JPH03264250A - Production schedule planning device - Google Patents

Abstract

PURPOSE:To maintain a rule even by a user, not knowing rule expression inherent to a system, without correcting an inference method by providing a production schedule planning means to effect inference of a production schedule and plan the schedule by using a pattern rule corrected by a room maintenance means. CONSTITUTION:A specified pattern rule needing addition and variation is read from an acknowledge base by means of rule maintenance means 12 to display it on a CRT 11b of a user interface 11. Thereafter, by adding and varying a necessary variable part as the displayed pattern rule is watched, an acknowledge base is corrected. Meanwhile, a production schedule planning means 16 previously handles a production plan as one rule at each of patterns 1, 2,...n and effects inference on the assumption that the given spot of a pattern rule has a variable part. Therefore, a production schedule can be inferred and planned by using a corrected pattern rule by the rule maintenance means 12 without especially correcting an inference method.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention is used when an expert system is used to formulate a production plan for, for example, paper production or turbine production. The present invention relates to a production planning device, and particularly to a production planning device that can easily add and modify rules.

(Prior Art) This type of production planning device has a knowledge base 1 that stores rules consisting of conditions and conclusions, and a system that receives order data and stores it in the knowledge base 1, as schematically shown in FIG. An inference engine 2 that creates a production plan based on rules, a user interface 3 that inputs necessary data from a keyboard 3a and displays necessary processing data on a CRT 3b, and a and a rule maintenance means 4 for modifying the knowledge base 1 in response to instructions.

Therefore, in this device, when rules stored in the knowledge base 1 are added or changed, the inference rules and analogy methods of the inference engine 2 are used to draw a conclusion based on the added or changed rules. A fix was being done.

(Problem to be Solved by the Invention) Therefore, in the production planning device as described above, when the knowledge base 1 is modified by adding or changing rules, the inference engine 2 and the rule maintenance means 4 are also modified accordingly. However, it is extremely difficult for an operator who does not know the rule expression method and inference method unique to the expert system to make the above modifications. In particular, most of the people who actually handle expert systems are field operators, and although they know what kind of maintenance should be done for the current knowledge, they do not know the system-specific rules expression method or inference method. It is not practical because there are many

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a production planning device that allows even a person who does not know the system-specific rule expressions to maintain rules without modifying the inference method etc. do.

[Structure of the Invention] (Means for Solving the Problems) The present invention has been made in view of the above circumstances, and includes a plurality of pattern rules having variable parts classified in advance to suit various production processes. A knowledge base to be stored, a rule maintenance means for modifying the knowledge base by adding or changing a variable part of a pattern rule required by this knowledge base, and a pattern rule modified by the rule maintenance means. This configuration includes a production planning means for inferring and formulating a production plan.

(Function) Therefore, by taking the above-described measures, the present invention adds information to the knowledge base using the rule maintenance means.
A specific pattern rule characterized by a change is read out and displayed on a display means. Thereafter, the knowledge base is modified by adding or changing necessary variable portions while looking at the displayed pattern rules. On the other hand, in the production planning means, each pattern is treated as one rule in advance, and the inference method is based on the assumption that the pattern rule has a variable part at a predetermined location, so the inference method etc. is particularly modified. It is possible to infer and formulate a production plan using pattern rules modified by the rule maintenance means without having to do so.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings. FIG. 1 shows the basic configuration of a production planning device. In the figure, 11 inputs necessary data manually from a keyboard 11a, and further interacts with various components described later.
In addition to various special editors, this is a user interface that displays stored data and processed data on a CRTllb. It has a function of reading a specific pattern rule stored in the CRT13 and displaying it on the CRTllb. This pattern rule knowledge base 13 stores a plurality of pattern rules that are classified in advance to suit the production process. The rule maintenance means 12 displays predetermined pattern rules and adds and changes only the data within the sono pattern rules without being aware of the rule expression for each pattern rule in the pattern rule knowledge base 13.

In other words, it has a correction function.

Reference numeral 14 denotes an order data processing means, which stores order data inputted from the user interface 11 and serves as the basis for production in an order data memory 15, and reads the order data from the order data memory 15 as necessary. CRT
Display on I 1 b.

Note that this order data includes, for example, period, delivery date, materials, etc.

16 is a production planning means, specifically an inference engine is the main body, which treats each pattern rule of the knowledge base 13 as one rule, and can make inferences by considering the variable part of the rule in advance. The system is programmed to perform inferences based on order data and formulate a production plan.

The production plan data created by the production plan planning means 16 is stored in the production plan data memory 17. 1
8 is a production plan display processing means for displaying production plan data.

Next, FIG. 2 is a diagram showing the simplest hardware configuration for realizing the apparatus of the present invention. In the figure, a computer 21 executes programs of a rule maintenance means 12, an order data processing means 14, a production planning means 16, and a production plan display processing means 18 stored in a memory 22. Reference numeral 23 denotes an external storage device, which stores knowledge bases, order data, production plan data, and the like.

Next, in order to explain the operation of the apparatus configured as described above, an explanation will be given by taking as an example the formulation of a production plan for a plant that produces a plurality of different products using a plurality of manufacturing machines. First, when we focus on the binary relationships between two successive products, that is, the binary relationships between the products, most of them can be classified into the following seven patterns. That is, the seven patterns are, for example, the following patterns.

■ Immediately assigned Product 2 is manufactured immediately after Product 1.

■Continuous allocation Product 2 is manufactured immediately before or after Product 1.

■、Ensuring spacing Product 1 and Product 2 are manufactured with a certain interval or more between them.

■, Assigned after a certain period of time Product 2 is manufactured after a certain period of time of Product 1.

■、Cabinet allocation for a certain period of time After product 1, product 2 is manufactured within a certain period of time.

(2) Product 1 and Product 2, which are prohibited from parallel allocation, are prohibited from being manufactured during the same period. (2) Product 1, which is designated for an allocation period, is manufactured within a specific period.

Therefore, as described above, in this apparatus, the rule of the knowledge base 13 is to classify and store information into a plurality of patterns as described above.

Then, when the operator performs rule maintenance processing,
When a command to read pattern "1" is issued from the user interface 11, the rule maintenance means 12 receives the command and reads the corresponding pattern "1" from the knowledge base 13.
1" is read out and displayed on the screen of CRTllb. 31 in FIG. 3 shows an example of the CRT screen display of the immediately after assignment rule. Therefore, from the immediately after assignment rule in this display example 31, (a) Machine No. 2, B should be manufactured immediately after Machine No. 1, A.

(b) Machine No. 1, D should be manufactured immediately after machine No. 2, C.

It is clear that there are two rules, and among the pattern rules of Display Example 31, the machine name and product name can easily be made into variable parts.

Therefore, when the operator changes the product name of the rule to E after displaying the assignment rule immediately after pattern "1" on the CRT screen by the rule maintenance means 12,
The variable part of the pattern "1°", 1 product, is changed to E. As a result, display example 32 in Figure 3 is displayed on the CRT screen, so the operator confirms the changed state and then changes the knowledge The pattern on the base 13 is stored at 1°.

Here, the production planning means 16 treats each pattern rule as one rule in advance, and performs inference assuming that the machine name and product name are changed parts, so that the changed parts in the pattern rule Even if additions or changes are made, a production plan can be drawn up based on the modified pattern rules and order data in the knowledge base 13 without having to modify the inference method or the like at all. 33 in Figure 3
FIG. 2 is a diagram showing an example of a production plan that has been created.

Therefore, according to the configuration of the embodiment as described above, a plurality of pattern rules classified in advance to suit every production process are stored in the knowledge base 13, and a variable part is provided in advance for each pattern rule. With this, the operator can read out and display the desired pattern rule using the rule maintenance means 12, and modify the knowledge base 13 by adding or changing the variable part, and the operator can modify the inference engine while being aware of the method of expressing the rule. There's no need to. This means that the rules in the knowledge base 13 can be easily maintained even by an on-site operator who does not particularly know the rule expressions and inference methods unique to the expert system, and who is not very skilled.

Although we have described an example of producing different products using multiple machines, even when producing one product using multiple machines, or when producing multiple products using one machine while exchanging tools, it is possible to If the pattern rules are classified into pattern rules and stored in the knowledge base 13, and variable portions are provided in advance in these pattern rules, it can be applied in the same manner as in the above embodiment. In addition, the present invention includes various other methods without departing from the gist thereof.
It can be modified and implemented.

[Effects of the Invention] As explained above, according to the present invention, even a person who does not know the unique rule expressions can easily modify only the knowledge-based rules without modifying the inference method etc. to create a production plan. We can provide a production planning device that can create a production plan.

[Brief explanation of drawings]

1 to 3 are shown to explain one embodiment of the present invention, FIG. 1 is a diagram showing the basic configuration of the device of the present invention, and FIG. 2 is a diagram showing the basic configuration of the device of the present invention. FIG. 3 is a diagram showing the simplest hardware configuration, FIG. 3 is a diagram schematically representing everything from rule maintenance to inference processing, and FIG. 4 is a schematic configuration diagram of a conventional general production planning device. 11... User interface, 12... Rule maintenance means, 13... Knowledge base, 14...
Order data processing means, 15...Order data memory, 1
6... Production planning means, 17... Production plan data memory, 18... Production plan display processing means.

Claims (1)

[Claims] A knowledge base that stores a plurality of pattern rules with variable parts that are classified in advance to suit various production processes, and a knowledge base that stores the variable parts of the pattern rules required by this knowledge base by adding and changing the variable parts. 1. A production planning device comprising: rule maintenance means for modifying a base; and production planning means for inferring and formulating a production plan using pattern rules modified by the rule maintenance means.