JP2002157011A - Process management method and device for process - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a process management device and process management method for adaptively and most suitably managing the main process of a process on the basis of the measurement data in the preprocess or after-process of the process when the main process has any relation with the preprocessor after- process. SOLUTION: This process management device comprises a measurement part for measuring data in each of the main process and the preprocess or after-process; a calculation part for outputting the expected value, upper limit value and lower limit value of the data measured by the measurement part of the main process on the basis of a relational equation between the measurement data; and a judgment part for judging the quality of the data measured by the measurement part of the main process on the basis of the output value of the calculation part.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a process management apparatus and a process management method for optimally managing the availability of process conditions or inspection results in a manufacturing process including a plurality of processes by using a process control chart.

[0002]

2. Description of the Related Art In a process such as a semiconductor diffusion process or a liquid crystal array process which is manufactured through a plurality of processes,
Process conditions and inspection results are managed to improve product quality. However, in the conventional management method, the setting conditions in each process or the evaluation of the inspection result are not associated with each other, and each process is managed by an independent process control chart. For example, in a manufacturing process in which multiple processes are connected in a series, process conditions or semi-finished products or products are measured in each process, and each measured data is compared with each fixed target value, upper limit value, and lower limit value. The operator automatically determines the quality or displays the result of comparison, and the operator determines the quality.

[0003] Conventionally, for example, different judgment criteria are set or can be set in different places or different management departments. Therefore, the quality of process conditions or semi-finished products is judged by a uniform judgment criteria. Often not. In addition, even if the same manufacturing process is used, if a plurality of manufacturing processes are located in different places, the process is often controlled based on different criteria.

However, the conventional method cannot be said to be the best management method when the process conditions and the like are mutually related between a plurality of steps included in the manufacturing process.
For example, even if variations in the process conditions and inspection results of semi-finished products in each of a plurality of processes accumulate, the variations in the process conditions and inspection results of semi-finished products in each process are narrow so as not to produce defective products. Suppose we manage to the allowable limit. According to such a management method, if the inspection result of a semi-finished product in a certain process is good and the process condition of the next process exceeds an allowable limit, a good product can be completed from the semi-finished product by a combination of these. Also, there is a possibility that the semi-finished product is discarded as a defective product because the measurement data of the process conditions in the next step exceeds the allowable limit. As a result, the product yield deteriorates, and the cost of the product increases.

[0005] For example, it is assumed that measurement data of process conditions in a certain process is lower than a target value. The target value (fixed value) of the inspection in the next step has been determined on the assumption that the process is executed under the target value process condition in the certain step. However, if the measurement data of the process conditions in a certain process is lower than the target value, a non-defective product can be finally completed if the target value of the inspection result of the semi-finished product in the next process is slightly higher than the fixed value. Or the quality of the finished product may be better. However, with the conventional process management method,
Since the management of each process was performed at a fixed target value, the target value or the allowable limit such as the upper limit or the lower limit could not be adaptively changed, and the yield of the product could not be improved.

As described above, in the conventional process control apparatus or process control method, each process condition is set within an unnecessarily strict allowable range (or not optimal), or in an unnecessarily strict allowable range (or non-optimal allowable range). Or (to the extent possible) control of test results. In addition, there is a possibility that the judgment of the quality of the process condition or the semi-finished product is erroneous. Further, for example, since the management method is different in different places or different management departments, it is difficult to secure a certain quality in all places or management departments and to improve the quality at an early stage.

[0007]

As described above, with the conventional process control method, it is difficult to perform optimal control, and it is difficult to ensure constant quality and to improve quality at all places or control departments at an early stage. There was a problem. The present invention provides a process management for adaptively and optimally managing the present process based on measurement data in the previous process or the subsequent process when there is some relationship between the main process and the previous process or the subsequent process of the process. An object is to provide an apparatus and a process control method. Another object of the present invention is to provide a process management apparatus and a process management method for performing the same management that can ensure a certain quality and improve quality at an early stage even in different places or different management departments.

[0008]

In order to solve the above-mentioned problems, the present invention has the following arrangement. The invention according to claim 1 of the present invention provides a first method in a manufacturing process including a plurality of processes.
A first measurement unit that measures data in the step of, a second measurement unit that measures data in a second step that is a pre-step or a post-step of the first step, and the first measurement unit It has a relational expression between the measured data and the data measured by the second measurement unit, and the data measured by the second measurement unit and the data measured by the first measurement unit based on the relational expression A calculating unit that outputs at least one of an expected value, an upper limit value, and a lower limit value, and a determining unit that determines whether the data measured by the first measuring unit is good or bad based on an output value of the calculating unit. And a step management device.

According to a fifth aspect of the present invention, there is provided a first measuring step of measuring data in a first step in a manufacturing process including a plurality of steps, and a step before or after the first step. A second measurement step of measuring data in a second step, and a relational expression between the data measured in the first measurement step and the data measured in the second measurement step. A calculating step of outputting at least one of an expected value, an upper limit value, and a lower limit value of data measured in the first measuring step based on the data measured in the measuring step and the relational expression; A judging step of judging the quality of the data measured in the first measuring step based on the output value in the step;
It is a process management method characterized by having the following.

According to the present invention, when there is some relation between each process or each measurement data of a plurality of processes in a manufacturing process, it is possible to realize a process management apparatus or a process management method for performing optimal process management. Having. In a manufacturing process such as a semiconductor diffusion process and a liquid crystal array process which are manufactured through a plurality of processes, process conditions and the like of each process are closely related to each other. Therefore, the present invention is particularly suitable for a process management apparatus and a process management method for a semiconductor diffusion process or the like. By applying the present invention to a process control device or a process control method for a semiconductor diffusion process or the like, the manufacturing process conditions in each process and the criterion for determining whether or not an inspection result (measurement data) as a result of the manufacturing process can be determined. It is possible to realize a management of adaptively changing based on measurement data in a process or a post-process, and it is possible to improve a product quality of a product, a production yield, and the like.

For example, if the previous step (the second step described in the claims) is a process apparatus, measurement data of process conditions is used. If the previous step is an inspection apparatus, measurement data of inspection results is used. To plot. To judge the quality of the measurement data in the present step based on a numerical relational expression between the measurement data in the present step (first step described in claims) and the measurement data in the previous step Calculate a process management reference value such as an expected value (sometimes called an average value or a target value), an upper limit value, or a lower limit value. As a result, the process management reference values such as the expected value of the present process are set adaptively according to the situation of the previous process. This step is managed based on the expected value or the like. The same applies to the case where the second step described in the claims is a later step and the first step described in the claims is the present step.

Although a technique for adaptively changing the parameters of the process in the present process in order to produce a certain test result in accordance with the result of the test in the previous process has been known, the management standard of the process in the present process is known. (For example, the upper limit value) was always constant. The present invention is characterized in that the control standard itself in this step is changed according to the inspection result in the previous step.
"Expected value" means the predicted average value.

According to a second aspect of the present invention, there is provided a first measuring section for measuring data in a first step in a manufacturing process including a plurality of steps, and a step before or after the first step. A second measuring unit that measures data in the second step, and a relational expression between the data measured by the first measuring unit and the data measured by the second measuring unit. Based on the data measured by the measuring unit and the relational expression.
A calculation unit that outputs at least one of an expected value, an upper limit value, and a lower limit value of data measured by the measurement unit, and displays an output value of the calculation unit and data measured by the first measurement unit. And a display unit for performing the process.

According to a sixth aspect of the present invention, there is provided a first measuring step of measuring data in a first step in a manufacturing process including a plurality of steps, and a step before or after the first step. A second measurement step of measuring data in a second step, and a relational expression between the data measured in the first measurement step and the data measured in the second measurement step. A calculating step of outputting at least one of an expected value, an upper limit value, and a lower limit value of data measured in the first measuring step based on the data measured in the measuring step and the relational expression; A display step of displaying an output value in the step and data measured in the first measurement step.

In the invention as set forth in claim 1, the process management device adaptively changes a management reference value such as an upper limit value based on the data in the second step, and based on the management reference value such as the upper limit value. Then, the propriety of the measurement data in the first step (pass / fail of semi-finished products, etc.) was automatically determined. In the invention of claim 2 or the like, the process management device adaptively changes the management reference value such as the upper limit value based on the data in the second process, and the process management device sets the management reference value such as the upper limit value and the like. A process control chart including the measurement data in the first process is automatically displayed on a display screen, for example, so that the operator can make an optimal decision. The present invention provides a process management device or a process management device that performs optimal process management when there is some relation between each process or each measurement data of a plurality of processes in a manufacturing process, similarly to the invention of claim 1 and the like. It has the effect that the management method can be realized. The present invention is particularly suitable for manufacturing steps such as a semiconductor diffusion step and a liquid crystal array step which are manufactured through a plurality of processes.

For example, when the measurement data in the previous step (the second step described in the claims) is plotted, the present step (the first step in the claims) is performed based on the plotted measurement data. A process management reference value such as an expected value, an upper limit value or a lower limit value for judging the quality of the measurement data in the step (1) is calculated. By displaying the measurement data in this step, the expected value, and the like on the screen, it is possible to easily perform the step management of this step based on the optimal expected value, upper limit value, and lower limit value. The same applies to the case where the second step described in the claims is a later step and the first step described in the claims is the present step.

In the invention according to a third aspect of the present invention, the process management device has one centralized management unit including the calculation unit, and a plurality of the manufacturing processes are respectively performed by the first measurement unit and the first measurement unit. It has the 2nd measurement part, and the 1st measurement part and the 2nd measurement part of each of a plurality of the manufacturing processes, and the centralized management part are connected by a communication line, It is characterized by the above-mentioned. A process management apparatus according to claim 1 or 2.

According to the present invention, by centrally managing a plurality of manufacturing processes, it is possible to realize a process management apparatus capable of securing a constant quality even in different places or different management departments and improving quality early. It has the action of:

For example, a centralized management unit (generally a computer for centralized management) includes a pre-process (a second process in claims) and a main process (second process in claims) of a plurality of manufacturing processes at remote locations. The measurement data in the first step described) is obtained through a communication line, and a process management reference value such as an expected value, an upper limit value, or a lower limit value for judging the quality of the measurement data in each main process is calculated. Next, the centralized management unit sends the measurement data in this process and the data such as the expected value to a plurality of manufacturing processes via a communication line. The display unit arranged in each manufacturing process displays bitmap image data (process control data) for displaying the measurement data in the present process and the expected value on the display based on the data sent from the central management unit. (Fig.)
Is generated, and the bitmap image is displayed on the display.

The operator looking at the display judges the quality of the measurement data in this step. The determination unit or the display unit may be arranged in each manufacturing process as described above, or may be included in the central management unit. The latter is suitable when the central management unit manages all processes including control of process conditions of each process. The above process management device manages the manufacturing processes of different places or different management departments based on process management reference values such as expected values calculated by a single calculation unit. Therefore, the quality of the measurement data such as the process conditions of the manufacturing process in all places and the inspection results of semi-finished products and the like is judged by the judgment criteria generated by the single calculation unit. The process management of each manufacturing process is substantially unitarily managed. As a result, it is possible to ensure constant quality and to improve quality at an early stage in a plurality of manufacturing processes arranged at a plurality of places. The second step described in the claims is a post-process and the first step described in the claims is performed.
The same applies to the case where this step is performed in this step.

According to a fourth aspect of the present invention, the process management apparatus has one centralized management unit having a program for driving the calculation unit, and a plurality of the manufacturing processes are respectively performed by the calculation unit. , The first measuring unit and the second measuring unit, each of the calculation unit of the plurality of manufacturing processes, the centralized management unit is connected by a communication line, the centralized management unit, The process management apparatus according to claim 1, wherein the program is transmitted to each of the calculation units through the communication line.

According to the present invention, a plurality of manufacturing processes are centrally managed, and a calculation unit and the like are arranged in each manufacturing process, so that a centralized management computer at a remote place and a computer at a place where the process exists are always connected. Even if it is not, there is an effect that it is possible to realize a process management device that can secure a certain quality in different places or different management departments and can improve the quality at an early stage.

For example, if it is necessary to change the relational expression for calculating the expected value or the like in the present step (first step in the claims), the central management unit transmits a program including the changed relational expression to the communication line. Through the calculation section arranged in each manufacturing process. Each calculation unit calculates an expected value and the like of this process according to the updated program (program sent from the central management unit). As a result, the process management of the manufacturing processes arranged at different locations can always be made uniform, and a new version of the process management can be quickly introduced into each manufacturing process without omission. For example, if you create a new factory and lay the same manufacturing processes as other factories,
The centralized management unit sends a program including a relational expression that has been used in another manufacturing process to a calculation unit arranged in a new manufacturing process via a communication line. As a result, the process management of the new manufacturing process can be promptly managed at the same high quality level as the process management of the existing manufacturing process.

[0024]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a preferred embodiment of the present invention. << Embodiment 1 >> Hereinafter, a method and an apparatus for managing a process according to an embodiment of the present invention will be described with reference to the drawings. In addition, “this step” in the description of the specification means a step in which the process management apparatus or the process management method of the present invention performs a process management unique to the present invention. The “pre-process” means a process before the “main process”, and the “post-process” means a process after the “main process”. "This process" may not be the main process,
For example, the “post-process” may be a main process than the “main process”.

FIG. 1 is a block diagram of a process management method for the process steps according to the first embodiment. 10 is a pre-process control procedure, 20 is a pre-process to main process definition procedure, 30 is a pre-process to main process calculation procedure, 12 is a post-process process control procedure, 22 is a post-process to main process definition procedure, and 32 is a post-process. This is the process calculation procedure. First, in the pre-process control procedure 10, the quality of the measurement data X in the pre-process (the second process described in claims) is determined. If the pre-process is a process device, the measurement data of the process conditions is plotted on the process control chart of the pre-process, and the measured data of the inspection result is managed on the process control chart of the pre-process.

Next, in the pre-step to main step definition procedure 20, a relational expression C1 = f (X, X) between the measurement data X of the previous step and the measurement data Y of the present step (first step in the claims).
Y) (eg, C1 = 2X + Y). For example, C1 is a target value of specific measurement data that fluctuates based on the values of X and Y. The definition procedure is generally executed by an operator inputting a relational expression into a computer using a keyboard or the like. Next, in the preceding step to the present step calculation procedure 30, the measured data X of the preceding step and the defined relational expression C1 = f (X, Y) are used (C1 is a target value that is a constant value). The expected value (target value) y for judging the quality of the measurement data Y in this step, the upper limit value, and the lower limit value are calculated. Therefore, the expected value y of the measurement data Y in this step is obtained.
And the like, adaptively change according to the measurement data X in the previous process.

A process control chart of the present process is created based on the expected value y and the like calculated in this way and the measurement data X in the present process, and is displayed on a display. The display (X) of the previous process displays the actual measurement value (each triangle and its connection line) X, the target value (the middle solid line) x, the upper limit value and the lower limit value (upper and lower broken lines) of the measurement data in the previous process. . The display (Y) of the present process indicates the actual measurement value (each triangle and its connection line) Y, the target value (each middle solid line) y, the upper limit value and the lower limit value (each upper and lower broken line) of the measurement data in this process. indicate. The display of the post-process (Z) displays the actual measurement value (each triangle and its connection line) Z, the target value (the middle solid line) z, the upper limit value and the lower limit value (upper and lower broken lines) of the measurement data in the post-process. .

In the pre-process and post-process in which the same process control as that of the related art is performed, target values (expected values) x and z, an upper limit value, and a display (X) of the pre-process and a display (Z) of the post-process. The lower limit is constant without depending on the object to be measured (displayed in a straight line). In the present step of the present invention, the target value (expected value) y, the upper limit value, and the lower limit value of the step (Y) change depending on each measurement object (for example, each liquid crystal panel).
The operator can execute appropriate process management by looking at the display. Alternatively, based on the created process control chart of the present process, the process control device automatically determines the quality of the semi-finished product or the like as the measurement target. Thus, every time a product is manufactured in the previous process, it is possible to execute appropriate management according to the state of the previous process in the next process (this process).

For example, by substituting the target value for C1 and the measured data in the previous process of the semi-finished product to be managed for X,
An optimum expected value (in this case, also referred to as a target value) y of the measurement data Y in this step can be obtained.
For example, by substituting the upper limit value or lower limit value for C1 and the measurement data in the previous process of the semi-finished product to be managed for X, the upper limit value and the lower limit value of the measurement data Y in this process can be obtained (another value). Calculation may be performed.) For example, in a manufacturing process of a liquid crystal panel or the like that requires a small pixel size and high-precision process management, a manufacturing number is generally assigned to each liquid crystal panel sheet to manage the manufacturing process. For example, four liquid crystal panels can be obtained from one sheet. Therefore, the expected value y and the like can be calculated for each sheet based on the actual measurement data X in the previous process. Similarly, in a semiconductor manufacturing process in which fine processing is required and high-precision process control is required, a management number is generally assigned to each semiconductor wafer to control the manufacturing process. 1
For example, 100 semiconductor chips can be obtained from one wafer. Therefore, the expected value y and the like can be calculated for each individual wafer based on the actual measurement data X in the previous process.

In the manufacturing process of small-sized ICs and the like, which have relatively low processing accuracy and are inexpensive, it is permissible to manage the process under less strict control conditions than in the management of the manufacturing process of a liquid crystal panel, which requires high-precision process control. Is done. In a manufacturing process of an inexpensive small-sized IC or the like, a management number is not assigned to each individual wafer, and the manufacturing process is managed based on an average value of measurement data for each lot.
For example, one lot is composed of 25 wafers. The average value of the measurement data for each lot in the previous process is expressed by the relational expression C1 = f
The expected value y of the measurement data Y in this step is obtained by substituting X of (X, Y). Since the dispersion of individual ICs in the same lot is relatively small (or does not matter), the expected value y obtained in this way is used for determining the quality of the measurement data in the present process, thereby making it possible to determine whether the preceding process and the present process are good. Optimum process management utilizing the relationship can be performed. Similarly, for example, in a manufacturing process of a liquid crystal panel in which the number of pixels is small and the process control under a loose control condition is permitted, a control number is not assigned to each sheet, and the manufacturing is performed based on an average value of measured data for each lot. Manage the process. For example, one lot is composed of 20 sheets.

The same applies to the subsequent steps. First, in the post-process control procedure 12, the quality of the measurement data Z in the post-process (the second process described in claims) is determined. If the post-process is a process device, the measurement data of the process condition is used. If the post-process is an inspection device, the measurement data of the inspection result (the measurement data of the inspection result of the same product made before the object to be measured in this process) Is plotted on a post-process control chart and managed.

Next, in the post-process to main process definition procedure 22, the relational expression C2 = g (Z, Z,) between the measured data Z in the post-process and the measured data Y in the present process (first process in claims).
Y) (variable Y of relational expression C1 = f (X, Y))
And the variable Y of the relational expression C2 = g (Z, Y) are assumed to be separate measurement data. ). For example, C2 is a target value (constant value) of arbitrary measurement data determined based on the values of Z and Y. Next, in the post-process to main process calculation procedure 32, the measurement data Z in the post-process and the defined relational expression C2
= G (Z, Y), the expected value y for determining the quality of the measurement data Y in this step, the upper limit value, and the lower limit value are calculated. Therefore, the management reference value such as the expected value y of the measurement data Y in the present process changes adaptively according to the measurement data Z in the subsequent process.

Based on the expected value y and the like calculated in this way and the measured data Z in the present process, a process control chart of the present process is created and displayed on a display. The operator
By looking at the display, appropriate process management can be performed. Or, based on the created process control chart of this process,
The process management device automatically determines the quality of the semi-finished product or the like as the measurement target. Thus, every time a product is manufactured in a post-process, it is possible to execute appropriate management according to the state of the post-process in the previous process (this process).

The pre-process and the main process, and the main process and the post-process are not necessarily required to be continuous processes, and another process may be present between them. Further, there is a relational expression C3 = f (X, Y, Z) between the measurement data in the previous step, the measurement data in the post-step, and the measurement data in the present step. It is also possible to calculate an expected value or the like of the measurement data in the present process based on the measurement data in the subsequent process. The process control chart to be displayed may be not only the xR control chart of FIG. 1 but also another type of control chart.

FIG. 5 is a detailed explanatory diagram of calculation procedures 30 and 32 of the process management method for the process steps of the first embodiment. Calculation of a relational expression between the previous step and the present step will be described. First, the measurement data (actually measured value) x0 in the previous process of the object to be managed (a liquid crystal panel having a management number assigned to each panel) is read (step 80). Next, a numerical relational expression C1 = f defined by the preceding process to the present process definition procedure 20
(X, Y), average value of C1 value (substantially target value) c
1 and the standard deviation σc1 of the target of the value of C1 are read.
(Step 82). Finally, measurement data x0 (X = x
0), the relational expression C1 = f (X, Y), the average value c1 of the value of C1 and the standard deviation σc1 of the target value of C1. A target value) y and a target standard deviation σy are calculated (step 84). The illustrated Y = f ′ (C1, X) is obtained by C1 =
This is an equation obtained by transforming f (X, Y) into a form for obtaining Y.

A calculation example (schematic calculation example) at step 84 is shown at 86. The measured data (actually measured value) x0 in the previous process of the management target object is set to 3, the relational expression C1 = 2X + Y, the average value c1 of the value of C1 is set to 5, and the standard deviation σc1 of the target of the value of C1 is set to 2. The average value x of the measured values X in the previous process of the liquid crystal panel to be managed is x = 3 because the actual measured value is x0 = 3 with the parameter 1. Standard deviation σx because the parameter is 1
= 0. From Y = C1-2X (average value of Y) =
(Average value of C1) −2 · (Average value of X)
Substituting the average value c1 of C1 = 5 and the average value x of X = 3, the average value of Y becomes y = 5-2 × 3 = −1. In other words, if the liquid crystal panel whose measurement data in the previous step is x0 is to be processed in this step so that C1 of the liquid crystal panel is 5, the average value y of the measurement data in this step is obtained.
However, process management and the like in this step are executed so that y = 5-2 × 3 = −1.

Similarly, (standard deviation of Y) = (standard deviation of C1) −2 · (standard deviation of X). Substituting the standard deviation σc1 of C1 = 2 and the standard deviation σx = 0 of X, the standard deviation of Y) σy = 2-2 × 0 = 2. Process management and the like in this step are executed so that the standard deviation σy of Y becomes 2. The numerical relational expression is not limited to a simple expression as in this example, and may be a complicated linear expression or a non-linear expression. The process control chart displayed on the display is not limited to the xR control chart, and may be any other type of process control chart.
When different process control charts are displayed on the display, the calculation unit calculates data necessary for creating each process control chart.

The calculation of the relational expression between the post-process and the present process will be described. Measurement data (actual measurement value) z0 in the post process of the liquid crystal panel
Is read (step 90). Next, a numerical relational expression C2 = g defined by the main process to post-process definition procedure 22
Average value of (Z, Y) and C2 (substantially the target value) c
2 and the standard deviation σc2 of the target of the value of C2 are read.
(Step 92). Finally, the measurement data z0 (Z = z
0), the relational expression C2 = g (Z, Y), the average value c2 of the values of C2 and the standard deviation σc2 of the target value of C2, and the average value (data) necessary for the management of this process ( A target value) y and a target standard deviation σy are calculated (step 94). The illustrated Y = g ′ (C2, Z) is represented by C2 =
This is an equation obtained by transforming g (Z, Y) into a form for obtaining Y.

The invention of Embodiment 1 can be applied to, for example, a process control method in a liquid crystal manufacturing process. The illustrated liquid crystal manufacturing process of FIG. 7 includes four processes of (1) a bead feeding process, (2) a liquid crystal feeding process, (3) a top plate bonding process, and (4) a curing process. First, a conventional process control method will be described. (1) The bead loading step (previous step) is a step of placing beads 76 of a fixed volume (volume) X in a lower plate (shallow box with an open top) 75 of a liquid crystal plate. In the bead charging step, the capacity X of the beads 76 placed in the liquid crystal plate is controlled so as to fall within a target range (a certain range around the target capacity x).

(2) The liquid crystal charging step is a step of charging a liquid crystal 77 having a fixed capacity (volume) V into the lower plate 75 containing the beads 76. In the lower plate 75, the beads 76 and the liquid crystal 77 are mixed. Accordingly, the lower plate 75 contains beads 76 and liquid crystal 77 having a fixed capacity (X + V). (3) In the upper plate bonding step (this step), the upper plate 78 is bonded to the lower plate 75 while applying a constant pressure Y (target pressure y) to the beads 76 and the liquid crystal 77 in the lower plate 75. . Since the bottom area S of the lower plate 75 is constant, the liquid crystal plate constituting the outer shape of the lower plate 75 and the upper plate 78 is filled with beads 76 and liquid crystal 77 having a constant depth (height) C1. I have. When the unit of Y is atmospheric pressure, (X + V) = S × C
1 × Y holds.

(4) The curing step is a step of curing the adhesive for bonding the lower plate 75 and the upper plate 78. In the liquid crystal plate, the liquid crystal display surface (the upper plate 78 is
In addition, reducing the variation in the liquid crystal amount (V / S) per unit area of the liquid crystal 77 and the portion in contact with the liquid crystal 77) and the variation in the depth C1 of the liquid crystal are necessary for achieving stable quality and uniform performance. is important. Therefore, in the conventional process management method, management is performed such that all the values to be fixed in each process fall within a narrow allowable range.

On the other hand, in the liquid crystal manufacturing process to which the present invention is applied, the following process control method is used. The basic work content of each process is the same. (1) The bead loading step (previous step) is a step of placing beads 76 of a fixed volume X in the lower plate 75 of the liquid crystal plate.
In the bead charging step, the capacity X of the beads 76 placed in the liquid crystal plate is controlled so as to fall within a target range (a certain range around the target capacity x). In the process control method of the present invention, as described later, the volume X of the beads 76 is substantially reduced.
Is wider than conventional process control methods. (2) In the liquid crystal charging step, the lower plate 75 containing the beads 76
In this step, a liquid crystal 77 having a fixed capacity (volume) V is charged. In the lower plate 75, beads 7 of a fixed capacity (X + V) are provided.
6 and a liquid crystal 77.

(3) The upper plate bonding step (this step)
The upper plate 78 is attached to the lower plate 75 while applying a pressure Y (a target pressure y) adaptively changed according to the bead capacity X to the beads 76 and the liquid crystal 77 in the lower plate 75. (X +
V) = S × C1 × Y holds, so that the measured data of the bead capacity X = x0 and y = (X + V) / (S × C1)
(Y = (x0 + V) / (S × C1)). Since the bottom area S of the lower plate 75 is constant, the outer shape is constituted by the lower plate 75 and the upper plate 78 by applying the above target pressure y, even if the measured data X = x0 of the bead volume varies somewhat. The liquid crystal plate is filled with beads 76 and liquid crystal 77 having a certain depth (height) C1.

(4) The curing step is a step of curing the adhesive for bonding the lower plate 75 and the upper plate 78. According to the above process management method, it is possible to manufacture a liquid crystal plate having small variations in the liquid crystal amount (V / S) per unit area of the liquid crystal display surface and the liquid crystal depth C1.

In the conventional process control method, control is performed such that all the values to be fixed in each process fall within a narrow allowable range. Conventionally, the target value x of the volume X of beads in the preceding step and the target value y of the pressure Y in the present step were both fixed values. In the process management method of the present invention, the target value x of the liquid crystal capacity X in the previous process is a fixed value, but the target value y of the pressure Y in the present process is determined by the measured data (bead capacity measurement) in the previous process. It changes according to the value X). By performing the process control by adaptively changing the target value y of the pressure Y, the yield in this process can be increased, and the quality and performance of the product (liquid crystal plate) can be improved.

Embodiment 2 FIG. 2 is a block diagram of a process management apparatus for the process steps of Embodiment 2. A measuring unit is provided in each of a pre-process, a main process, and a post-process of the manufacturing process (not shown). Each measurement unit measures data in each step and transmits the measured data to a computer 4
Transmit to 0.

The process control device of the second embodiment is a computer
0. The computer 40 performs each step of the process management method of the first embodiment (the pre-process control process 10, the pre-process to the process definition process 20, the pre-process to the process calculation process 30, the post-process control process 12, the main process to the process process). The means for executing the post-process definition procedure 22, and the steps from the present process to the post-process calculation procedure 32) are incorporated as the program, and these procedures are executed by software processing. Generally, the computer 40 calculates a relational expression between the measurement data X in the previous step and the measurement data Y in the present step,
And measurement data Z in the subsequent process and measurement data Y in the present process
Is stored in the memory. Therefore, except for the case of inputting a new relational expression and changing the relational expression, the preceding process to the present process defining procedure 20 and the present process to the subsequent process defining procedure 2
There is no need to perform Step 2.

The computer 40 calculates the expected value y, the upper limit value, and the lower limit value of the measured data in this process, the measured data X in the preceding process, the measured data Z in the subsequent process, and the relational expressions C1 and C2.
Calculate based on 2. The expected value y, the upper limit value, and the lower limit value of the measurement data in this step are adaptively changed based on the measurement data X in the previous step and the measurement data Z in the subsequent step.

The display unit (display) 52 in the preceding process is
The actual measurement value (each triangle and its connection line) X, the target value (middle solid line) x, the upper limit value and the lower limit value (upper and lower broken lines) of the measurement data in the previous process are automatically displayed. The display unit (display) 50 of the present process includes an actual measurement value (each triangle and its connection line) Y, a target value (each middle solid line) y, an upper limit value and a lower limit value (each upper and lower broken line) of the measurement data in the present process. ) Is displayed automatically. The target value, the upper limit, and the lower limit adaptively change for each measurement target. Display unit (display) 5 for post-process
Reference numeral 4 automatically displays the actual measurement value (each triangle and its connection line) Z, the target value (middle solid line) z, the upper limit value and the lower limit value (upper and lower broken lines) of the measurement data in the post-process.

As a result, a process control chart having a variable average value y and upper and lower limit values can be created very easily and immediately without manual intervention. The display means 50 is not necessarily a screen (display), but may be a printing means such as a printer or a plotter.

FIG. 6 is an explanatory diagram of a process for displaying the average value y and the like on the display unit of the process management apparatus according to the second embodiment. The computer 40 further has a program for executing the display data calculation and display procedure 56. The computer 40 executes the program. Based on the average value x and the standard deviation σx of the measurement data in the previous step set or calculated by executing the previous step process management procedure 10, and the actual measurement data in the previous step, the computer 40 performs UCL (upper limit value) and LCL (lower limit value) of the measurement data are calculated (in the embodiment, the upper limit value and the lower limit value are constant values), and the measurement data X in the previous step and the measurement data X in the previous step are calculated. Bitmap image data for visually displaying the average value x, the upper limit value, and the lower limit value is generated. In FIG. 6, the upper limit value and the lower limit value are values 3σ (σ is the standard deviation) away from the average value. The display 52 visually displays the average value x, the upper limit value, and the lower limit value in the previous process.

Similarly, the average value y and the standard deviation σy of the measurement data in the present process, which are set or calculated by executing the pre-process to the main process calculation procedure 30 and the main process to the post-process calculation procedure 32, are calculated. Based on the measured data Y in the process, the computer 40 determines the UCL of the measured data Y in the process.
(Upper limit value) and LCL (lower limit value) are calculated, and bitmap image data for visually displaying the measurement data Y in this process, the average value y in this process, the upper limit value, and the lower limit value are generated. In FIG. 6, the upper limit value and the lower limit value are values 3σ (σ is the standard deviation) away from the average value. The display 50 illustrated in FIG. 6 visually displays the measurement data (triangular marks and their connection lines) Y in this process, the average value y, the upper limit (UCL), and the lower limit (LCL) in this process. ing.

Similarly, based on the average value z and the standard deviation σz of the measured data in the post-process set or calculated by executing the post-process control procedure 12, and the actually measured data Z in the post-process, The computer 40 calculates the UCL (upper limit value) and the LCL (lower limit value) of the measurement data in the post-process, and visually displays the measurement data Z in the post-process, the average value z, the upper limit, and the lower limit in the post-process. Generate bitmap image data to be displayed. In FIG. 6, the upper limit value and the lower limit value are values 3σ (σ is the standard deviation) away from the average value. The display 54 visually displays the measured data Z in the post-process, the average value z, the upper limit, and the lower limit in the post-process.

The computer 40 transmits the generated bitmap image data to each of the display units (displays) 50, 52, 5 through an internal communication line such as Ethernet (registered trademark).
4 Each display unit 50, 52, 54 displays the transmitted bitmap image data. In the case where it is desired to increase the strictness (severity) of the management reference value (quality judgment criterion) or when it is clear that the distribution is nonlinear, etc.
The formulas for CL and LCL are different. The invention of the second embodiment can be applied to, for example, a process management apparatus for a control process of a liquid crystal amount of a liquid crystal.

Third Embodiment FIG. 3 is a block diagram of a process management apparatus for a process according to a third embodiment. The process control device of the third embodiment is similar to the process control device of the second embodiment,
It is characterized in that a plurality of manufacturing processes (not shown, but arranged at a factory A, a factory B, and a factory C) are centrally managed. In the process management apparatus according to the third embodiment, the computer 40 is included in a central management unit 60 located at a remote place. The central management unit 60 is a server including a large-capacity hard disk device and the like.

Each of the factories A, B and C has a manufacturing process (the same manufacturing process or separate manufacturing processes. In the third embodiment, all the manufacturing processes are the same). Each manufacturing process has a pre-process, a main process, and a post-process. The pre-process, main process and post-process of each factory
Each has a measurement unit and a display unit (display).
Each measurement unit and each display unit of each factory and the central management unit 60 are mutually connected by, for example, the Internet 64. Instead of the Internet, an intranet or the like can be used.

Each measurement unit transmits the measurement data to the central management unit 60 via the Internet 64. Central Management Department 6
The computer 40 included in 0 calculates the expected value y, the upper limit value, and the lower limit value of the main process in each manufacturing process based on the received measurement data and the built-in relational expression. The central management unit 60 calculates the expected value y, the upper limit value, the lower limit value, and the measured data (actual measurement value) Y of the calculated main process in the main process.
Is transmitted to the display unit 50 of the main process of each manufacturing process via the Internet 64.

The display unit (display) 50 of the main process in each manufacturing process includes an actual measurement value (each triangle and its connection line) Y, a target value (each middle solid line) y, an upper limit value of the measurement data in the main process. The lower limit (the upper and lower broken lines) is automatically displayed. The target value, the upper limit, and the lower limit adaptively change for each measurement target.

For the pre-process and post-process in which the process is controlled with a fixed target value, upper limit value and lower limit value, the process control device of the third embodiment uses the same method as the measurement data of the present process to measure each process. The measurement data is transmitted from the unit to the central management unit 60, and the measurement data is transmitted from the central management unit 60 to the display units 52 and 54 of the respective processes. Instead, the measurement data may be directly transmitted from the measurement unit in each step to the display units 52 and 54. In this case, for example, the display unit 5
2 and 54 each have a built-in execution program for the pre-process control procedure 10 and the post-process control procedure 12,
2, 54 perform these procedures.

The display unit (display) 52 in the previous process is
The actual measurement value (each triangle and its connection line) X, the target value (middle solid line) x, the upper limit value and the lower limit value (upper and lower broken lines) of the measurement data in the previous process are automatically displayed. The display unit (display) 54 in the post-process includes the actual measurement values (each triangle and its connection line) Z of the measurement data in the post-process, and the target value (solid line in the middle).
z, the upper limit value and the lower limit value (upper and lower broken lines) are automatically displayed.

By connecting the centralized management means 60 to a plurality of factories (process steps) via a communication line (for example, the Internet), it is possible to ensure a constant quality in all factories. Further, in a new factory or the like that newly starts up, the quality can be improved at an early stage by utilizing the calculation unit 40 including a relational expression that has been used in another factory that has already started up.

Fourth Embodiment FIG. 4 is a block diagram of a process management apparatus for the process steps of a fourth embodiment. The process management apparatus of the fourth embodiment is similar to the process management apparatus of the third embodiment, and centrally manages a plurality of manufacturing processes (not shown, but arranged at a factory A, a factory B, and a factory C). are doing.

The process management apparatus of the third embodiment receives measurement data from each measurement unit in each manufacturing process, calculates expected values y and the like in this process based on the transmission data and a built-in relational expression. The expected value y of the present process and the like are transmitted to each display unit 50. In the process management apparatus according to the fourth embodiment, a computer 40 is arranged in each manufacturing process. Central Management Department 7
0 has a relational expression of the measurement data in each step. Each time the relational expression is changed, the distribution unit 72 of the centralized management unit 70
Is a computer 40 for each manufacturing process.
To be transmitted. The process management device in each manufacturing process is similar to the process management device of the first embodiment, except that the computer 40 receives the relational expression from the central management unit 70.

The central management unit 70 at a remote location is a server including a large-capacity hard disk device and the like. Factory A, B
The factory and the factory C each have a manufacturing process (either the same manufacturing process or separate manufacturing processes. In the fourth embodiment, all the manufacturing processes are the same). Each manufacturing process has a pre-process, a main process, and a post-process. The pre-process, main process, and post-process of each factory include a measuring unit, a computer 40, and display units (displays) 50, 52, and 54, respectively.
Having. The measuring units, the computer 40, and the display units 50, 52, and 54 of each factory are connected to each other by an internal communication line such as Ethernet. The computer 40 and the central management unit 70 are connected to each other via, for example, the Internet 64. Instead of the Internet, an intranet or the like can be used.

The central management unit 70 calculates a relational expression C1 between the measurement data X in the preceding step and the measurement data Y in the present step and a relational expression C2 between the measurement data Z in the subsequent step and the measurement data Y in the present step. Having. When it is necessary to send the relational expression to the computer in the manufacturing process of each factory, the distribution unit 72 of the centralized management unit 70 performs the same processing as when an application service provider (ASP) distributes application software. Then, the program of the relational expression is sent to each computer 40.

For example, when a new manufacturing process is created, the distribution unit 72 of the centralized management unit 70 transmits the relational expression via the Internet 64 to the computer 40 of the new manufacturing process. For example, when the relational expression is changed, the distribution unit 72 of the centralized management unit 70 transmits the changed relational expression to the computers 40 in the manufacturing processes of all factories via the Internet 64.

In each manufacturing process of each factory, each measuring unit transmits the measured data to the computer 4 through the internal communication line.
Transmit to 0. The computer 40 calculates the expected value, the upper limit value, and the lower limit value of each process based on the received measurement data and the relational expression distributed from the distribution unit 72 of the centralized management unit 70. The computer 40 displays the calculated expected value, upper limit value, lower limit value, and measured data (actually measured value) of each process in each process through the internal communication line.
2, 54.

Display unit (display) 50, 5 for each process
Reference numerals 2 and 54 automatically display actual measured values (each triangle and its connection line), target values (solid lines), upper limit values and lower limit values (upper and lower broken lines) of measurement data in each process. The target value, the upper limit value, and the lower limit value displayed by the display unit (display) 50 in this step are adaptively changed for each measurement target.

Since the centralized management unit 70 substantially centrally manages the relational expressions of the computers 40 of a plurality of factories (process steps), it is possible to secure a constant quality in all factories. Further, in a new factory or the like that newly starts up, by transmitting to the computer 40 a relational expression that has been used in another factory that has already started up, it is possible to achieve early quality improvement in the new factory. In the process management apparatus of the fourth embodiment, the central management unit 70 at a remote location and the computer 40 at each manufacturing process do not need to be always connected (the connection is made only when the program of the relational expression is transmitted). No need.) For example, after trying a new relation in one factory that plays a pilot role and confirming the reliability of the relation (or after sufficiently increasing the precision of the relation), all the other factories Can transmit the program of the relational expression of high accuracy to the computer 40 of the present invention.

[0070]

According to the present invention, there is provided a process control apparatus or a process control method for performing optimum process control when there is some relationship between each process or each measurement data of a plurality of processes in a manufacturing process. The advantageous effect that it can be realized is obtained. Particularly, in a manufacturing process having a plurality of processes such as a semiconductor diffusion process or a liquid crystal array process, when the process conditions and the like in this process are related to the process conditions and inspection results in a preceding process or a subsequent process, the optimal process is performed. An advantageous effect is obtained that a process management method and a process management device for a process process that performs the above process management can be realized.

According to the present invention, by providing means for automatically calculating and automatically displaying the average value and the upper and lower limit values of the measured data for the process management, the process management for performing the appropriate process management very easily and immediately. An advantageous effect that a method and a process control device can be realized is obtained.

According to the present invention, there is provided a process control apparatus which centrally controls a plurality of manufacturing processes, thereby ensuring a certain quality even in different places or different management departments, and enabling early quality improvement. The advantageous effect that it can be realized is obtained.

According to the present invention, a plurality of manufacturing processes are centrally managed, and a calculation unit and the like are arranged in each manufacturing process, so that a centralized management computer in a remote place and a computer in a place where a process exists are always present. Even if they are not connected, there is an advantageous effect that it is possible to realize a process management apparatus that can ensure a certain quality in different places or different management departments and can improve quality at an early stage.

According to the present invention, for example, a new relational expression is tried in one factory that plays a pilot role,
After confirming the reliability of the relational expression, an advantageous effect that a process management apparatus that transmits a program of the relational expression with high accuracy to computers in all other factories can be realized is obtained. When it is clear that a certain quality can be secured or an early quality improvement is possible by applying a specific relational expression, the program of the relational expression is immediately transmitted to the process control device of all manufacturing processes. I can do it.

[Brief description of the drawings]

FIG. 1 is a block diagram of a process management method of a process according to a first embodiment of the present invention.

FIG. 2 is a block diagram of a process management device for process steps according to a second embodiment of the present invention.

FIG. 3 is a block diagram of a process management device for process steps according to a third embodiment of the present invention.

FIG. 4 is a block diagram of a process management device for process steps according to a fourth embodiment of the present invention.

FIG. 5 is an explanatory diagram of a calculation procedure of a process management method of a process process according to the first embodiment of the present invention.

FIG. 6 is an explanatory diagram of a process of displaying an average value and the like on a display unit of the process management device according to the second embodiment.

FIG. 7: (1) Bead input step of the liquid crystal manufacturing process,
The figure which shows (2) a liquid crystal input process, (3) upper board bonding process, and (4) hardening process.

[Explanation of symbols]

 10 Pre-process control procedure 12 Post-process control procedure 20 Pre-process to main process definition procedure 22 Main process to post-process definition procedure 30 Pre-process to main process calculation procedure 32 Main process to post-process calculation procedure 40 Computer 50, 52, 54 display unit (display) 56 calculation and display procedure of display data 60, 70 centralized management unit 62 terminal 64 internet 72 distribution unit 75 lower plate of liquid crystal plate 76 beads 77 liquid crystal 78 upper plate of liquid crystal plate

Claims (6)

[Claims] 1. A first step in a manufacturing process including a plurality of steps.
A first measurement unit that measures data in the step of: a second measurement unit that measures data in a second step that is a pre-step or a post-step of the first step; and the first measurement unit It has a relational expression between the measured data and the data measured by the second measurement unit, and the data measured by the second measurement unit and the data measured by the first measurement unit based on the relational expression A calculating unit that outputs at least one of an expected value, an upper limit value, and a lower limit value; and a determining unit that determines whether the data measured by the first measuring unit is good or bad based on an output value of the calculating unit. And a process management device comprising: 2. A first step in a manufacturing process including a plurality of steps.
A first measurement unit that measures data in the step of: a second measurement unit that measures data in a second step that is a pre-step or a post-step of the first step; and the first measurement unit It has a relational expression between the measured data and the data measured by the second measurement unit, and the data measured by the second measurement unit and the data measured by the first measurement unit based on the relational expression A calculation unit that outputs at least one of an expected value, an upper limit value, and a lower limit value; and a display unit that displays an output value of the calculation unit and data measured by the first measurement unit. A process management device characterized by the above-mentioned. 3. The process management device has one centralized management unit including the calculation unit, and the plurality of manufacturing processes each include the first measurement unit and the second measurement unit. The said 1st measurement part and said 2nd measurement part of each of the said several manufacturing process, and the said central management part are connected by the communication line, The said, The Claim 1 or Claim 2 characterized by the above-mentioned. The process control device described in the above. 4. The process management apparatus has one centralized management unit having a program for driving the calculation unit, and the plurality of manufacturing processes are respectively performed by the calculation unit, the first measurement unit, and the first measurement unit. A plurality of measurement units, wherein each of the calculation units of the plurality of manufacturing processes is connected to the centralized management unit via a communication line, and the centralized management unit is connected to each of the calculation units via the communication line. The process management apparatus according to claim 1, wherein the program is transmitted to a computer. 5. A first step in a manufacturing process including a plurality of steps.
A first measurement step of measuring data in the step of; a second measurement step of measuring data in a second step that is a pre-step or a post-step of the first step; and the first measurement step. It has a relational expression between the measured data and the data measured in the second measurement step, and the data measured in the second measurement step and the data measured in the first measurement step based on the relational expression A calculating step of outputting at least one of an expected value, an upper limit value and a lower limit value; and the first step based on an output value in the calculating step.
A determining step of determining whether the data measured in the measuring step is good or bad. 6. A first step in a manufacturing process including a plurality of steps.
A first measurement step of measuring data in the step of; a second measurement step of measuring data in a second step that is a pre-step or a post-step of the first step; and the first measurement step. It has a relational expression between the measured data and the data measured in the second measurement step, and the data measured in the second measurement step and the data measured in the first measurement step based on the relational expression A calculation step of outputting at least one of an expected value, an upper limit value, and a lower limit value; and a display step of displaying an output value in the calculation step and data measured in the first measurement step. A process management method characterized by the above-mentioned.