JPH06325109A - System and method for evaluating product design specification - Google Patents

Abstract

PURPOSE:To generate relation among defective phenomena of past trial products, evaluation items at the time of an analysis of a product, and various elements of respective designs of the product and to evaluate and optimize a design specification plan for a new product on the basis of the relation. CONSTITUTION:A trial product data management part 24 has data including design specification defect information on existent trial products and a design knowledge management part 25 has respective analytic programs for the evaluation items. Feature quantities of a product shape and material are extracted from the design specification plan for the new product, data having feature quantities similar to the extracted feature quantities are take out of the trial product data management part 24, and a defect causal relation generation part 26 generates the relation (causal relation) among three elements, i.e., the defect phenomena, evaluation items, and various design elements. On the basis of the defect causal relation, an evaluation order generation part 28 generates the order of the evaluation items to be evaluated. In this order, the evaluation items are analyzed by a simulation program stored in the design knowledge management part 25 and a design plan evaluation part 42 decides whether or not the values of various design elements of the inputted product design specification plan need to be corrected on the basis of the analysis results.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is a product for judging whether a design plan of a product design specification (product shape, structure, material, etc.) is good or bad, and if so, generating an appropriate change plan. The present invention relates to a design specification evaluation system.

[0002]

2. Description of the Related Art Conventionally, many analysis methods or analysis systems for evaluating and determining product design specifications have been announced. For example, as a system for performing a flow analysis and a cooling analysis for designing a mold, "mold technology-
2 of "Plastic Injection Mold Design Data Book-"
It is discussed in Volume 2, No. 11, Chapter 2, pp. 16 to 19 (published by Nikkan Kogyo Shimbun (October 20, 1987)). The system sequentially performs a flow analysis, a heat transfer analysis, and a structural analysis according to a molding process, and confirms whether or not design specifications are satisfied by individual analysis results.

In this case, not all items of each analysis affect the product equally. Therefore, the designer himself has decided the importance of each analysis item in the product and selected evaluation items corresponding to the product. Then, the actual evaluation is performed centering on the selected evaluation item.

Note that, as a related analysis system of this type, for example, Japanese Patent Laid-Open Nos. 63-85871 and 2-27856.
No. 3-1-27271, No. 3-147023, No. 3-265974, No. 4-
75108, 4-137073, and 4-153775.

[0005]

However, now that the products have become complicated, the number of evaluation items is increasing, and it becomes difficult to select the evaluation items to be actually evaluated. Moreover, since the importance of each evaluation item is different, it is not efficient to simply follow the order without prioritizing. Now that the design period is being further shortened,
This problem is becoming more and more serious.

The prior art described above does not consider the following points and is applied to the design and evaluation of design specifications of a new product that is approaching the limit design (the unit of metrology at the time of design is extremely small). However, the sufficient effect could not be obtained.

In products approaching the limit design,
In order to determine product specifications, various design specifications (for example, chip position, chip thickness, etc.) that are related to design departments with different technical fields are different (each technical field is different).
It is necessary to evaluate from the viewpoint.

In order to make an overall evaluation in this way,
It is necessary to select the evaluation items necessary for the evaluation of the design specification, determine the evaluation order of the selected evaluation items, and determine the evaluation criteria according to the determined evaluation items. However, in many cases, it is not possible to use all evaluation techniques because evaluation techniques related to different specialized techniques are required. Further, even when it can be used well, the evaluation item, the evaluation order, and the evaluation standard must be determined by the designer's unique intuition and experience (without objectivity).

As a result, the specifications finally obtained vary greatly depending on the designer in charge, and the evaluation man-hours and the period also vary greatly. In addition, there are problems that the evaluation items important for determining the design specifications are overlooked, and the evaluation man-hours and the period are significantly increased. In the past, the design margin was large, and evaluation based on the experience peculiar to the designer did not have a large effect on the final specifications.However, in the design of products approaching the limit design, the evaluation criteria The fact that the decision is not objective and ambiguous has a great influence on the occurrence of product defects, which is a cause of lowering the design quality.

Further, in the prior art, even if the evaluation items are evaluated based on the determined evaluation standard and the possibility of occurrence of defects is grasped, the design parameters to be changed concretely are determined, or this design is changed. Determining the amount of change in specifications is not objective and must rely on the experience and intuition of the designer. For this reason, there are situations in which design specifications to be changed cannot be appropriately determined, or changes must be repeated many times before appropriate determination. For this reason, the design quality is deteriorated, and the repetition from design to trial production occurs frequently.

An object of the present invention is to solve the above problems, to evaluate a design specification proposal of a product approaching a marginal design, and modify the design specification proposal based on this evaluation to determine the design specification proposal. It is to provide a product design specification evaluation system that can.

More specifically, the first object of the present invention will be described.
The purpose of is to provide a product design specification evaluation system that can objectively select the evaluation items for the design specification proposal, the evaluation order of the selected evaluation items, and the evaluation criteria for each evaluation item. is there. For this purpose, the relationship between design specifications, evaluation items, and defective defects in prototype data similar to the design specification proposal is quantified, and the evaluation item of the design specification proposal is selected and selected based on this relationship. The evaluation order of the evaluation items and the evaluation criteria for each evaluation item are determined.

Next, a second object of the present invention is to evaluate the evaluation items based on the determined evaluation criteria and grasp the possibility of occurrence of defects, and then determine the design parameters to be changed concretely. Alternatively, it is to provide a product design specification evaluation system capable of objectively determining the amount of change in the design specifications.

[0014]

In order to solve the above-mentioned problems, according to the present invention, at least the product shape, structure,
An input means for inputting a product design specification proposal including one of material, manufacturing condition, and design constraint, a prototype specification that is a design specification of a product already prototyped, a type of defect for each prototype specification,
Prototype data management means for preliminarily holding and managing prototype data consisting of defect information including at least one of occurrence rate and occurrence site, and evaluation items to be analyzed when evaluating a product design specification proposal. Each of them can be provided with design knowledge management means having a simulation program for performing analysis. Further, a product feature amount is extracted from the product design specification proposal, and trial production data having a feature amount similar to the feature amount is taken out from the trial production data managing means, and a first evaluation item and a defective phenomenon are obtained. And the second relationship between the design items which are the design data of the product according to the product design specification proposal and the evaluation item, and the first and second
It is possible to provide a failure causal relationship generating means for generating a failure causal relationship which is a relationship among the three elements of the failure phenomenon, the evaluation item, and the design specification from the relationship of. Further, the order of the evaluation items to be analyzed is generated from the defect causal relationship obtained from the defect causal relationship generation means, in consideration of at least one of the size of the defect occurrence rate, the strength of the design constraint, and the difficulty of modifying the specifications. According to the evaluation order generating means, and the order generated by the evaluation order generating means, the simulation items stored in the design knowledge managing means are used to analyze the evaluation items, and the analysis items are input based on the analysis result. It is also possible to provide a design plan evaluation means for determining whether or not the values of the design specifications of the product design specification plan need to be corrected.

[0015]

[Operation] As a pre-processing before operating the product design specification evaluation system, the prototype data management means is provided with specifications (prototype specifications) such as the shape, structure, material and manufacturing conditions of the prototype already manufactured, Prototype data consisting of defect information, such as the type of defect, occurrence rate, and occurrence site of each work, is registered in advance.

Next, the input means is used to input the design specification proposal such as the shape, structure, material and manufacturing condition of the new product to be evaluated and the constraint of the design specification.

The defective causal relationship generating means extracts the characteristic amount of the design specification draft of the new product inputted by the inputting means, and based on the characteristic amount and the constraint of the design specification, the prototype data managing means. A prototype specification similar to the design specification draft is extracted from the prototype specifications registered in. The design parameters of this prototype specification, the values of the evaluation items for these parameters, and the relationship between defects (defect phenomena) (defect causal relationship) are generated.

With respect to the relationship between the defect phenomenon and the evaluation item that constitute the defect causal relationship generated by the defective causal relationship generating means, the evaluation criterion generating means determines the defect occurrence rate and the value of the evaluation item. A failure occurrence rate prediction graph is generated by plotting the relationship of. In this defect occurrence rate prediction graph, the evaluation criterion is generated by setting the range of the value of the evaluation item having the defect occurrence rate of 0 as the allowable range for this evaluation item. That is, when the value of the evaluation item calculated from the design specifications of the design specification proposal is within this allowable range, the design specification proposal is judged to be good.

The evaluation order generating means selects an evaluation item for evaluating the design specification proposal based on the defective causal relationship generated by the defective causal relationship generating means,
Determine the evaluation order of the selected evaluation items. In this case, the defect phenomenon having the highest defect occurrence rate is taken out, and the evaluation item related to this defect phenomenon is taken out as the first evaluation item.
If there are multiple evaluation items that have been taken out, the evaluation items that have a large defect occurrence rate in the relationship between the defect phenomenon and the evaluation item and the evaluation items that are not related to other defective phenomena should be evaluated first. To generate an evaluation order.

The value of the evaluation item is obtained by the simulation executing means in accordance with the evaluation order generated by the evaluation order generating means. This is performed by giving the simulation program prepared in the evaluation executing means for each evaluation item the values of each design specification of the design specification draft as data, and executing the simulation program. Be done.

The design plan evaluation executing means evaluates the design specification proposal based on the values of the evaluation items obtained by the simulation executing means and the evaluation criteria generated by the evaluation reference generating means. Specifically, when the value of the evaluation item is within the allowable range based on the evaluation standard, the design specification proposal is evaluated as good, and the design specification proposal does not need to be modified. If the value of the evaluation item is not within the allowable range based on the evaluation criteria, the design specification proposal is evaluated as no and the design specification proposal needs to be modified.

When the design proposal evaluation executing means evaluates that the design specification proposal needs to be modified, the amount of change of the design specification proposal is further calculated. Specifically, the difference between the value of the evaluated evaluation item and the allowable range of the evaluation standard (this is called margin) is calculated. Further, for each of the design parameters related to this evaluation item in the failure causal relationship, the sensitivity, which is the ratio of the variation of the evaluation item value to the variation of the design parameter value, is obtained.

The countermeasure plan generating means has no influence on the evaluation items already evaluated / countermeasured from the sensitivity, the margin, and the relationship between the design specifications and the other evaluation items shown in the defect causal relationship. Design specifications, or if there is an influence, select design specifications with high sensitivity in the design specifications with a large margin of evaluation items, and evaluate the product sum of sensitivity and correction amount of each selected design specification. The correction amount of each design specification is determined so that the reference value of the item value is the protruding amount (the value of the evaluation item before the correction is outside the allowable range).

If the information on the design process is held, the evaluation result of the design proposal, the design process and the correction result of the design specifications are sequentially left and predicted together with the generated defect causal relationship and the accompanying information. Since the intentions related to the specifications and the amount of modification for avoiding defects are accumulated and reused, it is possible to easily understand the influence of the specification change on the evaluation result of other specifications.

Further, in a workstation including a bus, a bus control unit, a central processing unit, a disk control unit, a main storage unit, a display unit, a keyboard, and a disk, a prototype data forming the above product design specification evaluation system. A data management means, an input means, a failure causal relationship generation means, an evaluation reference generation means, an evaluation order generation means, a design plan evaluation means, and a countermeasure plan generation means are stored in the main storage device, and the prototype data management means is used. Managed prototype data
Data is stored on the same disk, and the central processing unit executes each means on the main storage device, so that the design specifications such as the shape, structure, material, and manufacturing conditions of the new product and constraints of the design specifications can be obtained. Based on the above, the cause of the failure that is implicitly included in the prototype data of the developed product can be extracted by the causal relationship between the failure phenomenon, the evaluation item, and the design specifications, and the cause of the failure can be systematically avoided.
The design specifications such as the shape, structure, material, and manufacturing conditions of the new product can be comprehensively and efficiently evaluated, and if necessary, the design specifications can be efficiently and easily corrected.

Further, in the workstation / computer device, trial data management means, input means, defect causal relationship generation means, evaluation reference generation means, evaluation order generation means, design plan which constitute the product design specification evaluation system. The evaluation means and the measure proposal generating means are stored in the main storage device on the workstation side, and the defective causal relationship generating means, the evaluation reference generating means, the evaluation order generating means, the design proposal evaluation which constitute the product design specification evaluation system are stored. Means and countermeasure plan generation means are stored in the main storage device of the computer side, and the prototype data managed by the prototype data management means are stored on the disk of the computer side, thereby the shape and structure of the new product. , The material, manufacturing conditions, and other design specifications and constraints of the design specifications, the cause of the failure that is implicitly included in the prototype data of the developed product is the failure phenomenon, the evaluation item, and the setting. Extracted by the causal relationship between specifications, the cause of the failure to systematically avoided, the shape of the new product, structure, material, comprehensively and efficiently evaluate the design specifications such as manufacturing conditions,
If necessary, the design specifications can be modified efficiently and easily, and the response performance can be improved by using the central processing unit on the computer side depending on the processing load when executing each means. It can also be used for large-scale prototype data.

[0027]

The present invention will be described below with reference to the drawings.

First, a basic product design specification evaluation system according to the present invention will be described. FIG. 1 shows the hardware configuration of the product design specification evaluation system according to the present invention, and FIG. 2 shows the software configuration of the product design specification evaluation system. As shown in Fig. 1, the hardware configuration equipped with the product design specification evaluation system is
It is under the control of the bus controller 12. In addition to the central processing unit 13, the multi-bus 11 accommodates a disk unit 18 via a disk control unit 17, and the central processing unit 13 has a main storage unit 14, a display unit 16 and a keyboard.
15 have been housed. As a result, the data input from the keyboard 15 is stored in the main memory 14 by the central processing unit 13 and at the same time the display unit 16
Is displayed in. Further, the data on the main storage device 14 is transferred to the multi-bus 11 and the disk control device 17 by the central processing unit 13.
Data is transferred and received by being stored in the disk device 18 via the.

As shown in FIG. 2, the system software includes a data input section 22, a data output section 23, and a data management section.
40, the evaluation procedure generation unit 41, the evaluation reference generation unit 27, the design plan evaluation unit 42, the countermeasure plan generation unit 30, the control unit 21, and other processing / control software. Furthermore, the data management unit 40 is a prototype data management unit 24.
From the design knowledge management unit 25, the evaluation procedure generation unit 41, the defective causal relationship generation unit 26, the evaluation order generation unit 28, the design plan evaluation unit 42
Is composed of a design plan evaluation executing unit 29 and a simulation executing unit 31.

Here, the relationship between each processing unit in FIG. 2 and the block diagram in FIG. 1 will be described. The product design specification evaluation system according to the present invention comprises a multi-bus 11, a bus controller 12, a central processing unit 13, a main memory unit 14, a keyboard 15 shown in FIG.
It is composed of hardware of a display device 16, a disk control device 17, and a disk device 18, and each processing unit of the software of the product design specification evaluation system shown in FIG. The disk device 18 stores prototype data managed by the prototype data management unit 24 shown in FIG. 2 and design know-how managed by the design knowledge management unit 25. By operating each processing unit of the product design specification evaluation system on the main storage device 14, design specification evaluation, countermeasures, and determination are performed.

Hereinafter, these will be described more specifically.

The data input unit 22 stores the design specification proposal such as the shape, structure, material characteristics, and manufacturing conditions, and whether or not the contents in the design specification proposal can be changed or the design constraint such as the change allowable range in the system. It is designed to be taken into.

The data management unit 40 is composed of a prototype data management unit 24 and a design knowledge management unit 25. The prototype data management unit 24 is composed of prototype specifications such as the shape, structure, material and manufacturing conditions of the already developed product, and defect information such as the type of defect, the occurrence rate, and the location of occurrence of each prototype. It is designed to hold and manage prototype data. This prototype data is stored in the disk device 18 in advance before operating the system.

Further, the design knowledge management section 25 is adapted to hold and manage the know-how of the designer regarding the defect occurrence process in the manufacturing process, and the know-how data is stored in the disk device 18 in advance before operating the system. It is supposed to be done. An example of the data managed by the design knowledge management unit 25 will be described with reference to FIG. 3 (a). The flow velocity difference 3001 between the top and bottom of the chip is large, the flow imbalance 3002 is large, and the stress difference is large.
03, the know-how that a void defect 3005 will occur after the physical state of the lead frame deformation 3004 affects the parameters of the physical quantity in each state and the design specifications (3006, 3007 , 3008, 3009).

The evaluation procedure generator 41 is a defective causal relationship generator
26 and an evaluation order generation unit 28. The defect-causal relationship generating unit 26 is configured to design the draft design specifications such as the shape, structure, material characteristics, and manufacturing conditions input from the data input unit 22, and whether or not the contents in the draft design specifications can be changed or the allowable change range can be changed. A feature amount is calculated from the constraint. On the other hand, the feature amount is calculated for each of the trial production data stored in the trial production data management unit 24.

Here, the feature amount will be specifically described with reference to (a) to (e) below.

(A) Shape (complexity) Chip position, package internal structure (eg, FIG.
It is related to the flow state as shown in FIG. 16, and includes the chip shape, the lead frame shape, the chip size, and the thickness),
Whether there is a tab, the size of the tab, the shape of the tab, etc.

(B) Size: Package volume, package aspect ratio, etc.

(C) Material Resin material (sugar content parameter, specific heat, density, etc. of the material),
Lead frame material (whether the material is copper or 42 alloy, etc.).

(D) Manufacturing conditions: molding temperature, speed, mold type (single pot, multi-pot, etc.), mold volume, cavity position, gate position, gate angle.

(E) Manufacturing process This affects the flow of resin during molding.
It can be substituted in connection with the analysis model shown in FIG.

After that, the feature quantity of the design specification plan and the feature quantity of the prototype data are compared, and the defect phenomenon (type of defect, occurrence rate, occurrence site) of this prototype data having similar feature quantity is evaluated. A defective causal relationship that is a relationship between items and design specifications is generated.

A method of generating a defective causal relationship will be described. First, the correlation between the defective phenomenon and the evaluation item is obtained, and all the two-element relationships between the defective phenomenon and the evaluation item are generated. The relationship between these two elements is shown in FIG. For correlation,
"Calculation Probability Statistics" (published by Sunouchi, Terada, Funane, published by Science Co.), p. 7-p. 8 is mainly described.
The relationship between the defect phenomenon and the evaluation item can be obtained from the defect occurrence rate obtained from the prototype data. Regarding the evaluation items that form the relationship between these two elements,
A two-element relationship between the evaluation item and the design specification is generated from the design specification that affects the evaluation item in the design specification proposal and can be changed under the constraint condition. The relationship between these two elements is shown in FIG.
Shown in. A defect-causal relationship representing the relationship among the three elements of the defect phenomenon, the evaluation item, and the design specifications is generated from the relationship between the two elements of the defect phenomenon and the evaluation item and the relationship between the two elements of the evaluation item and the design specifications.

This defective causal relationship will be described with reference to FIG.

FIG. 3 (a) shows a product state up to a product defect at the time of molding. 3001, 3002, 3003, 3004, 3010, 3013 surrounded by an ellipse indicate a progress state, and 3005, 3012 surrounded by a double ellipse indicate a defective product state. 3006, 3007, 3008, 3009, and 3011 indicate elements (design specifications) that influence when the state changes.

For example, a large vertical chip flow velocity difference 3001, a flow imbalance 3002, a large stress difference 3003, and lead frame deformation
After the state of 3004, a void state 3005 is reached.
Further, the flow imbalance 3002 may directly lead to a defective state of void generation 3005. Further, the state of high viscosity 3010 may lead to a defective state of void generation 3005.

As shown in FIG. 3A, defective phenomena 3005, 3012
And evaluation items 3002, 3004, 3010 and design specifications 3009, 300
If there is a relationship with 6, 3011, sort out these relationships (sort by leaving things with a strong degree of correlation etc.),
A bad causal relationship can be generated as shown in FIG. In FIG. 3 (b), the occurrence rate of defective phenomenon,
The values of design specifications are obtained from past prototype data. The value of the evaluation item is obtained from the value of the design specifications of the past prototype data by simulation (a program for simulation is prepared for each evaluation item). Figure 3
(B) shows a part of the defective causal relationship regarding the formability 3121 of FIG. 4-1, which will be described later. That is, regarding the evaluation of the moldability 3121 in FIG. 6, it indicates that the evaluation items such as the flow balance, the lead frame deformation amount, and the viscosity in FIG. 3B are evaluated. 6, thermal stress 3122, heat dissipation 3123, lead frame pattern 3131, lead frame strength 3132, bonding process 3141, molding process
For each of the 3142, the range material 3151, and the lead frame material 3152, it is possible to generate the failure causal relationship as shown in FIG. 3B (of course, the content of each failure phenomenon, evaluation item, and design specification). Is different).

Here, the defective causal relationship will be more generally described with reference to FIG. For example, the defective phenomenon of void occurrence 901 is the lead frame deformation amount 911, the flow balance 9
12 and viscosity 913 are related to the evaluation item, and the value of the evaluation item of lead frame deformation amount 911 is chip position 92.
0, lead frame thickness 922 and lead frame suspension position
It is shown that there is a relation with the design parameter 923 (the analysis parameter can be obtained by inputting these design parameters). FIG. 18 shows that even if the relationship between design specifications and the occurrence of defects is complicated, a simple relationship can be expressed by selecting an appropriate evaluation item and interposing this evaluation item.

Returning to FIG. 2, the evaluation order generation unit 28 takes out the trial production data having the feature quantity similar to the design specification proposal from the trial production data management unit 24 for each evaluation item, and the failure rate for each failure phenomenon. Is calculated and is set as the defect rate of the defect phenomenon shown in the defect causal relationship.

Further, for each two-element relationship between the failure phenomenon and the evaluation item shown in the failure causal relationship, prototype data having a characteristic amount similar to the characteristic amount obtained from the design specification proposal is produced. From the data management unit 24, the defect rate for each defect phenomenon relating to the evaluation item is calculated, and is set as the defect occurrence rate of the two-factor relationship between the defect phenomenon and the evaluation item indicated in the defect causal relationship. There is. Then, the failure phenomenon having the maximum failure occurrence rate set as the failure phenomenon related to the failure cause-effect relationship is extracted, the evaluation item related to this failure phenomenon is extracted, and this evaluation item is set as the first evaluation target. If there are multiple evaluation items that have been extracted, the evaluation items are evaluated in the order of the evaluation item with the highest failure rate of the two-element relationship between the failure phenomenon and the evaluation item, and the evaluation item that is not related to other failure phenomena or has a low relationship. The evaluation procedure is generated as follows.

The evaluation criterion generating section 27 is a defective causal relationship generating section.
For all the relationships between the two elements of the failure phenomenon and the evaluation items that make up the failure causal relationship generated by 26, for each of the same relationships, the prototype data having similar feature quantities to the design specification proposal is created. , Plot the relationship between the value of the evaluation item and the failure rate,
A defective rate prediction graph is generated.

FIG. 21 shows a specific example of this defect rate prediction graph. The defect rate prediction graph shows the relationship between past prototype data (defect occurrence rate) and the values of evaluation items obtained by simulation. That is, it is a quantitative expression of the relationship between the two factors of the failure phenomenon and the evaluation item shown in FIG. 19, and can be considered as an intermediate result for generating the evaluation standard. Note that FIG. 21 is a defect rate prediction graph showing the relationship between the lead frame deformation amount and the void defect rate.

After generating this graph, the values of the evaluation items are divided into a range of evaluation items having a defect occurrence rate of "0" and a range of other items, and the range having a defect occurrence rate of "0" is subjected to this evaluation. Generate the evaluation criteria as the allowable range for the item.

FIG. 2 shows the method of generating this evaluation standard.
2 is used for the explanation.

First, in process 1701, the relationship between the value of the evaluation item (evaluation value) and the occurrence of defects is derived. Specifically, the simulation is performed using the prototype specifications (design specifications) in the prototype data as input data, and the value of the evaluation item for each time is obtained. Then, the prototype specification in which the defect has occurred and the prototype specification in which the defect has not occurred are classified.

Next, in process 1702, a time period in which it is possible to easily determine the presence or absence of a defect is determined from the result of process 1701, and an evaluation window is set. In process 1703, the relationship between the evaluation item and the defect rate is derived. Regarding the prototype specifications in the evaluation window, the correspondence between the evaluation item value and the defect rate is graphed.

In process 1704, the allowable range of the evaluation item in which no defect occurs is obtained from the defect rate prediction graph generated in process 1703. Which of the values of the evaluation items the defect occurrence rate of which is “0” (either 1705 or 1706 in the processing 1704 of FIG. 22) depends on how the design margin is taken.

FIG. 23 shows an example in which the evaluation standard determined in the process 1704 of FIG. 22 is reflected in the process 1701 (an example of how to set the evaluation standard).

Returning to FIG. 2, the design plan evaluation section 42 is composed of a design plan evaluation execution section 29 and a simulation execution section 31. The design plan evaluation executing unit 29 executes the simulation in accordance with the evaluation order of the evaluation items generated by the evaluation procedure generating unit 28.
A simulation is carried out in order to obtain the value of the evaluation item via the operation executing section 31 (a simulation program such as a flow analysis, a deformation analysis and a structural analysis is provided, and the simulation is performed. Option execution section 31
According to the type of evaluation item for which a value is to be selected, the program is selected and executed from the above programs).

At the time of executing this simulation program, the design plan evaluation executing unit 29 obtains data such as the package shape, structure, lead frame shape, molding condition, material characteristic, and chip position of the design specification proposal. The data is converted into a format suitable for execution of the simulation program, and the simulation program is executed using this data. Further, in the design plan evaluation executing unit 29, the evaluation standard generated by the evaluation standard generating unit 27 and the value of the evaluation item of the design specification proposal obtained by the simulation executing unit 31 are compared,
It is determined whether or not the value of this evaluation item is within the evaluation criteria, and whether or not the design specification proposal needs to be changed is determined based on this determination. In other words, it is judged that it is not necessary to change the design specifications if the value is within the evaluation standard (value within the allowable range), and it is necessary to change the design specification if the value is outside the evaluation standard (value outside the allowable range). It is judged that there is. When it is determined that the evaluation item needs to be changed, the margin (calculated as the difference between the evaluation standard and the value of the evaluation item) of the evaluation item is calculated. In addition, regarding the design specifications expressed in the defect causal relationship and related to the relevant evaluation item (if there are multiple, each design specification), fluctuations in the design specifications (position, thickness) Sensitivity, which is the ratio of fluctuations in the values of evaluation items to the fluctuations in the

A method of obtaining this sensitivity will be described.

First, the outline of the concept of sensitivity analysis and the definition of sensitivity will be described. The sensitivity analysis is for quantitatively grasping the correction amount for correcting the design specifications when the evaluation criteria (allowable range) is not satisfied in the evaluation of the design plan.

The sensitivity can be defined as follows. Evaluation item F obtained from simulation results
Is regarded as a function of design specifications a, b, c, ....

F = F (a, b, c, ...) The sensitivity to the design specifications a, b, c is, for example, as shown in FIG.
Define as follows. 34 (1) will be referred to as the sensitivity of the parameter a in the sense that the change amount (Δa) of the parameter a is affected. Actually, there are a plurality of parameters, and F is included in the evaluation standard by a combination of changing the plurality of parameters. A specific example of sensitivity is shown in FIG.

FIG. 36 shows four types of sensitivity definitions (sensitivity expression methods) including the above sensitivity definitions.

The basis of the sensitivity calculation method is to calculate the first partial differential of the function F with the parameter a. That is, the approximate expression obtained by Taylor expansion,

[0067]

[Equation 1]

Therefore, the first partial differential of the function F with respect to a is

[0069]

[Equation 2]

It is That is, the parameter a and the parameter (a + Δa) can be simulated to calculate the value of the first partial differential of the function F with the parameter a. FIG. 37 shows a method of calculating the value of the first partial differential by software.

Since the sensitivity calculation requires a huge calculation cost, it is necessary to improve the efficiency. However, in the method shown in FIG.
o. In the learning data method of No. 3, since data can be prepared in advance, efficiency can be improved. In the present embodiment, No. 1 among the methods shown in FIG. Although it is assumed that the direct method of No. 1 is used, no. 2, No.
Method 3 may be used.

When the sensitivity is calculated, the designation items shown in FIG. 38 are designated by the user.

Returning to FIG. 2, the countermeasure plan generating section 30 determines the sensitivity and margin of each design specification with respect to the evaluation item obtained by the design plan evaluation executing section 29, and the design specification and other factors indicated by the defect causal relationship. The amount of modification of each design specification is determined based on the relationship with the evaluation item. As a candidate for design specifications to be modified, design specifications that do not affect the evaluation items that have already been evaluated and taken countermeasures, or if there is an impact, design specifications that have a large margin for evaluation items and have high sensitivity. Select design specifications. The correction amount of each selected design parameter is the product sum of the sensitivity and the correction amount, and the standard protrusion amount of the evaluation item value (the difference between the evaluation item value before correction and the threshold value of the allowable range). Decide to become.

In determining the correction amount, when the relationship between the value F of the evaluation item and the time t is shown in FIG. 39, (1) The deformation amount is finally determined like the lead frame deformation amount. If it is enough to be within the reference range, only the portion 3901 in the figure is considered (only the portion 3901 is within the reference range).

(2) When it is necessary to keep the velocity within the reference range in the entire region like (velocity × velocity), only the portions 3901 and 3902 in the figure are considered (the portions 3901 and 3902). Make sure that the part of is within the reference range.At this time, the part of 3901 is within the reference range.
Next, make sure that 3902 is within the reference range. ).

A method of correcting design specifications using the result of the sensitivity analysis will be described with reference to FIG. As shown in the left figure, when the simulation result F for the evaluation item goes out of the standard, the modification amount (a, b, c, ...) Of the design parameters (a, b, c, ...) Δa, Δb, Δc,
...) is required appropriately. That is, the protrusion amount ΔF
The correction amount Δa, Δb, ... Of each item is obtained from (the difference between F and the reference) and the sensitivity Ka, Kb ,. That is, the following (Equation 3) is satisfied (Δa, Δb, Δ
c, ...) is obtained.

[0077]

[Equation 3]

The design specifications (a, b, c, ...) Have some constraints, and the solution that satisfies the constraints from the same combination is the solution.

[0079]

[Equation 4]

FIG. 41 shows the correction amounts Δa and Δ of the above specifications.
The constraint expression regarding b, ... is shown.

Returning to FIG. 2, the data output unit 23 requests the operator to select a plurality of countermeasure plans generated by the countermeasure plan generator 30 and one of these countermeasure plans. The message and are displayed on the display screen of the display device 16.
In response to this request, the operator can select one from a plurality of countermeasures using the keyboard 15 and give an instruction.
This instruction determines the countermeasure.

The control unit 21 controls the execution of each of the above processing units to manage the flow of processing of the entire product design specification evaluation system.

Now, the case where the product design specification evaluation system according to the present invention is applied to the design of a semiconductor plastic package will be described as an example.

First, the structure of the semiconductor plastic package will be described. FIG. 4A is an explanatory diagram of the structure of the semiconductor plastic package before LOC (Lead on Chip), and FIG. 4B is an explanatory diagram of the LOC (Lead on Chip) structure. The package 500 of the conventional structure is a lead frame for electrically connecting the package 501 with an electronic circuit-baked chip 501, a tab 504 for mounting and fixing the chip 501, and an external device such as a printed circuit board and the package 500. 502, a gold wire 505 for electrically connecting the chip 501 and the lead frame 502, and a resin 503 for protecting internal structures such as the chip 501 from the external environment.

On the other hand, the LOC structure package 500 'has a structure in which a lead frame 502' and a bus bar 506 are arranged on the surface of a chip 501 'via an insulating film 507. Unlike conventional structure, gold wire 505 'makes tip 501'
Since the bonding pad on the chip 501 'connecting to the lead frame 502' and the lead frame 502 'is in the center, there is no need to lay the lead frame 502' at the end of the chip 501 ',
The outer shape of the package can be reduced accordingly. Further, since the bus bar 506 is provided on the surface of the chip on which the electronic circuit is burned, noise can be reduced.

The reason why the above-mentioned package having a new structure has been developed is that the package must be miniaturized and thinned due to the diversification of electronic equipment and the need for high-density mounting. For this reason, the inside of the package is inevitably a low-rigidity and complicated structure, and there is a high possibility that a product defect will occur due to a slight difference in (parameter) value of design specifications. Has become.

As examples of product defects, as shown in FIG. 5, voids 601, lead frame deformation 602, wire bending 603, and other product defects caused by resin flow during molding, and packaging that occurs during board mounting. − Jicrac 604
Such as product defects, adhesive peeling 605, etc. that occur due to temperature cycles due to turning on and off of electronic devices.

As described above, in order to cope with the situation in which product defects are likely to occur, the product design specification evaluation system according to the present invention has design specifications and evaluation items required from past prototype data. Evaluate the design specifications of the new product's design specifications based on the relationship with the failure phenomenon, and give instructions on the design specifications where the failure phenomenon is predicted to occur and the amount of modification of these design specifications. It is possible to prevent product defects.

Next, the development process of the semiconductor plastic package and the position of the package design in this development process will be described.

FIG. 6 shows a development process in a semiconductor plastic package, and particularly, details of the contents of the package design. As shown in FIG.
In the development of semiconductor plastic packages, products such as outer shape, number of pins, required grade, allowable power consumption, electrical characteristics, allowable chip size, package thickness, mounting guarantee (reflow resistance), etc. Determine the draft specification (Process 30)
1). Next, the package is designed (process 302) and the mold is developed (process 303) based on the product specifications, and a prototype is manufactured and evaluated (process 304). In process 304 (prototype / evaluation), the presence or absence of defects such as lead frame deformation, void generation, wire bending, package crack, and adhesive peeling is evaluated. If the evaluation result is NG (defective occurrence), the process returns to step 302, the package design is examined again, and the mold development 303 and the trial production / evaluation 304 are performed. If the evaluation result is OK (no defect occurs), the designed package is certified (process 305), and the package is mass-produced (process 306).

The package design 302 will be described in detail. The package design 302 is a comprehensive package internal structure and process specifications such as assemblability, moldability, reflowability, moisture resistance, temperature cycle resistance, board mountability, heat dissipation and electrical characteristics. Evaluation Review (Process 31
After performing 0), outer shape design 311, structure design 312, lead frame design 313, process design 314, and material design 315 are individually performed. 3121, 3122, 3123, 3131, 3132, 314
1, 3142, 3151, and 3152 indicate the contents to be considered in each of the designs 311 to 315 (for example, structural design 312
Examines formability 3121, thermal stress 3122, and heat dissipation 3123). A detailed description of these designs 311-315 follows.

(1) In the external design 311, the Electronic Industries Associations of Japan (EIAJ)
n), Electronic Devices Joint Committee of the American Electronics Industry Association (JEDE
C: Joint Electron Device Engineering Council) etc. outline standard, pin pitch and mounting height based on customer requirements,
And design the package thickness based on the results of the examination of the internal structure.

(2) In the structural design 312, the formability 3121 at the time of molding, the thermal stress 3122 related to the temperature cycle property and the board mountability, and the heat dissipation property 31 are based on the examination result of the internal structure.
23 designs.

(3) In the lead frame design 313, the outer shape specification is determined in consideration of the compatibility of the assembly line with other packages, the tab suspension lead width, the tab size, and the lead routing. The design of the inner lead pattern 3131 in consideration of the above, and the determination of the lead frame material 3132 based on the examination result of the internal structure are performed.

(4) In the process design 314, the assembly process (boding process, molding process) from dicing to selection is determined and its problems are identified based on the examination result of the structural design 312.

(5) In the material design 315, the resin material 3151 and the lead frame material are considered in consideration of the reliability based on the customer's request, the mounting guarantee and the electrical characteristics, and the heat dissipation regarding the power consumption.
Determine the material such as 3152.

In the package design 302 in this development process, product defects such as void generation 601, lead frame deformation 602, wire bending 603, package crack 604 and adhesive peeling 605 shown in FIG. In order to prevent the generation, the designer has to combine various conditions in order to determine the package design specifications such as the package outer shape, structure, lead frame shape and material, and the process conditions in detail. Design with consideration.

Here, the mutual relationship between the design departments 311 to 315 will be described. First, the mold used for molding the mold will be described with reference to the drawings.

FIG. 7 shows an ordinary semiconductor plastic package.
It is explanatory drawing which illustrates the mold for die molding of J, a molding machine, and a molding process. In contrast to the multiple lead frame 403 in which the semiconductor chip 401 is bonded with the gold wire 402, a resin mold is used.
The laid lead frame 420 is cut and processed into each chip, and a product 421 is manufactured. 409 and 405 are upper and lower molds for performing this resin mold molding. The lower die 405 is provided with a cavity 404 for simultaneously performing mold forming on the multiple lead frame 403, and a runner groove 406 for pouring the molten resin into this and a gate of each cavity. To 407 is carved. In the upper mold 409, resin
A pot 410 for accommodating 411 and a plunger 413 for pushing in the resin are fitted together. Also,
A heater 408 for melting the resin is embedded in the upper and lower molds.

In the structural design (department) 312 shown in FIG. 6, the thermal stress applied to the package is made up of a member having a different coefficient of thermal expansion such as metal, silicon or resin against the thermal stress applied to the package. Etc. will be evaluated and a structurally strong member layout will be designed.

In the process design (department) 314, the viscosity and flow velocity of the inflowing resin are evaluated so that the gold wire is not pushed and bent by the resin flowing into the cavity, and the insert (chip , Lead frames, etc.)
Design the mold specifications such as the flow path shape and molding conditions while evaluating the resin flow state by. Here, for example, the following relationships exist among design specifications (parameters) handled by each design department.

(1) Regarding the chip position, the larger the distance from the lower surface of the package to the lower surface of the package, the smaller the thermal stress generated by the solder reflow at the time of mounting the printed circuit board. Sometimes insert
The resin flow is likely to be unbalanced (affects the flow balance). The flow imbalance causes void defects in which air remains in the package and defective lead frame deformation.

(2) With respect to the shape of the lead frame, if the lead frame pattern is set so that the chip and the lead frame are electrically connected efficiently, the lead frame pattern may be different depending on the pattern. -It is necessary to reduce the plate thickness of the frame, and the reflow due to resin flow unbalance during molding
It may cause defective deformation. Further, if the pattern is restricted, the distance from the bonding pad on the chip to the bonding position on the lead frame side becomes long, and the gold wire is easily bent by the resin flow at the time of molding.

This will be described in detail with reference to FIG. Figure 8
Is a graph showing the relationship between the resin flow front end position l ₁ above the chip and the resin flow front end position l ₂ below the chip with the chip position h as a parameter, and the unbalance degree k ( When l ₁ > l ₂ , k = 1−l ₂ / l ₁ , l ₁ <
When l ₂ is a graph showing the relationship between _{k = 1-l 1 / l} 2). h, l ₁ and l ₂ are shown in the upper part of the graph. In this figure, l ₁ and l ₂ are represented as the distance from the origin with the gate position as the origin, but the origin may be set freely and the tip of the chip may be the origin.

FIG. 8A shows the relationship obtained from the results of the flow analysis simulation.
A procedure for obtaining the relationship of FIG. 8A from the result of the operation will be described. FIG. 33A is a graph showing the change over time in the flow front end position, which is one of the results of the flow analysis. In FIG. 33 (a), the position of the flow front on the chip and
The time change with the position of the flow tip under the tip is shown.
From the result of FIG. 33 (a), when the flow fronts on the tip and the tip at the same time are plotted again, FIG.
It becomes like (b). In FIG. 33, at time t = 0.
The position for 8 is specifically shown.

FIG. 9 is a graph (a) showing the relationship between the thermal stress and time during solder reflow and the graph (b) showing the relationship between the chip position h and the maximum thermal stress, with the chip position h as a parameter. ).

As shown in FIG. 8B, the graph showing the resin flow unbalance degree k shows a downward convex shape, and the lower limit of the graph is the optimum chip position. On the other hand, since the graph showing the thermal stress in FIG. 9B shows the characteristic of monotonically decreasing, the value of the chip position h that minimizes the resin flow unbalance and the thermal stress are minimized. The value of the chip position h to be reduced does not match.

As described above, since the design specifications spanning a plurality of design departments span a plurality of evaluation items, the relationship between the evaluation items and the design specifications is grasped as a whole and the evaluation is performed. It is necessary to determine the specifications (values, etc.).

The product design specification evaluation system according to the present invention has a specialized field so that the repetition of the package design stage 302 to the trial production / evaluation stage 304 performed in the package development process shown in FIG. 6 can be realized on a computer. Based on the design specifications of past prototype data, the values of evaluation items calculated from these design specifications, and the defect phenomenon of the prototype data, It is intended to be understood not only from the perspective of

FIG. 10 is a conceptual diagram of a semiconductor plastic package development process using the product design specification evaluation system according to the present invention. As shown,
In this system, when the product specifications are determined (process 301) and the design proposal of the new package is input by the designer (process 301).
331), using the past prototype data accumulation 337 similar to the input design plan, package design evaluation such as lead frame design, structural design and material design, and mold development evaluation are performed sequentially (processing). 332), based on this evaluation result, the specifications of the design proposal are changed / determined, and the design solution (corrected design proposal) is output (process 336). Based on this design solution, prototypes and prototypes are evaluated (Process 30
4) If the result of this evaluation is good, the design solution is certified as a design plan (process 305), and the products are mass-produced by the certified design plan (process 306). At this time, the following problems occur.

(1) Since the package design is approaching the limit design in which the design margin cannot be taken sufficiently, the product specifications that satisfy the required specifications cannot be determined depending on the specification determination and change order. . In addition, since each design department has different specialized fields, it is not possible to fully understand the relationship between specifications related to multiple specialized fields and product defects.
The designer decides the evaluation contents and evaluation order based on his intuition and experience, and designs. The combination of optimum solutions of individual designs such as lead frame design, structural design and material design is not necessarily the optimum solution of the entire package design.

(2) When the evaluation simulation required for each design is performed, the criteria for judging whether a product defect occurs from the evaluation simulation result is not clarified, and the designer's personal It's up to the decision.

(3) When it is determined that a product defect will occur, the method for determining the parameters to be changed and the amount of change is not established.

The product design specification evaluation system according to the present invention aims to solve the above-mentioned problem of package design, and can comprehensively grasp the design specification parameters related to the design departments having different specialized fields. In order to ensure that the evaluation contents required by the design departments can be easily taken into consideration in relation to the design parameter,
This is made possible by effectively using the prototype data 337 of the already-developed product that is always obtained by 304. As a result, it is possible to eliminate the repetition from the package design stage to the trial production / evaluation stage shown in FIG. 6, which significantly shortens the design period and improves the efficiency of design / evaluation of design specifications related to the design department. It can be carried out well and the design quality can be greatly improved (of course, the present invention can exhibit the same effect in the mold development).

The product design specification evaluation system of the present invention will be described by taking as an example the case of being applied to the design of such a semiconductor plastic package. The outline of the processing procedure of this system is as shown in FIG. 28, and the details of the processing procedure are as shown in FIGS. Further, FIG. 27 shows an explanatory view showing an outline of processing of this system. The processing procedure will be described with reference to the hardware configuration shown in FIG. 1, the software configuration shown in FIG. 2 and the explanatory diagram shown in FIG.

As shown in FIGS. 11 to 14, first, the operator operates the keyboard 15 to enter the same design specifications as the design specifications such as package shape, structure, lead frame shape, molding conditions, material characteristics, and chip position. Design constraint data such as the changeable range of the plan and the size relationship between design specifications are input. These data are fetched into the product design specification evaluation system by the data input unit 22 based on the execution control of the control unit 21 as external input data (process 102). This process corresponds to the process 2801 of FIG.

Next, in process 2802 of FIG. 28, a defective causal relationship is generated. This processing will be described in detail with reference to FIG. In the defect causal relationship generation unit 26 in the evaluation procedure generation unit 41, based on the execution control of the control unit 21, from the design specifications and design constraints from the data input unit 22, the package shape, structure, material, manufacturing conditions. Constraints and feature quantities related to the prototype specifications such as are extracted and calculated (process 104). Then, all evaluation items managed by the design knowledge management unit 25 of the data management unit 40 and stored in advance in the disk device 18 are stored in the main storage device via the disk controller 17, the multi-bus 11, and the central processing unit 13. 14 and extract the constraints for each of these evaluation items with the prototype data that is managed by the prototype data management unit 24 in the data management unit 40 and stored in advance in the disk device 18. Prototype data having trial specifications similar to the feature amount calculated under the conditions, together with the calculated values of the evaluation items,
The data is loaded into the main memory 14 via the disk controller 17, the multibus 11, and the central processing unit 13 (process 106), and the type of defect in the prototype data, the occurrence rate, the correlation between the occurrence site and the evaluation item. Is calculated (process 108), and all two-element relationships between the defect phenomenon and the evaluation item are generated (processes 110 and 111). Regarding the evaluation items that form the relationship between these two elements,
In the design specifications of the new product, a two-element relationship between the evaluation item and the design parameter is generated from the parameter (design parameter) that affects the evaluation parameter and can be changed under the extracted constraint conditions ( processing
112), based on the relationship between the two elements of the failure phenomenon and the evaluation item and the relationship between the two elements of the evaluation item and the design specifications, a failure causal relationship representing the relationship between the failure phenomenon, the evaluation item, and the three elements of the design specifications is determined. Generate (process 114).

Next, as shown in the process 2803 of FIG. 28, an evaluation standard is generated. This process will be described in detail with reference to FIG. After the defective causal relation generating unit 26 generates the defective causal relation, the evaluation criterion generating unit 27, based on the execution control of the control unit 21, the design specification of the new package generated by the defective causal relation generating unit 26. For all the relationships between the two elements of the failure phenomenon and the evaluation item that constitute the failure causal relationship with respect to, the prototype data management unit 24 in the data management unit 40 manages each relationship. , The prototype data stored in advance in the disk device 18, which has a prototype specification similar to the calculated feature quantity under the extraction constraint condition, the package shape, structure, material, manufacturing condition, defect in the prototype data. Type, defect occurrence site data, together with the calculated value of the evaluation item, is taken out to the main storage device 14 via the disk controller 17, the multi-bus 11, the central processing unit 13, and the defect occurrence rate and the calculated value of the evaluation item Plot the relationship and generate a defect rate prediction graph ( 116, 118). Then, the calculated value of the evaluation item is divided into the range of the evaluation item having the defect occurrence rate of “0” and the range not thereof, and the range having the defect occurrence rate of “0” is the allowable range for the evaluation item, that is, the evaluation. A standard (standard for judging quality of design specification) is generated (processing 120, 122).

Next, in process 2804 of FIG. 28, the evaluation order is generated. This processing will be described in detail with reference to FIG. In the evaluation order generation unit 28 in the evaluation procedure generation unit 41, based on the execution control of the control unit 21, from the design specifications of the new package from the data input unit 22 and the constraints of the design specifications, the evaluation order generation unit 28 relates to the prototype specifications. Constraints and features are extracted and calculated,
For each evaluation item, prototype data managed by the prototype data management unit 24 in the data management unit 40 and stored in advance in the disk device 18 is calculated with the calculation feature amount under the extraction constraint condition. The disk controller 17, which is a prototype data having similar prototype specifications,
It is taken out to the main storage device 14 via the multi-bus 11 and the central processing unit 13, the defect rate for each defect type of the prototype data is calculated, and the defect rate is generated by the defect causal relationship generation unit 26. The failure rate is set as a failure rate for each failure phenomenon that constitutes a failure causal relationship (processes 124, 126, 128). Furthermore, for each two-element relationship between the failure phenomenon and the evaluation item that constitute the failure-causal relationship, the feature amount obtained from the input design specifications from the data input unit 22 includes the amount related to the evaluation item, and again From the disk device 18 to the disk control device 17, the trial data having the trial specifications similar to the feature amount under the constraint condition is provided.
It is taken out to the main memory device 14 via the multi-bus 11 and the central processing unit 13, and the defect rate for each type of defect related to the evaluation item is calculated, and the defect rate for each defect related to the calculated evaluation item is It is set as the failure occurrence rate of the two-element relationship between the failure phenomenon forming the relationship and the evaluation item (the rate at which the failure phenomenon occurs due to the evaluation item) (Process 13).
0). Then, the failure phenomenon with the maximum failure rate set for the failure phenomenon related to the failure cause-effect relationship is extracted, the evaluation item related to the failure phenomenon is extracted, the evaluation item is set as the first evaluation target, and the extracted evaluation is performed. When there are multiple items, the defect occurrence rate of the two-element relationship between the defect phenomenon and the evaluation item is large, and the evaluation item is not related to other defect phenomena or has a small relationship (the defect rate between the two elements is small. ) An evaluation procedure is generated so as to be evaluated in the order of evaluation items (process 132).

Next, the simulation is executed as in the process 2805 of FIG. This process will be described in detail with reference to FIG. In the design plan evaluation unit 42, in the design plan evaluation execution unit 29, based on the execution control of the control unit 21, according to the evaluation order of the evaluation items generated by the evaluation order generation unit 28 of the evaluation procedure generation unit 41, the evaluation item concerned. In order to obtain the value of, the simulation is performed via the simulation executing unit 31. Simulation execution unit
31, the simulation program required for the evaluation of the design plan is determined, and the required simulation program is selected from the simulation program group stored in the disk device 18 in advance and executed. Disk device 18 to disk controller 1 to put it in a ready state
7, a multi-bus 11, a central processing unit 13 through the main memory 14
Heroed and executed (process 134). As a result, the simulation executing unit 31 loads the loaded simulation.
In order to execute the simulation program, data of design specifications such as package shape, structure, lead frame shape, molding conditions, material characteristics, and chip position from the data input section 22 can be input to the simulation program. After converting to the format, run the simulation program (Process
136).

Here, a model example of a simulation program stored in advance in the disk device 18 is shown in FIG.
9, and will be described with reference to FIGS. 15 and 16. The simulation executing unit 31 shown in FIG. 2 stores the simulation program shown in FIG. Each program of FIG. 29 is a program for analyzing each analysis item of FIG. 15 by simulation. FIG. 15 and FIG. 16 are simulation analysis items related to the evaluation of formability, an image diagram of the analysis model, and a simulation by the analysis model.
It is an explanatory view showing the evaluation items obtained from the application, and the main design specifications to be evaluated and determined by the evaluation items. The contents are as follows.

(1) Tip Position Considering Balance Analysis Model This model is for the flow tip position, flow velocity, viscosity, etc. when the resin flow is divided into upper and lower parts at the insert part (chip, tab, etc.) during molding. It is a model that simulates the change of pressure with time, and can evaluate design specifications related to the flow path shape such as an appropriate chip position and chip thickness.

(2) Balance Analysis Model Considering Flow Control Plate This model is an extension of the balance analysis model of (1) above so that the effect of the flow control plate can be evaluated. The flow control plate controls the resin flow above and below the chip by installing a structure that acts as a flow resistance near the gate in order to eliminate the resin flow imbalance above and below the chip.
It is built into the product as part of the dframe.

(3) Balance analysis model considering laminated structure This model is an extension of the balance analysis model described in (1) above so that a laminated structure in which multiple chips are mounted in one package can be evaluated. Is.

(4) Balance Analysis Model Considering Side Flow This model is an extension of the balance analysis model described in (1) above so that the resin flow in the channels on the side as well as the top and bottom of the chip can be considered. .

(5) Chip stress analysis model This model is a model that simulates the time distribution of stress distribution, moment, and load from the pressure distribution obtained by the balance analysis model of (1) above. -Doffret-
It is a model that can evaluate design specifications such as the hanging position and plate thickness.

(6) Chip Position Displacement Analysis Model This model is an extension of the balance analysis model described in (5) above, and is a model that can reflect the time change of the displacement amount of the chip. Therefore, not only the design specifications of the lead frame, but also chip exposure, gold wire exposure, through voids, and lead frame deformation due to chip displacement can be evaluated.

The procedure for obtaining the value of the evaluation item using the above simulation program will be described with reference to FIG. First, a simulation program related to the desired evaluation item is selected (process 3002). Next, input data for simulation is generated (process 3004), and the selected simulation program is executed (process 300).
6). The results of the executed simulation programs are integrated to generate evaluation item values (process 3008).

For example, when the flow balance degree is taken as an evaluation item, the flow analysis program is selected as the simulation program (process 3002). Processing 3004,
After processing 3006, a flow balance (chip vertical flow balance) is calculated from the flow state (time change of distribution of concentration, viscosity, flow velocity, etc.) obtained as a result of simulation (processing 3008). The flow balance is in an optimal state as shown by the dotted line in FIG. 32, and how much the flow balance deviates becomes the point of evaluation.

In order to analyze the lead frame deformation (lead frame suspension position, plate thickness, etc.) shown in FIG. 17, a simulation program of the model shown in No. 5 and 6 shown in FIG. (Chip stress analysis program and chip displacement analysis program) will be used.

First, the contents of this model will be described with reference to the drawings. The resin flow state above and below the chip generated during molding is simulated, and the resulting pressure P ₁ , P ₂ , P ₃ , ... On the chip surface and the area ΔS _{1 of the} surface receiving the pressure. , ΔS ₂ , ΔS ₃ , ... From the stress P ₁ · ΔS ₁ , P
₂ · ΔS ₂ , P ₃ · ΔS ₃ , ... Are calculated, and the moment is calculated from the stress to calculate the chip displacement. The lead frame supporting the chip is deformed by the displacement of the chip.

Further, when the lead frame deformation amount is taken as the evaluation item, the flowchart of the process for obtaining the value of the evaluation item by simulation is as shown in FIG. First, the simulation program to be used is selected. In this case, the flow analysis program and the deformation analysis program are selected (process 3102). Next, input data for simulation is generated. First, input data for flow analysis is generated (process 3104). Then, the flow analysis simulation is executed (process 3106).

From the output result of the executed flow analysis simulation, the input data used in the next deformation analysis simulation is generated. That is, the load (force) distribution is obtained from the pressure distribution of the flow analysis result (process 3110). Then, the deformation analysis simulation by the load is executed (process 3111). From this simulation result, the time change of the deformation amount is obtained (process 3108).

As described above, the program relating to the formability can be evaluated by the program stored in the disk device 18 in advance.

Returning to the description of the processing procedure of the product design specification evaluation system according to the present invention, in the design plan evaluation executing unit 29 of FIG. 2, the evaluation standard generated by the evaluation standard generating unit 27 and the simulation executing unit 31 are used. By comparing with the value of the obtained evaluation item, it is determined whether or not the value of the evaluation item is within the evaluation criteria, and the new package from the data input unit 22
It is determined whether or not it is necessary to change the design specifications (see FIG. 27, it is determined that the design specifications need to be changed because the chip displacement amount n exceeds the reference). Process 138). This corresponds to the process 2806 in FIG.

If it is determined in this process that the design specification proposal need not be changed, the process returns to process 132, and the evaluation procedure generation unit 28 selects the next evaluation target. When it is determined that the change is necessary, the countermeasure plan generation unit 30 generates the countermeasure plan (process 142 in FIG. 13). This corresponds to the process 2807 in FIG.

Then, the margin of the evaluation item calculated as the difference between the evaluation standard and the evaluation item value is obtained, and a plurality of design parameters expressed in the defective causal relationship graph and related to the evaluation item are obtained. For each of the design specifications, the sensitivity, which is the change ratio of the value of the evaluation item to the change in the design specifications, is obtained (process 140 in FIG. 13). This is process 2 in FIG.
It corresponds to 808.

Next, in process 2809 of FIG. 28, a countermeasure plan is generated. This process will be described in detail with reference to FIGS. 13 and 14. In the countermeasure plan generation unit 30, the sensitivity of each design specification to the evaluation items obtained by the design plan evaluation execution unit 29 of the design plan evaluation unit 42 based on the execution control of the control unit 21 (the countermeasure plan generation unit 30 in FIG. 27). (Refer to the image diagram for explaining the sensitivity shown in)), the margin, and the relationship between the design specification and other evaluation items indicated by the causal relationship of defects, and the evaluation that has already been evaluated / measured. Design specifications that do not affect the item, or design parameters that have a large margin of the evaluation item and that have a large sensitivity (in the case of FIG. 27, lead frame
By selecting the suspension position P ₁ ) and determining the correction amount of each design specification so that the product sum of the sensitivity and the correction amount of each selected design specification is the protrusion amount of the evaluation item value. , A candidate of design specifications to be modified and a candidate of design specifications that holds the evaluation item value within the evaluation criteria are obtained (processing 144, 146, 14).
8). Then, the plurality of generated countermeasures are displayed on the screen of the display device 16 by the data output unit 23 based on the execution control of the control unit 21 and request the operator to select (process 150). .

In response to the request, the operator selects one from a plurality of countermeasures and gives an instruction using the keyboard 15. Next, based on the instruction from the control unit 21, the defective causal relationship generating unit 26 of the evaluation procedure generating unit 41 determines whether or not the evaluation / countermeasures have been completed for all the evaluation items in the defective causal relationship (processing. 152). If not completed, the defective rate of the defective causal relationship is improved by the information newly obtained by the simulation, and the process returns to the process 132 (process 15).
Four).

When the evaluation and countermeasures for all the evaluation items are completed, the design specifications of the new package have been optimally decided. The determined design specification draft is output as a design solution (process 156). This corresponds to the process 2811 of FIG.

As described above, the moldability relating to the internal structure design, the lead frame strength relating to the lead frame design,
Like molding conditions related to manufacturing process design, evaluation contents that are mutually related can be grasped together with design specifications, and evaluation items and evaluation orders can be determined according to the importance of design specifications to determine optimum design specifications.

As described above, according to the present invention, in the development of the semiconductor plastic package as shown in FIG. 6, the formability 3121 in the structural design 312 in the package design 302 and the lead in the lead frame design 313 are selected. The frame strength 3121 and the process design 314 have been specifically described by taking an example directly related to the part relating to the molding process 3142.

As can be seen from the above example, when designing a semiconductor plastic package, it is possible to generate a failure causal relationship that matches a failure occurrence state that can be predicted by past prototype data, and from the failure causal relationship, It is possible to grasp the evaluation contents necessary for obtaining the design solution of the new package and related specifications as a whole, and to systematically determine the evaluation items and the evaluation order according to the importance of the design specifications. is there. Further, the design specifications such as the shape, structure, material and manufacturing conditions of the new package can be comprehensively and efficiently evaluated and, if necessary, the design specifications can be efficiently and easily corrected.

Incidentally, FIGS. 24, 25 and 26 show another embodiment of the present invention. The system of this embodiment is shown in FIG.
28. The processing is almost the same as the flowchart of FIG.
28 is different from that of FIG. 28 in that the processing of (1) is performed, and that the countermeasure plan is generated and then fed back to the defective causal relationship generation. The three modes are: (1) Design proposal evaluation / countermeasure mode (normal processing for designing) (2) Evaluation standard generation mode (processing when generating evaluation standard) (3) Prototype data management / operation mode (Preparation process for enabling operation of this system), which corresponds to FIGS. 24, 25, and 26, respectively.

Further, FIGS. 24, 25 and 26 also show the data flow together with the processing procedure.

Further, in the product design specification evaluation system according to the present invention, the defect occurrence rate is "0" in the evaluation reference generating means.
The evaluation criteria are divided into a plurality of ranks in the range of the evaluation items that are "0", the range of the evaluation items that is greater than "0" and less than "1", and the range of the evaluation items that are "1" or more, and if necessary, defects occur. The design specification and the design specification of the input new product are provided by substituting the range in which the rate is "0" to "1" into a plurality of ranks and adding the function of dividing the evaluation standard into a plurality of ranks. Based on the above restrictions, the cause of defects is systematically avoided by the quantitative evaluation criteria obtained from the cause and rate of occurrence of defects that are implicitly included in the prototype data of developed products, and the shape of new products It is also possible to obtain a design solution by easily and efficiently evaluating the design specifications such as the structure, the material, the manufacturing conditions, etc. and appropriately determining them.

Further, a bus serving as a signal transmission path between the connected devices, a bus control device for controlling the bus, a central processing unit connected to the bus, and a disk control device,
In a workstation including a main storage device, a display device, a keyboard connected to the central processing unit, and a disk connected to the disk control device, an input means, a prototype data management means, a failure causal relationship generation means The evaluation order generation means and the evaluation execution means are stored in the same main storage device, and the trial production data managed by the trial production data management means is stored on the disk, and the central processing unit performs the main processing. It is also possible to execute each means on the storage device.

Further, a communication control device which is connected to the bus and transmits data sent from the bus to another device and receives data from the other device is added, and data communication is performed with the workstation through the modem. A computer comprising a bus, a bus controller for controlling the bus, a central processing unit, a disk control unit, a communication control unit connected to the bus, a main storage unit connected to the central processing unit, and a disk connected to the disk unit. In a workstation / computer device connected to the above, the input means, the prototype data management means, the defective causal relationship generation means, the evaluation order generation means, and the evaluation execution means according to claim 1 are stored in the main storage device on the workstation side. A defective causal relationship generating means according to claim 1,
The evaluation order generation means and the evaluation execution means are stored in the main storage device on the computer side and held by the prototype data management means,
By storing the prototype data to be managed on the disk on the computer side, the response performance can be improved by using the central processing unit on the computer side according to the processing load when executing the respective means. It is also possible to deal with large-scale prototype data.

[0149]

As described above, the product design specification evaluation system according to the present invention has the following effects.

(1) It is possible to generate a failure causal relationship in accordance with a failure occurrence situation that can be predicted by a design plan and past prototype data, and evaluate contents necessary for obtaining a design solution from the failure causal relationship and related design various matters. The elements can be grasped as a whole, and the evaluation items and the evaluation order according to the importance of the design specifications can be systematically determined.

(2) According to the input design plan and the failure occurrence state that can be predicted by the past prototype data so that the evaluation contents necessary for obtaining the design solution and the related design specifications can be grasped as a whole. A bad causal relationship can be generated.

(3) In order to systematically determine the evaluation items and the evaluation order according to the importance of design specifications for obtaining the design solution, it is possible to make predictions based on the input design plan and past prototype data. It is possible to generate a defect causal relationship that matches the defect occurrence situation.

(4) Since the evaluation result of the design proposal, the process of modifying the design specifications and the modification result can be sequentially left together with the generated defect causal relationship and the accompanying information, the correction specifications for avoiding the predicted defect can be obtained. Since the intention regarding the modification amount can be accumulated and reused, it is possible to easily grasp the influence of the specification change on the evaluation results of other specifications, so it is easy to improve the efficiency of design / evaluation / decision after the next time. Can be achieved.

(5) Based on the design specifications such as the shape, structure, material and manufacturing conditions of the new product and the constraints of the design specifications, the cause of the failure which is implicitly included in the prototype data of the developed product is the failure phenomenon. , Evaluation items, and causal relationships with design specifications,
Systematically avoid the cause of defects, comprehensively and efficiently evaluate the design specifications of the new product such as shape, structure, material, manufacturing conditions, etc., and if necessary, correct the design specifications efficiently and easily. It is a reward.

(6) Efficiently judge whether the inputted design specification proposal of the new product is good or bad from the registered prototype data of the developed product, and generate an appropriate correction proposal of the design specification if necessary. It is possible to quantitatively determine the evaluation standard as the allowable range of the analysis value of the evaluation item.

(7) It is possible to efficiently evaluate the design specifications of a new product and determine the order in which the new product design specifications are to be evaluated according to the types of defects in the prototype data of the already-developed product, the occurrence rate, and the location of occurrence. it can. (8) Using the prototype data necessary to efficiently evaluate the design proposal of the new product, the defect phenomenon related to the new product,
It is possible to easily generate a defective causal relationship between evaluation items and design specifications.

(9) The sensitivity required for determining the change candidate and the change amount can be efficiently obtained. (10) Change candidates and change amounts can be efficiently determined in design specifications.

(11) Large-scale prototype data can be used,
Based on the design specifications such as the shape, structure, material and manufacturing conditions of the new product and the constraints of the design specifications, the cause of the failure that is implicitly included in the prototype data of the developed product is the failure phenomenon, the evaluation item, the design It can be extracted based on the causal relationship with the origin, systematically avoids the cause of defects, can efficiently evaluate the design specifications of new products such as shape, structure, material, manufacturing conditions, etc., and if necessary, design appropriately The specifications can be modified efficiently and easily.

As described above, a product design specification evaluation system capable of objectively selecting the evaluation items for the design specification proposal, the evaluation order of the selected evaluation items, and the evaluation criteria for each evaluation item is provided. Can be provided.

After evaluating the evaluation items based on the determined evaluation criteria and grasping the possibility of occurrence of defects, the design parameters to be specifically changed or the amount of change of the design parameters are determined. It is to provide a product design specification evaluation system capable of objectively performing.

Further, a work station device capable of operating the product design specification evaluation system and a host computer are provided, and high-speed and highly accurate numerical calculation and data handling are performed on the host computer side. It is possible to provide a work station system capable of operating the above-mentioned product design specification evaluation system while carrying out interactive processing with the user on the work station device side.

[Brief description of drawings]

FIG. 1 is a hardware configuration diagram of a product design specification evaluation system according to the present invention.

FIG. 2 is a software configuration diagram of a product design specification evaluation system according to the present invention.

FIG. 3 is an explanatory diagram showing a relationship between a state transition to a defect occurrence and a defect causal relationship.

FIG. 4 is an explanatory view showing the structure of a semiconductor plastic package.

FIG. 5 is an explanatory diagram of a quality problem of a semiconductor plastic package.

FIG. 6 is an explanatory view of a development process of a semiconductor plastic package.

FIG. 7 is an explanatory view of a mold for molding, a molding machine and a molding process.

FIG. 8 is an explanatory diagram of a relationship between a tip position and a flow balance.

FIG. 9 is an explanatory diagram of the relationship between chip position and stress.

FIG. 10 is a conceptual diagram of a development process of a semiconductor plastic package using the product design specification evaluation system according to the present invention.

FIG. 11 is a detailed flowchart showing a processing procedure according to the embodiment of the present invention.

FIG. 12 is a detailed flowchart showing a processing procedure according to the embodiment of the present invention.

FIG. 13 is a detailed flowchart showing a processing procedure according to the embodiment of the present invention.

FIG. 14 is a detailed flowchart showing a processing procedure according to the embodiment of the present invention.

FIG. 15 is an explanatory diagram of a simulation model for flow analysis and deformation analysis.

FIG. 16 is an explanatory diagram of a simulation model for flow analysis and deformation analysis.

FIG. 17 is an explanatory view of the principle of lead frame deformation analysis.

FIG. 18 is an explanatory diagram of a defect causal relationship.

FIG. 19 is an explanatory diagram of a two-element relationship between a defect phenomenon and an evaluation item.

FIG. 20 is an explanatory diagram of a two-element relationship between evaluation items and design specifications.

FIG. 21 is an explanatory diagram showing an example of a defect rate prediction graph.

FIG. 22 is an explanatory diagram of a processing outline of evaluation criterion generation.

FIG. 23 is an explanatory diagram of a reference mode in evaluation standard generation.

FIG. 24 is an explanatory view of another embodiment of the present invention (design proposal evaluation /
Countermeasure mode).

FIG. 25 is an explanatory diagram of another embodiment of the present invention (evaluation reference generation mode).

FIG. 26 is an explanatory diagram of another embodiment of the present invention (prototype data management / operation mode).

FIG. 27 is an explanatory diagram of a process outline of a product design specification evaluation system.

FIG. 28 is a schematic flowchart showing the processing procedure of the embodiment of the present invention.

FIG. 29 is a functional block diagram of a simulation program.

FIG. 30 is a flowchart showing an example of processing for obtaining an evaluation item value by a simulation program.

FIG. 31 is a flowchart showing another example of processing for obtaining an evaluation item value by a simulation program.

FIG. 32 is an explanatory diagram of an optimum state of flow balance (chip vertical flow balance).

FIG. 33 is an explanatory diagram of a case where an evaluation item value of a chip vertical flow balance degree is obtained from a flow analysis simulation result.

FIG. 34 is an explanatory diagram of sensitivity.

FIG. 35 is an explanatory diagram of sensitivity and correction amount.

FIG. 36 is an explanatory diagram of a method of expressing sensitivity.

FIG. 37 is an explanatory diagram of a sensitivity calculation method by software.

FIG. 38 is an explanatory diagram of conditions for performing sensitivity calculation.

FIG. 39 is an explanatory diagram of determination of a correction amount.

FIG. 40 is an explanatory diagram of a method for correcting design specifications.

FIG. 41 is an explanatory diagram of a method for correcting design specifications.

[Explanation of symbols]

11 ... Multi-bus, 12 ... Bus control device, 13 ... Central processing unit, 14 ... Main storage device, 15 ... Keyboard, 16 ...
Display device, 17 ... Disk control device, 21 ... Control unit, 22 ... Data input unit, 23 ... Data output unit, 24
Prototype data management unit, 25 Design knowledge management unit, 26 Defect causal relationship generation unit, 27 Evaluation standard generation unit, 28 Evaluation sequence generation unit, 29 Design plan evaluation execution unit, 30 Proposed measure generation Section, 31 ... simulation execution section, 40 ... data management section, 41 ... evaluation procedure generation section, 42 ... design plan evaluation section, 3
31 ... Package design proposal, 336 ... Package design solution,
500 ... Conventional structure package, 501 ... Chip, 502
... lead frame, 503 ... resin, 504 ... tab, 5
05 ... Gold wire, 506 ... Bus bar, 507 ... Insulating film, 508 ... LOC structure package, 601 ... Void,
602 ... Frame deformation, 603 ... Wire bending, 604 ...
Package crack, 605 ... Adhesive peeling, 301 ... Product specification determination, 302 ... Package design, 303 ... Mold development, 304 ... Prototype / evaluation, 305 ... Certification, 306 ... Mass production, 311 ... External design, 312 ... Internal Structural design, 313
... lead frame design, 314 ... manufacturing process design, 3
15 ... Material design, 3121 ... Moldability, 3122 ... Thermal stress, 3123 ... Heat dissipation, 3131 ... Lead frame pattern
3132 ... Lead frame strength, 3141 ... Bonding process, 3142 ... Molding process, 3151 ...
Resin material, 3152 ... Lead frame material, 401 ... Chip, 402 ... Gold wire, 403 ... Lead frame, 404 ...
Cavity, 405 ... Lower mold, 406 ... Runner, 407 ...
Gate, 408 ... Heater, 409 ... Upper mold, 410 ... Pot, 411 ... Resin, 413 ... Plunger, 421 ... Product, 331 ... Design proposal, 332 ... Specification evaluation, 333 ... Re-
Draft design, 334 ... Structural design, 335 ... Material design, 336 ... Design solution, 337 ... Prototype data, 338 ... Designer, 901 ... Void, 902 ... Gold wire exposure, 903 ... Gold wire deformation, 911 ... Red frame deformation amount, 912 ... Flow balance degree, 913 ... Viscosity, 914 ... Velocity, 920 ... Chip position, 921 ... Chip thickness, 922 ... Red frame thickness, 923 ... Red frame suspension Position, 924 ... Insulating film size, 925 ... Insulating film thickness, 926 ...
Gold wire length, 927 ... Gold wire thickness, 928 ... Gold wire position, 92
9 ... Gate shape.

 ─────────────────────────────────────────────────── ─── Continuation of front page (72) Kunihiko Nishi Nishi 5-20-1, Kamimizuhonmachi, Kodaira-shi, Tokyo Inside semiconductor design development center, Hitachi, Ltd. (72) Inventor Toshihiro Yasuhara Komizu, Kodaira-shi, Tokyo Honcho 5-chome No. 20-1 Incorporated company Hitachi Ltd. Semiconductor Design Development Center

Claims (14)

[Claims] 1. Input means for inputting a product design specification proposal including at least one of a product shape, structure, material, manufacturing condition and design constraint, and a prototype specification which is a design specification of a product already prototyped. And a prototype design data managing means for preliminarily holding and managing prototype data consisting of defect information including at least one of defect type, occurrence rate and occurrence site for each prototype specification, and a product design specification plan. A simulation for performing analysis for each of the evaluation items that should be analyzed when evaluating.
A design knowledge management means having an application program, and a feature amount of the product extracted from the product design specification draft, and the trial production data having the feature amount similar to the feature amount is produced as the trial production data
The first of the evaluation items and failure phenomena
And the second relationship between the design items which are the design data of the product according to the product design specification and the evaluation item, and the defect phenomenon, the evaluation item and the design item are determined from the first and second relationships. Defect causal relationship generating means for generating a defective causal relationship that is a relationship between the three elements, and the defect causal relationship obtained from the defective causal relationship generating means, the magnitude of the defect occurrence rate, the strength of design constraints, and the difficulty of modifying the specifications. Considering at least one of them, an evaluation order generating means for generating an order of evaluation items to be analyzed, and a simulation stored in the design knowledge managing means in accordance with the order generated by the evaluation order generating means. A design proposal evaluation means for analyzing evaluation items using a program and determining whether or not the values of design specifications of the product design specification proposal input based on the analysis result are necessary to be modified, Product design specification evaluation system and butterflies. 2. The defect causal relationship generating means according to claim 1, wherein the first relationship is generated based on a correlation between a failure phenomenon of the prototype data and a value of an evaluation item, and the second relationship is determined. A product design specification evaluation system, which is generated based on a relationship between an evaluation item and design specifications managed by the design knowledge management means and a constraint condition of the design specification proposal. 3. The evaluation order generating means according to claim 1,
In the causal relationship of defects, a sequence of evaluation items to be evaluated in order of magnitude of defect occurrence rate is generated, and when the defect occurrence rates are equal, the correlation between the defect phenomenon and the evaluation item is obtained, and the strength of the correlation is calculated. A product design specification evaluation system characterized by generating an order of evaluation items to be evaluated in order. 4. The measure plan according to claim 1, wherein when the design plan evaluation means determines that the values of design specifications need to be modified, the design specifications to be modified and the amount of modification are determined. A product design specification evaluation system comprising a generation means. 5. The method according to claim 1, wherein for each of the prototype data,
A failure rate prediction graph is generated based on the relationship between the value of the evaluation item obtained by simulation based on the design specifications of the prototype data and the failure rate, and the failure rate is "0" in the graph. The design plan is provided with an evaluation criterion generating unit that obtains an evaluation criterion value range, and generates an evaluation criterion with the evaluation item value range in which the defect occurrence rate is “0” as an allowable range for the evaluation criterion. When the analysis result is within the allowable range, the evaluation means determines that it is not necessary to correct the values of the design specifications of the product design specification proposal, and the analysis result is not within the allowable range. The product design specification evaluation system is characterized by determining that the values of the design specifications of the product design specification proposal need to be corrected. 6. The design plan evaluation executing means according to claim 4, wherein the design specification variation is input to the simulation,
The sensitivity, which is the variation ratio of the value of the simulation result to the variation, is calculated, and the countermeasure plan generating means, the sensitivity, the margin that is the difference between the value of the simulation result and the evaluation standard, and the cause of failure. From the relationship between the design specifications shown in the relationship and other evaluation items,
Design specifications that do not affect the evaluation items that have already been evaluated / measured,
Alternatively, if there is an impact, select a design parameter with a large sensitivity in the design parameter with a large margin of the evaluation item, and the product sum of the sensitivity and the correction amount of each selected design parameter is the evaluation item value. A product design specification evaluation system characterized in that the correction amount of each design specification is determined so that the standard becomes equal to the protruding amount. 7. The evaluation order generation means according to claim 3, wherein the evaluation order of the evaluation items related to the failure phenomenon having the maximum failure rate in the failure causality is first, and the failure phenomenon having the maximum failure rate. When there are a plurality of evaluation items related to, the evaluation order of the evaluation item having the maximum correlation with the failure phenomenon having the maximum failure rate is first among the plurality of evaluation items, and the maximum failure is When there are a plurality of evaluation items that have the highest correlation with the defective phenomenon having the rate, the evaluation order of the evaluation item that has the smallest relation with other defective phenomena is first among the plurality of evaluation items. A product design specification evaluation system characterized by the following. 8. A product design specification proposal including at least one of a product shape, structure, material, manufacturing condition and design constraint, and a prototype specification which is a design specification of a product already prototyped, Evaluation items to be analyzed when evaluating and proposing a product design specification proposal in advance by preserving and managing prototype data consisting of defect types for each prototype specification, occurrence rate, and defect information including at least one of the occurrence sites. For each of the
Application program is provided, the feature quantity of the product is extracted from the product design specification proposal, and the prototype data having the feature quantity similar to the feature quantity is created as the prototype data.
The first of the evaluation items and failure phenomena
And the second relationship between the design items which are the design data of the product according to the product design specification and the evaluation item, and the defect phenomenon, the evaluation item and the design item are determined from the first and second relationships. The defect causal relationship which is the relationship between the three elements is generated, and at least one of the defect occurrence rate, the strength of the design constraint, and the difficulty of modifying the specifications is determined from the defect causal relationship obtained from the defect causal relationship generating means. In consideration of this, the order of the evaluation items to be analyzed is generated, the evaluation items are analyzed in accordance with the order using the simulation program, and the design specifications of the product design specification input based on the analysis result are input. A product design specification evaluation method characterized by determining whether or not the original value needs to be corrected. 9. The method according to claim 8, wherein the first relationship is generated based on a correlation between a defect phenomenon of the prototype data and a value of an evaluation item, and the second relationship is defined as the managed evaluation item. A product design specification evaluation method, wherein the defective causal relationship is generated based on a relationship with design specifications and a constraint condition of the design specification proposal. 10. In the defect-causal relationship according to claim 8, a sequence of evaluation items to be evaluated in order of magnitude of defect occurrence rate is generated, and when the defect occurrence rates are equal, correlation between the defect phenomenon and the evaluation item is generated. Is calculated and an order of evaluation items for evaluating in order of the strength of the correlation is generated, and a product design specification evaluation method. 11. The product design according to claim 8, wherein when it is determined that the values of the design specifications need to be modified, the design specifications to be modified and the amount of modification are determined. Specification evaluation method. 12. The defect rate according to claim 8, based on the relation between the value of the evaluation item obtained by simulation based on the design specifications of the prototype data and the defect occurrence rate for each prototype data. A prediction graph is generated, the range of the value of the evaluation item having the defect occurrence rate of “0” is calculated in the graph, and the range of the value of the evaluation item having the defect occurrence rate of “0” is set as an allowable range for the evaluation item. As an evaluation criterion, when the analysis result is within the allowable range, it is determined that the values of design specifications of the product design specification proposal need not be corrected, and the analysis result is within the allowable range. If not, the product design specification evaluation method is characterized in that it is determined that the values of the design specifications of the product design specification proposal need to be corrected. 13. The variation of design specifications input to the simulation according to claim 11,
The sensitivity that is the variation ratio of the value of the simulation result to the variation is obtained, and the sensitivity, the margin that is the difference between the value of the simulation result and the evaluation standard, and the design specifications shown in the failure causal relationship Based on the relationship between the evaluation items and other evaluation items, the design parameters that do not affect the evaluation items that have already been evaluated and taken countermeasures, or if there is an impact, the design parameters that have a large margin for the evaluation items Select a large design parameter and determine the correction amount for each design parameter so that the product sum of the sensitivity and the correction amount for each selected design parameter is approximately equal to the standard protrusion value of the evaluation item value. Characteristic product design specification evaluation method. 14. The evaluation item according to claim 10, wherein an evaluation order of evaluation items related to a defect phenomenon having a maximum defect rate in the defect causality is first, and a plurality of evaluation items related to the defect phenomenon having the maximum defect rate are provided. If present, the evaluation order of the evaluation item having the largest correlation with the failure phenomenon having the maximum failure rate is the first among the plurality of evaluation items, and the evaluation order having the maximum failure rate is the first. When there are a plurality of evaluation items having the maximum correlation, the evaluation order of the evaluation item having the smallest relation with other defective phenomena among the plurality of evaluation items is the first evaluation order. Product design specification evaluation method.