JP2010191744A - Management method and management system for processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide the management method and management system of a processor for precisely detecting the deterioration of a through-put, and for easily managing the processor. <P>SOLUTION: The management method of an in-line processor for performing the application processing, exposure processing and development processing of photo-resist to a substrate on which an electronic device is formed (or has been formed) includes: a process for actually measuring a required time from a preliminarily set start point via two or more processing to an end point, among the processes of the in-line processing; a process for comparing data related with the required time obtained by the measurement with a management value (UCL, LCL) to be set based on the past result of the data; and a process for determining whether or not an in-line processor is normal based on the result of comparison. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a processing apparatus management method and a management system thereof.

  Conventionally, statistical process control (SPC) has been used as a quality control method for products produced in a manufacturing process. Statistical process control is, for example, an Xbar-R control chart that is a control chart in which an average value (Xbar) of a group of quality characteristic values and a range (R) of variation within the group are plotted, or an average value ( Xbar-s is a technique for managing quality variations in the manufacturing process using an Xbar-s control chart, which is a combination of control charts of group standard deviation (s).

  In such a method, a control chart is created by acquiring quality characteristic values using products produced in the manufacturing process or test samples used in place of the products, and plotting the acquired values on the figure. . From this control chart, it is possible to grasp whether the manufacturing process itself is in a stable state or an unstable state in which the quality of the product can be maintained at a certain level (for example, (See Patent Document 1).

JP 2008-27184 A

By the way, according to the conventional management method of the manufacturing process, various quality characteristic values are measured using a product or the like, and a control chart is created based on the measured values. Based on this control chart, the state of each device (equipment) in the manufacturing process has been grasped, and the device has been maintained as necessary. According to such a method, variations in quality characteristic values can be suppressed, and product quality and yield can be maintained high.
However, the above method cannot detect a decrease in throughput due to an automatic retry (re-try) of the apparatus or inadequate product processing conditions. Here, the throughput is a production amount per unit time.

  For example, as shown in FIG. 8, the case where product processing is performed by arbitrarily selecting conditions A to C in each device (No. 1, No. 2, etc.) in an arbitrary manufacturing process is illustrated. As shown in FIG. 8, when the condition B is processed in the second machine, unlike the normal operation, if the wafer position detection is repeated many times or the end point detection is abnormally prolonged, the product for the wafer Processing and unloading are delayed, and the throughput of Unit 2 is reduced.

Such a decrease in throughput does not affect the quality or yield of the product, and cannot be detected in the control chart in which the quality characteristic values of the product or the like are plotted. In addition, when the time required for detecting the position of the wafer or the time required for detecting the end point does not reach the limit value set in the sequence on the apparatus side, the apparatus does not recognize it as an error. For this reason, the apparatus side has not been able to accurately detect a decrease in throughput.
Accordingly, some aspects of the present invention have been made paying attention to such a problem, and are capable of accurately detecting a decrease in throughput and easily managing the processing apparatus. The purpose is to provide a management method and a management system therefor.

  In order to achieve the above object, a management method for a processing apparatus according to one aspect of the present invention is a management of a processing apparatus that performs predetermined product processing on a substrate on which an electronic device is formed (or formed). A method of measuring a required time from a preset start point to a preset end point through a plurality of processes in the product processing step, and obtained by the measurement. Comparing the data relating to the required time with a management value set based on the past performance of the data, and determining whether the processing apparatus is normal based on the result of the comparison. It is characterized by this.

  Here, the “electronic device” is, for example, a semiconductor device, a liquid crystal display device, or an electrophoretic display (EPD). The “substrate” is, for example, a semiconductor substrate on which a semiconductor device is formed (for example, a silicon substrate) or an SOI (Silicon on Insulator) substrate, a glass substrate on which a liquid crystal display device is formed, or an EPD. It is a thin film substrate. Further, the “substrate” may be, for example, a printed circuit board on which a dicing semiconductor device (IC chip) is mounted. The printed circuit board includes a rigid substrate (that is, an insulating substrate having low flexibility), a flexible substrate (that is, a thin and flexible insulating substrate) used in the TAB technology, and the like.

  According to such a management method, it is possible to roughly manage the state of the processing apparatus only by actually measuring the time required for product processing without using any special means. Further, it is possible to accurately detect a decrease in product processing throughput. As a result, for example, even if a processing device abnormality that cannot be detected from the quality characteristic value of a product or test sample or a change in the state of the processing device leading to the abnormality occurs, these are detected through a decrease in throughput. can do. Management of the processing device is simple.

The management method may further include a step of setting an upper limit management value and a lower limit management value as the management value. Here, the upper limit management value and the lower limit management value are set based on, for example, an average value of past results related to the data and a degree of variation. The degree of variation is expressed by, for example, standard deviation σ. The upper limit management value (UCL) is expressed by, for example, equation (1), and the lower limit management value (LCL) is expressed by, for example, equation (2).
UCL = average value X + n * σ (1)
LCL = average value X−n * σ (2)
In the formulas (1) and (2), n is an arbitrary positive numerical value, and n = 3 as an example.
According to such a management method, the range of the management value can be defined by the upper limit management value and the lower limit management value. Then, it is possible to check whether or not the data obtained by actual measurement is within the range of the management value.

In the management method, in the step of determining whether or not the processing device is normal, the processing is performed when the data is within a management value range defined by the upper limit management value and the lower limit management value. The apparatus may be determined to be normal, and if the data is outside the management value range, the processing apparatus may be determined to be abnormal. Here, “within the range of the management value” means, for example, that it is in the range of the lower limit management value or more and the upper limit management value or less. According to such a management method, whether or not the processing apparatus is normal can be objectively determined based on statistical data.
The management method may further include a step of adding the data to the past performance and updating the management value range when the data is within the management value range. good. According to such a management method, the range of the management value can be updated, and the accuracy of determining whether the processing apparatus is normal or abnormal can be improved.

In the above management method, the start point is set when the loading of the substrate into the processing apparatus is started, and the end point is set when the unloading of the substrate from the processing apparatus ends. It may be characterized by. According to such a management method, product processing can be collectively monitored from almost the beginning to the end of the product processing, and the rough state of the processing apparatus can be grasped in a macro manner.
In the management method, when there are a plurality of product processing conditions, the management value may be set for each of the product processing conditions. With such a management method, the throughput of product processing can be grasped for each condition.

Further, in the above management method, when there are a plurality of the processing devices, the management value may be set for each processing device. With such a management method, the throughput of product processing can be grasped for each apparatus.
A processing apparatus management system according to another aspect of the present invention is a system for managing a processing apparatus that performs predetermined product processing on a substrate on which an electronic device is formed (or formed), Of the product processing steps, measuring means for measuring a required time from a preset start point to a preset end point through a plurality of processes, and data relating to the required time obtained by the measurement Comparing means with a management value set based on the past performance of the data, and determining means for determining whether or not the processing device is normal based on the result of the comparison It is what.

  According to such a configuration, it is possible to roughly manage the state of the processing apparatus only by actually measuring the time required for product processing without using any special means. Further, it is possible to accurately detect a decrease in product processing throughput. As a result, for example, even if a processing device abnormality that cannot be detected from the quality characteristic value of a product or test sample or a change in the state of the processing device leading to the abnormality occurs, these are detected through a decrease in throughput. can do. Management of the processing device is simple.

The figure which shows the management method of the processing apparatus which concerns on 1st Embodiment. An example of the control chart which concerns on 1st Embodiment. The figure which shows an example of the fluctuation | variation factor of required time. The figure which shows the structural example of the management system 10 which concerns on 1st Embodiment. The flowchart which shows an example of the management method which concerns on 1st Embodiment. An example of the control chart which concerns on other embodiment. The figure which shows the example of a setting of the starting point and end point which concern on other embodiment. The figure which shows a prior art example and its problem.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that, in each drawing described below, parts having the same configuration are denoted by the same reference numerals, and redundant description thereof is omitted.
(1) First Embodiment FIGS. 1A and 1B are conceptual diagrams showing a management method for a processing apparatus according to a first embodiment of the present invention. In the first embodiment, an inline processing apparatus used in a photolithography process will be described as an example. As shown in FIG. 1A, the inline processing apparatus is a single-wafer processing apparatus that performs, for example, a photoresist coating process, an exposure process, and a development process on a wafer surface in succession.

In the present invention, it is essential to measure the time required from a preset start point to a preset end point through a plurality of processes in the product processing steps. Therefore, in the first embodiment, the start point is set when starting to carry the wafer into the inline processing apparatus. In addition, the end point is set when the unloading of the wafer from the inline processing apparatus ends. Then, the time required from the start point to the end point through the photoresist coating process, the exposure process, and the development process is actually measured.
Here, there are a plurality of methods for measuring the required time. For example, as shown in FIG. 1A, the first measurement method is a method of measuring the time required from the start point to the end point for each wafer. Thereby, the required time for every wafer can be calculated | required.

  In addition, as shown in FIG. 1B, for example, the second measurement method includes wafer # 1 that is first subjected to product processing in a lot, and wafer #n (n is a product processing that is finally subjected to product processing). When the first wafer # 1 into the in-line processing apparatus is started, the starting point is the time when the last wafer #n is unloaded from the in-line processing apparatus. Set to end point. This is a method of actually measuring the time required from the start point to the end point. Thereby, the required time for every lot can be calculated | required. In the present invention, either the first measurement method or the second measurement method may be used.

  FIG. 2 is an example of a management chart according to the first embodiment of the present invention, and is a chart plotting the time required for each lot of inline processing. The horizontal axis in FIG. 2 indicates, for example, the number (No.) of the processed lot, and the vertical axis indicates the required time for each lot. UCL is an upper limit management value (line), and LCL is a lower limit management value (line). These UCL and LCL are values respectively set based on the past results for the time required for inline processing, and are, for example, values calculated by the above formula (1) or (2). UCL and LCL in FIG. 2 are values when n = 3 (that is, ± 3σ) in the formulas (1) and (2), for example.

In such a control chart, it is determined that the inline processing apparatus is normal while the data obtained by measurement is trending within the range of the management value sandwiched between UCL and LCL. However, for example, lot no. If the data is out of the management value range as in 61, the inline processing device is determined to be abnormal.
By the way, there are a plurality of reasons why the time required for inline processing varies, such as deterioration or misalignment of parts constituting the inline processing apparatus, and defects in processing conditions.

FIGS. 3A and 3B are diagrams showing an example of the variation factor of the required time. FIG. 3A is a diagram in which UCL and LCL are omitted from the control chart of FIG. 2 and a broken line is added to make it easier to understand the fluctuation amount of the required time. FIG. 3B is a diagram showing a trend different from that in FIG. 2, but the vertical and horizontal axes are the same as those in FIG. 2. That is, the horizontal axis of FIG. 3B indicates the lot number (No.), and the vertical axis indicates the required time for each lot.
For example, as shown in FIG. 3A, the time required for inline processing may be short-term fluctuation, long-term fluctuation, fluctuation due to poor adjustment, and the like. There are many.

  Here, the short-term fluctuation is due to, for example, deterioration of a mercury lamp used as a light source in the exposure process. Since the light emission intensity of the mercury lamp decreases with the accumulated lighting time, the light emission intensity is restored by performing maintenance such as adjustment and replacement at a certain frequency. Long-term fluctuations are caused by, for example, deterioration of the illumination system used in the exposure process. The influence of the deterioration of the illumination system is naturally small as compared with the deterioration of the light source, but the intensity of the illumination system also has a slight influence on the time required for in-line processing. Furthermore, the fluctuation due to poor adjustment is due to, for example, poor adjustment of the position of the mercury lamp. If the position of the mercury lamp is poorly adjusted, the light emission intensity and illuminance drop, and the required time may not be recovered to the expected level.

Further, as shown in FIG. 3B, the time required for the in-line processing may suddenly vary due to, for example, an increase in the automatic retry of the transport system (that is, transport retry). The conveyance retry is automatically performed according to a sequence when the wafer position is not correctly detected, for example.
With respect to the time required for inline processing that varies in this way, in the present invention, as shown in FIG. 2, for example, an upper limit management value and a lower limit management value are set based on past results, and based on these management values. Statistically manage the time required. For example, the time required from the start of the carry-in process to the inline processing apparatus to the end of the carry-out process is monitored for each lot. Thereby, the throughput of inline processing can be grasped for each lot. Next, a management system for a processing apparatus according to the first embodiment of the present invention and a method for managing an inline processing apparatus using this management system will be described with reference to FIGS.

  FIG. 4 is a block diagram illustrating a configuration example of the management system 10 according to the first embodiment of the present invention. As shown in FIG. 4, the management system 10 includes a first host computer 1, a second host computer 2, a first client 3, and a second client 4. Here, the host computers (hereinafter also simply referred to as “hosts”) 1 and 2 are a central processing unit (CPU) responsible for various types of control and arithmetic processing, a random access memory (RAM) and a read (ROM), respectively. Or a storage device such as a hard disk. The hosts 1 and 2 are connected to each other via a signal line such as a bus so that signals can be transmitted and received. The first client 3 is, for example, an inline processing device, and the second client 4 is, for example, a desktop or notebook computer device or a portable terminal device. The host 2 and the client 3 are connected so that signals can be transmitted and received with each other via a signal line such as a bus. Further, the host 2 and the client 4 are connected so that signals can be transmitted and received with each other via a signal line such as a bus or wirelessly, for example.

  FIG. 5 is a flowchart showing an example of the management method according to the first embodiment of the present invention. In step S (1) of FIG. 5, for example, a management chart in which an upper limit management value (UCL) and a lower limit management value (LCL) are set as shown in FIG. 2 is prepared. This control chart is stored in the storage device of the host 1, for example. Next, in step S (2) of FIG. 5, product processing by the client 3 is performed. As described above, the client 3 is, for example, an inline processing device. When the client 3 starts processing, the product processing conditions are transmitted from the host 2 to the client 3 (see arrow A in FIG. 4). Based on the transmitted conditions, the client 3 starts product processing and starts measuring the required time. After the measurement of the required time is completed, the process proceeds to step (S) 3 in FIG.

Next, in step S (3), the product processing conditions executed in step (S) 2 and the data relating to the required time obtained by measurement are transmitted from the client 3 to the host 2 (FIG. 4). (See arrow B). These data are transmitted from the host 2 to the host 1 (see arrow C in FIG. 4).
Next, in step S (4) of FIG. 5, it is determined whether or not the data obtained by the measurement is within the range between the UCL and the LCL set in the control chart (that is, the management value range). To do. Here, for example, the host 1 confirms the product processing conditions sent from the host 2 and the data obtained by the measurement. Then, the host 1 reads the control chart corresponding to the product processing condition from the storage device, compares the data obtained by the measurement with the control chart, and determines whether the data is within the range of the management value. To do. Then, this determination result is transmitted to the host 2 (see arrow D in FIG. 4).

  When the determination result by the host 1 is abnormal (that is, when it is determined that the data obtained by the measurement is out of the management value range), the process proceeds to step S (5) in FIG. Here, a determination result indicating that there is an abnormality is transmitted from the host 2 to the client 4 (see arrow E in FIG. 4). Upon receiving this signal, the client 4 generates an alarm, for example, and notifies the worker or the like. The worker who has received this notification stops, for example, the product processing for the next lot or later by the client 3, and searches each part of the client 3 for the cause of the increase in the required time (that is, the decrease in throughput). Can be dealt with.

On the other hand, in step (S) 4 in FIG. 5, if the determination result is normal (that is, if it is determined that the data obtained by measurement is within the management value range), go to step (S) 1. Return. In step (S) 1, the data obtained in step (S) 3 is plotted on the control chart, and the plotted data is added to the actual result to update UCL and LCL.
As described above, according to the first embodiment of the present invention, it is possible to collectively monitor from the start of loading a lot to the client (for example, inline processing apparatus) 3 until the end of unloading of the lot. Only by actually measuring the time required for product processing, the state of the client 3 can be roughly grasped without using any special means. Then, the throughput of product processing by the client 3 can be grasped, and a decrease in throughput can be detected easily and accurately. As a result, even if an abnormality of the client 3 that cannot be detected from the quality characteristic value of the product or the test sample or a change in the state of the client 3 that is a precursor of the abnormality occurs, these are detected through a decrease in throughput. Can be detected. Management of the client 3 is simple.

In the first embodiment, the host 1 corresponds to “measurement means” and “comparison means” of the management system of the present invention, and the client 3 corresponds to “measurement means”.
In FIG. 4, only one client 3 is shown, but the number of clients 3 is not limited to one, and a plurality of clients 3 may be provided. In FIG. 4, the hosts 1 and 2 and the clients 3 and 4 are described as separate configurations, but the form of the management system 10 is not limited to this. For example, the hosts 1 and 2 may be integrated into one host computer. Further, the clients 3 and 4 may be integrated to constitute one processing apparatus. Alternatively, the hosts 1 and 2 and the clients 3 and 4 may be integrated to constitute one processing apparatus.

  When there are a plurality of clients 3, management values such as UCL and LCL may be set for each client. Further, when there are a plurality of product processing conditions, management values such as UCL and LCL may be set for each product processing condition. Thereby, the throughput of product processing can be grasped for each client and for each product processing condition.

(2) Other Embodiments In the first embodiment described above, FIG. 2 is exemplified as a management chart for statistically managing the processing time. However, the control chart of the present invention is not limited to the trend chart as shown in FIG. A frequency distribution table may be used.
FIG. 6 is an example of a management chart according to another embodiment of the present invention, in which the processing time is represented not by a trend but by a frequency distribution. The horizontal axis of FIG. 6 shows, for example, the time required from carrying in a lot to carrying it out. The vertical axis indicates the number of product processes (that is, the number of processed lots). As shown in FIG. 6, even when the frequency distribution table is used, it is possible to grasp occurrences such as a halfway end (product processing) due to a trouble, an increase in conveyance retry, a temporary stop of the apparatus due to an alarm, and the like. Also in the frequency distribution table, management values such as UCL and LCL can be set based on past results, and the required time can be statistically managed based on these management values.

  In the first embodiment described above, for example, as shown in FIGS. 1A and 1B, the start of loading of the wafer into the inline processing apparatus is set as a starting point, and the inline processing apparatus The case where the end point of unloading of the wafer is set as the end point has been described. However, in the present invention, the setting of the start point and the end point is not limited to this. For example, as shown in FIGS. 7A and 7B, in an inline processing apparatus, the time when the coating process is started may be set as the start point, and the time when the development process is ended may be set as the end point. Even with such a method, it is possible to monitor from the start of the coating process to the end of the development process in a batch, and it is possible to grasp the throughput of the product processing excluding the wafer carry-in and carry-out processes. it can.

  In the first embodiment, the “product processing” of the present invention has been described by taking as an example the case of in-line processing in which photoresist coating, exposure, and development processing are continuously performed. However, the product processing of the present invention is not limited to this, and may be ion implantation processing for introducing impurities into the substrate, etching processing, film formation processing such as CVD, thermal diffusion / thermal oxidation processing, and the like. .

Here, when the product treatment is, for example, thermal diffusion / thermal oxidation treatment, as “multiple treatments” of the present invention, the substrate loading process to the heat treatment furnace, the temperature raising process, and the thermal diffusion / thermal oxidation process at a constant temperature. Examples of the temperature lowering process and the process of carrying out the substrate from the heat treatment furnace can be given. For example, the start point is set to the time when the loading of the lot into the heat treatment furnace is started, and the end point is set to the time when the removal of the lot from the heat treatment furnace is completed. In the case where the heat treatment furnace is a batch type apparatus, in FIG. 1A, the wafer unit may be read as the lot unit, and the start point and the end point may be set. Thereby, the same effect as the case of the inline processing apparatus explained in the first embodiment can be obtained. That is, it is possible to roughly grasp the state of the heat treatment furnace by actually measuring the time required for the heat treatment without using any special means. Then, the throughput of the heat treatment can be grasped, and a decrease in the throughput can be detected easily and accurately.
Furthermore, the “product processing” of the present invention is not limited to chemical processing, physical processing, and thermal processing on a substrate. The “product processing” of the present invention may be an inspection process for a substrate. The inspection process includes, for example, an electrical inspection using a prober or a tester, an appearance inspection using an imaging device, and the like.

  1, 2 host computers, 3 clients (processing devices), 4 clients, 10 management systems

Claims (8)

A processing apparatus management method for performing predetermined product processing on a substrate on which an electronic device is formed (or formed),
Of the product processing steps, a step of measuring a time required from a preset start point to a preset end point through a plurality of processes;
Comparing the data relating to the required time obtained by the measurement with a management value set based on the past performance of the data;
And a step of determining whether or not the processing apparatus is normal based on a result of the comparison.   The processing apparatus management method according to claim 1, further comprising a step of setting an upper limit management value and a lower limit management value as the management value. In the step of determining whether or not the processing apparatus is normal,
If the data is within the range of management values defined by the upper limit management value and the lower limit management value, determine that the processing device is normal,
3. The processing apparatus management method according to claim 2, wherein the processing apparatus is determined to be abnormal when the data is outside the range of the management value.   The processing apparatus according to claim 3, further comprising a step of, when the data is within the range of the management value, adding the data to the past performance and updating the range of the management value. Management method.   2. The start point is set when the loading of the substrate into the processing apparatus is started, and the end point is set when the unloading of the substrate from the processing apparatus is completed. The management method of the processing apparatus as described in any one of Claims 1-4. If there are multiple product processing conditions,
6. The processing apparatus management method according to claim 1, wherein the management value is set for each product processing condition. When there are a plurality of the processing devices,
7. The processing apparatus management method according to claim 1, wherein the management value is set for each processing apparatus. A system for managing a processing apparatus that performs predetermined product processing on a substrate on which an electronic device is formed (or formed),
Of the product processing steps, measuring means for measuring a required time from a preset start point to a preset end point through a plurality of processes,
Comparison means for comparing the data relating to the required time obtained by the measurement with a management value set based on the past performance of the data;
And a determination unit that determines whether or not the processing device is normal based on a result of the comparison.

Images