method according to the present invention is a rule base construction method that converts rules serving as constraints into a format that can be processed by a computer when verifying whether design data stored in a computer satisfies constraints. , To construct the rule base, input is in the form of a table, and that table is used as a general-purpose rule description table expressed in natural language to specify the processing function of the rule, and design data as the object to be processed for verification. An interaction processing unit that refers to these tables and implements interaction processing with workers, and a rule base that manages the creation, modification, and addition of rules to the rule base. The system includes a management section and a rule base that stores a group of rules based on the combination of the above-mentioned tables as one rule, and constructs a rule base by combining the tables.

[Effect]

In this invention, data verification rules are defined by freely combining the specified processing object and each row of the processing content table, and the defined rules are stored in the rule base by the base management unit that manages the rules. By specifying the above-mentioned object and processing content using the interaction processing unit, the rules are stored in the rule base one after another, and it is possible to theoretically create an infinite number of data verification rules. can.

〔Example〕

Hereinafter, this invention will be explained based on the drawings.

FIG. 1 is a system configuration diagram illustrating a rule base construction method showing an embodiment of the present invention. In the figure, (1
) to (3) have the same effect as explained in the conventional rule base construction method, and (4) shows the processing content described using numerical data and processing objects as parameters in natural language expression. General rule description table, (5)
is an object classification item table group, which is a table group that classifies processing objects from various viewpoints and shows each item thereof.

FIG. 2 is a specific example of the general-purpose rule description table (4), which does not limit the objects to be processed.
The object to be processed is also described as a parameter.

3(a) to (e) are specific examples of each classification item table in the object classification item table group (5), and FIG. 3(a) is a classification item table (5a) which is a major classification. ), (b) is a classification item table (5b) that shows the type of parts in the second row of the classification item table (5a), and (C) is a classification item table (5C) that also shows the thermal attributes of the parts. , the same figure (d) shows the classification item table (5d) showing the usage of the pattern in the fourth row of the classification item table (5a), and the same figure (e)
1 and 2 respectively show classification item tables (5e) representing the directions of parts.

Furthermore, Figure 4 shows an example of rule definition using the tables shown in Figures 2 and 3 above, and combines tables (4), (5a) to (5e) shown in Figures 2 and 3. ,
This is a specific example of a rule definition in which a numerical value and execution/non-execution of the rule are set for each row of the combination. Figure 5 shows a specific example of the contents of rule base (3).
a) is the route table of the rule base (3) that is referenced first, and includes the row number of the general rule description table (4), the start and end rows of the reference rows of the object instruction table (3b), and numerical values. The start and end rows of the reference rows of the table (3d) and flags for execution/non-execution of the rule are described.

(3b) is an object instruction table in which the start and end rows of the reference rows of the object classification table (3c) are described; (3c)
is an object classification table in which the table number and row number of each table in the object classification item table group (5) are described, and (3d) is a numerical table in which numerical data is stored.

The operation of the rule base construction method configured as described above will be explained with reference to the flowchart shown in FIG. The operator first selects one row of processing contents from the general rule description table (4) using the dialog processing section (1) (step Sl, 52). Next, in order to specify the processing object used in the selection row of the general rule description table (4), select how many car tables from the object classification item table group (5) (step S3.54). , select one row of classification items in each selected table (step S5).

The above process is performed for the number of processing objects being used (steps 56 and 57), and then the numerical value and the execution/non-execution status of the rule are input (steps S8 and 59).
One rule is defined by this, and a plurality of rules are defined by the above operations (step 510).

The defined rules are stored in a rule base (3) under the management of a rule base management section (2).

When modifying a defined rule, use the dialog processing unit (1) to show the contents of the rule base (3), for example the contents of the rule definition example shown in diagram N4, and specify the value to be modified, and then use the same method as the rule definition. If you want to delete it, specify the line number in the rule definition example above. For example, when defining the first row of the above rule definition example, first select the first row of the general rule description table (4) shown in FIG.
In order to specify object A therein, the second row of the classification item table (5a) shown in FIG. 3(a), the third row of the classification item table (5b) shown in FIG. 3(b), and Figure 3 (c
) Select the first row of the classification item table (5C) shown in
In order to specify object B, the above classification item table (5
The second row of a) and the second row of the classification item table (5b) above
After selecting the row and the second row of the classification item table (5C) above, input lO as numerical data to execute the rule.
This is done by turning non-execution into execution. The defined rules are stored in the form of tables (3a) to (3d) shown in FIG. 5 under the management of the rule base management section (2).

In the above embodiment, a case was described in which the rule base construction method according to the present invention is used to construct verification rules for design data, but the target world is known in advance, and the processing contents and If the object to be processed can be organized as a table, it can also be used for building a rule base.

〔Effect of the invention〕

As explained above, this invention has the advantage that each rule can be used as a document by using a general-purpose rule description table expressed in natural language for building a rule base. In addition, a verification system using a rule base based on this rule base construction can utilize complex rules for each object to be processed, so it has the effect of preventing small design mistakes and supporting high-quality design.

[Brief explanation of drawings]

FIG. 1 is a system configuration diagram of rule base construction using a rule base construction method showing an embodiment of this invention, FIG. 2 is an explanatory diagram of a specific example of the general-purpose rule description table shown in FIG. 1, and FIG. Figure 1 is an explanatory diagram of a concrete example of the object classification item table group, Figure 4 is an explanatory diagram of an example of rule definition using the tables in Figures 2 and 3, and Figure 5 is a concrete example of a rule base. FIG. 6 is a flow chart of rule base construction, and FIG. 7 is a system configuration diagram of rule base construction using a conventional rule base construction method. In the figure, (1) is the dialogue processing unit, (2) is the rule base management unit, (3) is the rule base, (4) is the general rule description table, (5) is the object classification item table group, and (3a) is the is a rule-based route table, (3b)
is a rule-based object instruction table, (3c) is a rule-based object classification table, and (3d) is a rule-based numerical value table. Note that the same reference numerals in the figures indicate the same or corresponding parts.

Claims (1)

[Claims] When verifying whether design data stored in a computer satisfies constraints, a rule base construction method that converts the rules that serve as constraints into a format that can be processed by the computer,
To build a rule base, input is in table format, and that table is used to specify a general-purpose rule description table expressed in natural language to specify the processing function of the rule, and design data to be processed for verification. An interaction processing unit that refers to these tables and implements interaction processing with workers, and a rule base management unit that manages the generation, modification, and addition of rules to the rule base. A rule base construction method comprising: a rule base that stores a group of rules based on a combination of the above-mentioned tables as one rule, and constructs a rule base using a combination of tables.