JPH07200662A - Experiment plan production support system having design knowledge base - Google Patents

Abstract

PURPOSE:To provide an experiment plan production support system which can know a process where a newly designed product has a defect and can previously and effectively produce the plans for analyses and experiments. CONSTITUTION:The knowledges including the corresponding relation between the defects of existing products and the occurring factors of these defects, the correlation of the defect occurring factors, and the corresponding relation between the defect occurring factors and a design parameter are inputted through a new knowledge input part 2 and storey in a design knowledge base 4. Meanwhile the specifications of a new product are inputted through a requested specification input part 1, and the knowledge of the products having the similar specifications to the product inputted through the part 1 are extracted out of the base 4 by a defect occurring process generation data extracting part 3. A defect occuring process is generated by a defect occuring process generating part 5 based on the knowledge extracted at the part 5. The corresponding relation and the correlation generated at the part 5 are displayed at a defect occurring process display part 6. If the knowledge extracted at the part 5 is not included in the base 4, the more detailed specifications of a product are received and inputted through a feature data input part 7.

Description

Detailed Description of the Invention

[0001]

[Industrial field of application] An experiment that has a design knowledge base that can efficiently store design knowledge required when developing a new product for an LSI package, and enables beginners who do not have design knowledge to obtain the defect occurrence process of a new product. Regarding the planning support system.

[0002]

2. Description of the Related Art There are many expert systems that support design based on a design knowledge database in which the knowledge and know-how of designers who design products are registered. For example, an expert system developed for the purpose of optimizing the shape of products and molds at the designing stage by using injection molding CAE is a “mold technology” Volume 6 No. 7 (published by Nikkan Kogyo Shimbun). June 1991). The design knowledge base of this system has accumulated knowledge about defects during molding, their causes, and countermeasures.

Further, Japanese Patent Laid-Open No. 4-77219 discloses that
A technique that includes a knowledge base, a neural network unit, an experiment planning unit, and a molding defect diagnosing unit and predicts and diagnoses molding defects and obtains optimum molding conditions based on an experimental plan is described.

Further, Japanese Laid-Open Patent Publication No. 3-122771 discloses a knowledge storage unit, a design data storage unit, and an inference processing unit which store design knowledge obtained from past experience in the form of a rule including a condition section and a conclusion section. The technology relating to an apparatus configured to automatically set design parameters, advise a designer, predict and diagnose a failure, and support design work is described.

In the conventional design knowledge base, it takes time to infer when the amount of data is large, and it is difficult to confirm the knowledge system by updating the knowledge.

[0006]

Each designer has knowledge of past designs when designing a product, but these are data that each person has and cannot be used freely. Further, in order to register the knowledge and confirm whether there is any inconsistent knowledge, it is necessary to retain data so that the entire design knowledge system can be easily confirmed. In addition, since the cause of failure is a combination of multiple phenomena that lead to failure, unless the designers' knowledge is to be interpreted together, what kind of analysis / experiment, etc. must be conducted to set the design plan. I don't know.

On the other hand, the design knowledge of product design varies depending on the specifications of the product, the amount of data is large, and it takes time to generate a defect occurrence process. However, after the required specifications are decided, in order to plan the experiment / analysis and perform the detailed design, it is necessary to generate the failure occurrence process predicted in a short time.

Further, since it takes both cost and time to carry out an experiment, it is necessary to carry out a simulation by a numerical analysis or the like in advance in order to select a defect potential which is highly likely to cause a defect in the defect generation process. is there.

A first object of the present invention is to store all the knowledge of each designer.

A second object of the present invention is to retain knowledge so that the entire design knowledge can be easily confirmed.

A third object of the present invention is to generate a defect occurrence process of a new product in a short time.

A fourth object of the present invention is to compress design knowledge data.

A fifth object of the present invention is to make a final determination of the defective potential predicted by the knowledge of each designer.

[0014]

In order to solve the above-mentioned problems, according to the present invention, a correspondence relationship between a defect obtained in the development process of a product already made and the cause of the defect, A new knowledge input means for inputting knowledge including mutual relations of occurrence causes and a correspondence relationship between the cause of failure and design parameters, and a design knowledge base for storing the knowledge input by the new knowledge input means are stored. Means and design knowledge display means for displaying the knowledge stored in the design knowledge base,
Requirement specification input means for inputting specifications of a product to be newly designed, and extraction means for extracting knowledge about a product having specifications similar to the specifications of products input to the requirement specification input means from the design knowledge base; Based on the knowledge extracted by the extraction means, a defect occurrence process generation means for generating an explanatory diagram of a correspondence relationship between the defect and the cause of the defect and a mutual relationship of the cause of the defect, and the defect occurrence process generation. It is possible to provide a defect occurrence process display means for displaying the explanatory diagram generated by the means.

Further, when the knowledge extracted by the defect occurrence process generation data extraction means does not exist in the design knowledge base, a feature data input means for inputting further detailed product specifications can be provided.

Further, the design knowledge base is based on at least one feature (partial feature or overall feature) of the outer shape and the internal structure of the product, the constraint of the design parameter, the reason for the control, and the design. Design information such as the necessary analysis model of the numerical analysis system, the predicted failure, and the cause of the failure generation process (hereinafter referred to as failure potential)
And a design parameter that changes the degree of the defective potential, and has a table structure showing a relationship between a defect and the defective potential, a relationship between the defective potentials, and a relationship between the design parameter and the defective potential. You can also

Further, when the relationship between the defect and the defect potential is not related to each other, the defect generation process generating means determines that the defect represented by each column is increased when the numerical value of the defect potential represented by each row increases. If it increases
In the case of decreasing, it is possible to have a matrix in which each is represented by a symbol.

Further, the defect generation process generation means may be arranged such that when there is no relationship between the defect potentials, there is a parallel relationship between the defect potentials, and when there is an increase in the numerical value of the defect potential represented by each row. If the number of defective potentials in a column increases or decreases,
It is also possible to have a matrix in which each is represented by a symbol.

Further, the defect generation process generation means represents the relationship between the defective potential and the design parameter in each column when the numerical value of the design parameter represented in each row increases when there is no relationship between the defective potential and the design parameter. If the number of bad potential increases or decreases,
It is also possible to have a matrix in which each is represented by a symbol.

[0020]

The correspondence relationship between the defect obtained in the development process of the already created product and the cause of the defect, the mutual relationship of the cause of the defect, and the correspondence between the cause of the defect and the design parameter are described. Upon receiving the contained knowledge, the new knowledge input means inputs the knowledge, and the design knowledge base that stores the input knowledge is stored. When the knowledge stored in the design knowledge base is displayed by the design knowledge display means and the specification of the product to be newly designed is received and input by the required specification input means, it is similar to the product specification input by the required specification input means. Knowledge regarding a product having specifications to be extracted is extracted from the design knowledge base by the defect occurrence process generation data extracting means. Based on the knowledge extracted by the defect occurrence process generation data extraction means, an explanatory diagram of the correspondence relationship between the defect and the cause of the defect and the mutual relationship of the cause of the defect is generated by the defect generation process generation means. The defect generation process display means displays the explanatory diagram generated by the generation process generation means. The knowledge extracted by the defect occurrence process generation data extraction means is
If it does not exist in the design knowledge base, the feature data input means receives and inputs more detailed product specifications.

[0021] Further, the restriction of the design specifications and the reason for each feature (partial feature, overall feature) such as the outer shape of the product and the internal structure, and the design of the analysis model of the numerical analysis system necessary for the design. Information, a predicted defect, a cause of the defect generation process (hereinafter referred to as a defect potential), and a design parameter that changes the degree of the defect potential. Relationship, and the relationship between design parameters and failure potentials, the design knowledge base that has a table structure that stores all the information necessary for product design, and all designers need to do the same before designing based on the required specifications. You can know the contents of various analyzes and experiments.

In the design knowledge base, the relationship between a defect and the defect potential, the relationship between the defect potentials,
Knowledge can be held so that the entire design knowledge can be easily confirmed by the design knowledge display means that displays the table structure showing the relationship between the design parameter and the defective potential.

In the design knowledge base, the relationship between the defect and the defect potential is (1) when there is no relationship between the defect and the defect potential, and (2) when the numerical value of the defect potential represented by each row increases, each column. The number of design knowledge data between the defect and the defect potential can be reduced by the matrix configured by respectively expressing the defect by (3) and the defect represented by (3) when the defect is represented by (3). The defect occurrence process generation means capable of generating the defect generation process in a short time compresses the amount of design knowledge data to generate the defect occurrence process of the new product in a short time.

Further, in the design knowledge base, the relations between the defect potentials are as follows: (1) If there is no relation between the defect potentials (in parallel relation),
(2) When the numerical value of the defective potential represented by each row increases when the numerical value of the defective potential represented by each row increases, and (3) when the numerical value of the defective potential represented by each column decreases, the matrix of the defective potentials between them is used. By the failure occurrence process generation means that can reduce the amount of design knowledge data of, and can generate the relationship between the failure potentials in a short time,
The amount of design knowledge data is compressed and the defect occurrence process of a new product is generated in a short time.

Further, in the design knowledge base, if the relationship between the design parameter and the failure potential is (1) there is no relationship between the design parameter and the failure potential,
(2) When the numerical value of the failure potential represented by each column increases when the numerical value of the design parameter represented by each row increases,
(3) When decreasing, the amount of design knowledge data between the design parameter and the defective potential can be reduced by expressing each by a matrix constituted by symbols, and the relationship between the design parameter and the defective potential can be generated in a short time. The defect generation process generation means can compress the amount of design knowledge data and generate the defect generation process of a new product in a short time.

Further, in the defect generation process generation means for generating the defect generation process based on the defect, the defect generation potential, and the design parameter by using the data of each matrix, the generated defect generation process and each defect potential are numerically analyzed. Based on the analysis means to be performed and this analysis result,
By having a tolerance range determining means which receives a user's instruction to predict the tolerance range of the defective potential,
It is possible to make a final determination of the defect potential in the defect occurrence process predicted from the characteristics of the required specifications of the product.

[0027]

[Embodiment] A design knowledge base in LSI package design will be described as an example.

First, the process in which the design knowledge base is used in the package development process will be described with reference to FIG. New package development plan 8
After 1 was performed, various experiments including the development of elemental technology were conducted, the detailed design 82 of the package was conducted, the mold development 83 of the package was completed, the prototype 84 was conducted, and certification 85 was passed through the certification test. Is done. This design knowledge knowledge base is based on the designer's knowledge of defects (described later).
This is applied to the part of creating the experimental plan for design after the new package development plan is decided.

Hereinafter, the term "experiment" means that the designer actually uses the experimental equipment to set conditions, and to perform numerical analysis to confirm the reliability of the product. Knowledge is knowledge acquired in the process of developing various products by design.

As shown in the table a of FIG. 1, the design knowledge base has the classification of the entire package according to the outer size of the package, the internal structure, the characteristics of details, and the like. For example, LQF
A package in which a chip, which is said to have a normal structure due to the difference in the internal structure, is located above the lead frame with respect to the outer shape of the package such as P, QFP, TSOP, LOC.
There is a package in which a lead frame is located on a chip called a structure.

The difference between the outer shape and the internal structure and the classification based on other characteristics are necessary because the defect and the defect generation process are different due to the difference and the classification, and it is necessary to retain individual data. .

The process of occurrence of defects will be specifically described with reference to FIG. Those enclosed by squares indicate defects in the package of this example, and void defects, gold wire deformation defects, and gold wire exposure defects may occur.
Then, how this occurs is shown in this defect generation process (defective potential).
50 in FIG. 2 indicates that the flow velocity between power lines is high when the central flow velocity in front of the chip is high, and the viscosity between power supply lines is likely to be high when the central viscosity in front of the chip is high. It means that if the temperature is high and the viscosity between the power supply lines is high, there is a high possibility that a gold wire deformation defect will occur.

The b, c, and d tables of FIG. 1 respectively show the relationship between defects and defective potentials in the example of FIG. 1a of the TSOP package, the relationship between defective potentials, and the relationship between defective potentials and design parameters. is there. The defect occurrence process is generated by the information in this table. The meaning of each table will be described.

FIG. 1B shows the relationship between the defect and the defect potential. The item of each row (record) shows the defect potential, and the item of each column shows the defect. . Each element of the matrix shows the relationship between the defect potential of each row and the defect of each column. When the element is 0, there is no relation between the defect potential and the defect, and +
The case of means that the number of defects increases when the value of the defect potential increases, and the case of-means that the number of defects increases when the value of the defect potential decreases.

FIG. 1C shows the relationship between defective potentials. Items that cause a plurality of defective potentials are items in each row (record), and results are items in each column. Is a table shown as. When the element of the matrix is 0, it means that there is no relation between the two defective potentials. When the number of elements in the matrix is 1, it means that the two defective potentials have a parallel relationship. When the element of the matrix is +, it indicates that the value of the defect potential of the corresponding column increases as the defect potential indicated by the item in each row increases. For example, in the flow rate before the tip and the flow rate between the bus bars as in the portion 51 surrounded by c in FIG. 1, it means that the flow rate between the bus bars is likely to increase as the flow rate before the tip increases.

FIG. 1d shows the relationship between the defective potential and the design parameters. Items in each row indicate design parameters, and items in each column indicate defective potential. An element of 0 in the matrix indicates that the design parameter is not related to the failure potential, and an element of + indicates that the value of the failure potential increases as the design parameter increases. The fact that the element is-indicates that the value of the defective potential increases as the design parameter decreases.

Next, a design method utilizing the design knowledge base will be described.

FIG. 3 shows the configuration of a design knowledge base utilization environment. First, there is a new knowledge input unit 2 that inputs new knowledge into the design knowledge base 3. Furthermore, a required specification input unit 1 for inputting the outer shape, internal structure, detailed characteristics of parts, etc. of an LSI package to be input, and data for generating a defect occurrence process from a design knowledge base 4 by inputting required specifications. A defect occurrence process generation data extracting unit 3 for extracting the defect occurrence process, a defect occurrence process generation unit 5 for generating a defect occurrence process from the extracted data, a defect occurrence process display unit 6 for displaying the generated defect occurrence process, and It is composed of a feature data input unit 7 that prompts input of other detailed features of the inside when there is no package of the external shape and the internal structure input in the design knowledge base. The design knowledge display unit 8 displays the design knowledge, and the defect occurrence process display unit 6 displays the defect occurrence process.

Next, a method of using the design knowledge base will be described with reference to FIG.

First, when the required specifications are determined, the designer inputs the external size of the package and the internal structure name into the required specifications input section 1 (process 101). At this time, all the input items may not be input by the designer, but there is no problem.
This is because the value and content of all items in a newly developed package do not match. Here, using the input data as a key, a table number representing the relationship between the defect and the defect occurrence potential, a table number representing the relationship between the defect potential and the defect potential, and a table representing the relationship between the defect potential and the design specification parameter. Based on the numbers, the defect occurrence process generation data extraction unit 3 from the design knowledge base 4 shows a relationship between a defect and a defect potential shown in FIG. 1b, and a table showing a relationship between defect potentials shown in c of FIG. , Data showing the relationship between the defective potential and the design parameter, which is shown in d of FIG. 1, is taken out (process 102). When the defect occurrence process generation data extraction unit 3 loads the data, the defect occurrence process generation unit 5
The defect occurrence process is displayed on the defect occurrence process display unit 6 using the data of the retrieved table (process 103). A specific method of generating a defect occurrence process will be described later with reference to FIG.
Will be explained. For example, the defect occurrence process as shown in FIG. 2 is displayed on the screen. As shown in FIG. 2, the designer can create an experiment plan by looking at the displayed defect occurrence process and perform an experiment in which reliability is sufficiently taken into consideration when designing. For example, it can be seen that it is necessary to increase the load on the chip or decrease the load under the chip in order to eliminate the predicted defective exposure of the gold wire. Therefore, the designer can think of one of them to satisfy one of them, make a prototype of the product, and perform an experiment to find the range of the design parameters in which no defect occurs. At this time, the designer also knows that the load on the chip is likely to increase as the viscosity of the central part in front of the chip increases, and from this information, the design parameters (material, shape and dimension) for increasing the viscosity It can be seen that an experiment for changing and adding design specifications) is necessary. Further, if a designer who cannot think of the parameter to be changed picks up the load on the chip on the screen, the related parameter is displayed on the screen as shown in FIG. 7C.
However, an experienced designer performs an experiment by changing the design parameters himself / herself in addition to the parameters recommended by the system based on the defect potential in the defect generation process.

A case where there is no corresponding package data in the design knowledge base when the external size and the internal structure are input to the required specification input unit 1 (see FIG. 3) will be described with reference to FIG. Defect occurrence process generation data extraction unit 3
Activates the feature data input unit 2 (process 201). The feature data input unit 2 presents some questions as shown in FIG. 6 in order to extract the features of a package having a new structure and to take out the defect occurrence process of a package having a similar structure and similar parts. Prompt for input (process 202). When the designer inputs data to the characteristic data input unit 2 for the questions, the defect occurrence process generation data extraction unit 3 for these data uses the design knowledge base 4 as a key for the characteristic data of each question. Fig. 1 showing the relationship between defects and defect potentials
3b, the table shown in FIG. 1c showing the relationship between defective potentials, and the table shown in FIG. 1d showing the relationship between defective potentials and design parameters are taken out. (Process 2
03). When data is loaded by the defect occurrence process generation data extraction unit 3, the defect occurrence process generation unit 5 displays the defect occurrence process on the defect occurrence process display unit 6 using the data of the retrieved table (process 204).

The questions in the feature data input section 2 are increased when a defect occurrence process for a new feature of the package is discovered and design knowledge is increased. Further, when the design of this new package is completed and the knowledge about the new structure is accumulated, the questions are deleted and registered as the knowledge of the new internal structure and outer shape.

Next, a method of generating a defect occurrence process will be specifically described with reference to FIG. First, according to the data of FIG. 1-b loaded in the central processing unit 23 by the processing described with reference to FIGS. 4 and 5, the defect and the defect generation potential are displayed as shown in FIG. Is displayed in. For example, since the values of the void defect 33, the gold wire deformation defect 34, and the gold wire exposure defect 35 of the record of the defective potential of the flow rate 31 between power lines in FIG. 1-b are 0, +, 0, The defect occurrence process of the portion 36 surrounded by the curve a is taken out and displayed. Similarly, a relationship between another defect and the defect occurrence process is generated. Further, whether the number of defects increases when the value of the defect potential increases or decreases, the data of each defect potential is determined.
Retained as data.

Next, in the case where the defective potentials downstream in the defective occurrence process are connected to the already displayed defective potentials by the data shown in FIG.
It is displayed on the defect occurrence process display unit 6 as shown in b. For example,
In the column of the flow velocity between the power supply lines 37 in FIG.
Since the value of 1 is +, 0, 0, 0, the defect occurrence process is displayed like the portion 42 surrounded by the curve in FIG. 7-b. Further, although the relationship between the design parameter and the defective potential is not displayed, when each defective potential is designated with a mouse or the like, a pull-down menu is displayed as shown in FIG. 7-c. For example, in FIG.
The gate volume 61 which is the design parameter of the row for
Front center volume 62 of the tip, left and right front volume 63 of the tip,
Since the values of the cross-sectional area 64 between the power supply lines are −, −, 0, 0, respectively, the flow velocity of the central portion 60 in the front of the chip of the defective potential is 60.
Is specified, the volume of the gate is 61, and the volume of the center of the front of the chip is 6
2 will be displayed as a pull-down menu.

Next, a method of registering in the design knowledge base when the designer acquires new design knowledge will be described. The designer obtains knowledge about the failure occurrence process and estimates the failure occurrence process through analysis and experiments.
The knowledge about the process of occurrence of defects is specifically the knowledge described below.

1. The A defect occurs due to the defect potential b.

2. The defective potential b occurs due to the defective potential a.

3. The defective potential c is caused by the defective potential d and the defective potential e.

4. The defective potential f changes depending on the design parameter x1.

When the above knowledge is obtained, the designer first
Knowledge is input to the new knowledge input unit 2 in FIG. In the new knowledge input section, the external shape, internal structure, or detailed feature of the package is input. For example, as the package outline, TS
OP, a normal structure as an internal structure, or the number of chips as a detailed feature. These are used as keys for extracting the table relating to the defect occurrence process from the design knowledge base 4 in the defect occurrence process generation data extraction unit 3 described above.

When the outline, internal structure, etc. are input, next,
Enter the location and name of the defect that is expected to occur in the package that has the outline and internal structure. For example, the defect name is input such as void defect on the chip. Then, the place where the defect potential, which is the cause of the defect, is observed, and the phenomenon thereof are input. For example, input is made on the chip as the location observed as the cause of the defect, and the flow velocity as the phenomenon name. If there are multiple causes of failure, that is, if there are multiple failure potentials,
Enter what kind of relationship the bad potentials have.

When these are in a parallel relationship, all the defective potentials in a parallel relationship are designated by a mouse or the like, and the command "parallel" is selected from the menu prepared in advance in the new knowledge input section 2. Choose. In the case of a defect potential having a cause-effect relationship, select the "series" command from the menu. In this case, which is the cause (which is the downstream defect potential in the defect occurrence process) and which is the result (which is the upstream defect potential in the defect occurrence process)?
Enter. Next, design parameters related to each defective potential are input. In this way, when relations and orientations have been established among defects, defect potentials, and design parameters, qualitative trends in the relation between defects and defect potentials, the relationship between defect potentials, and the relationship between defect potentials and design parameters are clarified. input.

For example, there is a tendency that defective gold wire deformation increases as the flow velocity between power lines increases. As a specific processing method, two defective potentials and defectives, defective potentials, and design parameters and defective potentials having a cause / result relationship are selected, and the resulting defectiveness or the value of the defective potential is selected. It suffices to input whether or not the value increases when the value of the cause increases or decreases.

[0054]

According to the experiment planning support system having the design knowledge base according to the present invention, all designers can similarly know the defect occurrence process of the newly designed product, and the designers can analyze and experiment in advance. The plan can be created efficiently.

Further, different design knowledge data for each product specification can be compressed and held by a matrix showing the relationship between defects and defective potentials, the relationship between defective potentials, and the relationship between defective potentials and design parameters.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a design knowledge base table.

FIG. 2 is an explanatory diagram of a defect occurrence process.

FIG. 3 is a configuration diagram of a design knowledge base use environment.

FIG. 4 is a flow diagram regarding a method of using a design knowledge base.

FIG. 5 is a flowchart showing how to use a design knowledge base.

FIG. 6 is an explanatory diagram of an example of input items on the feature input screen.

FIG. 7 is an explanatory diagram of a defect occurrence process.

FIG. 8 is an explanatory diagram showing a process of developing an LSI package.

[Explanation of symbols]

1 ... Required specification input section, 2 ... New knowledge input section, 3 ...
... Defect occurrence process generation data extraction unit, 4 ... Design knowledge base, 5 ... Defect occurrence process generation unit, 6 ... Defect occurrence process display unit, 7 ... Characteristic data input unit, 8 ... Design Knowledge display part, 31, 37, 38, 39, 40, 41, 6
0 ... defective potential name, 32, 33, 34 ... defective name, 36 ... part of relation between defective and defective potential, 42 ... partial relation of defective potential,
50: Part of defective process, 51: Value indicating the relationship between defective potentials, 61, 62, 63, 64
…… Design parameter name.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Toshihiro Yasuhara 5-20-1, Josuihonmachi, Kodaira-shi, Tokyo Incorporated company Hitachi Ltd. Semiconductor Division

Claims (6)

[Claims] 1. A correspondence relationship between a defect obtained in the development process of a product already created and a cause of the defect, a mutual relation of the cause of the defect, a correspondence between the cause of the defect and a design parameter. New knowledge input means for inputting knowledge including relations, means for storing a design knowledge base for storing the knowledge input by the new knowledge input means, and design knowledge display for displaying the knowledge stored in the design knowledge base Means, requirement specification input means for inputting specifications of a newly designed product, and extraction for extracting knowledge about a product having specifications similar to the specifications of the product input to the requirement specification input means from the design knowledge base Means and the knowledge extracted by the extracting means to generate an explanatory diagram of the correspondence between the defect and the cause of the defect and the mutual relationship of the cause of the defect. Experimental planning support system characterized in that it comprises a stage, and a failure process display means for displaying the diagram generated by the failure process generating unit. 2. The characteristic data input means for inputting further detailed product specifications when the knowledge extracted by the defect occurrence process generation data extracting means does not exist in the design knowledge base. An experimental plan creation support system characterized by being provided. 3. The design knowledge base according to claim 1, wherein the design knowledge base includes design parameter constraints for at least one feature (partial feature, overall feature) of the product outer shape and internal structure. The reason for the control, design information such as an analysis model of a numerical analysis system necessary for designing, a predicted defect, a cause that constitutes a defect generation process (hereinafter, referred to as a defect potential), and a degree of the defect potential are described. It has a table structure showing the relationship between defects and the defective potentials, the relationship between the defective potentials, and the relationship between the design parameters and the defective potentials. An experimental plan creation support system characterized by storing. 4. The defect generation process generation means according to claim 3, wherein when there is no relation between the defect and the defect potential between the defect and the defect potential, the defect generation process generation means determines each value when the numerical value of the defect potential represented by each row increases. A system for supporting the planning of an experiment, which has a matrix configured to represent by symbols each of a case where the number of defects represented by a column increases and a case where the number of defects decreases. 5. The defect generation process generation means according to claim 3, wherein the defective potentials have no relation between the defective potentials, and the defective potentials have a parallel relation.
An experimental plan creation support system characterized by having a matrix that represents each symbol when the numerical value of the defective potential represented by each row increases or decreases when the numerical value of the defective potential represented by each column increases. . 6. The defect generation process generation means according to claim 3, wherein when the relationship between the defective potential and the design parameter is not related to the defective potential and the design parameter, when the numerical value of the design parameter represented by each row increases. An experimental plan making support system characterized in that it has a matrix that represents each symbol by a symbol when the number of defective potentials in each column increases or decreases, respectively.