JP2007311581A - Process control method, and process control system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a process control method capable of specifying cause of yield drop at high probability. <P>SOLUTION: The process control method includes a step where a process history information is collected for each manufacturing device 6, a step where a process parameter when processing a wafer is calculated based on the process history information collected for each manufacturing device 6, a step for collecting the inspection information for calculating yield, and a step for identifying the process parameter that contributes much to yield by analyzing with the process parameter and inspection information. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a process management method and a process management system, and more particularly to a process management method and a process management system for managing a manufacturing process of a product manufactured by being processed in a plurality of manufacturing processes.

  2. Description of the Related Art Conventionally, there is known a process management method capable of discriminating a manufacturing process or a manufacturing apparatus that causes a decrease in yield (for example, see Patent Document 1). In the above-mentioned Patent Document 1, the device-specific failure occurrence information obtained experimentally in advance is compared with the yield information obtained from the inspection result of the actually processed product, which causes a decrease in yield. A method for discriminating one specific manufacturing process or manufacturing apparatus is disclosed.

JP 2000-353729 A

  However, since the above-described Patent Document 1 is configured to determine only one specific manufacturing process (manufacturing apparatus) that causes a decrease in yield, there are the following inconveniences. That is, for example, in a manufacturing process of an LSI or the like processed in hundreds of manufacturing processes, the cause of the yield reduction may straddle a plurality of manufacturing processes (manufacturing apparatuses). In this case, with the configuration of Patent Document 1, it is difficult to specify the cause of the yield reduction. In addition, the cause of the yield fall over a several manufacturing process (manufacturing apparatus) is the length of time after the process in a predetermined manufacturing process is complete | finished until the process in another manufacturing process is started, for example And so on. Further, in Patent Document 1, even if a manufacturing process (manufacturing apparatus) that causes a decrease in yield can be determined, a more detailed cause of a decrease in yield (for example, processing time, processing conditions, etc.) There is also the inconvenience that it is difficult to specify. As a result, the above-mentioned Patent Document 1 has a problem that it is difficult to specify the cause of yield reduction with high accuracy.

  The present invention has been made to solve the above-mentioned problems, and one object of the present invention is to provide a process management method capable of specifying the cause of yield reduction with high accuracy. is there.

  Another object of the present invention is to provide a process management system capable of specifying the cause of yield reduction with high accuracy.

  In order to achieve the above object, a process management method according to a first aspect of the present invention is a process for managing a manufacturing process of a product manufactured by being processed in a plurality of manufacturing processes each including at least one manufacturing apparatus. A management method for collecting processing history information for each manufacturing device, calculating a process parameter for processing a product based on the processing history information collected for each manufacturing device, and calculating a yield Collecting the inspection information to perform, and determining the process parameter having a large contribution to the yield by performing analysis using the process parameter and the inspection information.

  In the process management method according to the first aspect, as described above, the processing history information is collected for each manufacturing apparatus, and the process for processing the product based on the processing history information collected for each manufacturing apparatus. By calculating the parameters, for example, detailed process parameters such as the following (1) to (5) can be easily obtained.

  (1) Time from the end of processing of a product of interest in a predetermined manufacturing process to the start of processing of the product of interest in another manufacturing process.

  (2) In a predetermined manufacturing apparatus, the time from the end of processing of another product processed before the target product is processed to the start of processing of the target product.

  (3) In a predetermined manufacturing apparatus, the time from the start of processing of another product processed before the target product is processed to the start of processing of the target product.

  (4) When each of two or more of the plurality of manufacturing processes includes the same manufacturing apparatus, the product of interest is processed by the same manufacturing apparatus over the plurality of manufacturing processes. Number of times.

  (5) The number of times other products are processed under the same processing conditions as the product of interest before the product of interest is processed in a predetermined manufacturing apparatus.

  As a result, in the first aspect, by performing analysis using the process parameters and inspection information (yield) as described in (1) to (5) above, process parameters having a large contribution to the yield are determined. Thus, the cause of the yield reduction can be specified with high accuracy.

  In the process management method according to the first aspect, in the process parameter calculation step, the processing of the product of interest in another manufacturing process is started after the processing of the product of interest in the predetermined manufacturing process is completed. The method may include a step of calculating a process parameter including a time until the time. With this configuration, it is possible to examine the influence of the product leaving time on the yield.

  In the process management method according to the first aspect, the step of calculating the process parameter is focused on after processing of another product processed before the focused product is processed in a predetermined manufacturing apparatus. A step of calculating a process parameter including a time until processing of a product is started may be included. If comprised in this way, the influence which the idle time (idling time) of a manufacturing apparatus has on a yield can be investigated.

  In the process management method according to the first aspect, the step of calculating the process parameter is focused on after processing of another product that is processed before the focused product is processed in a predetermined manufacturing apparatus. A step of calculating a process parameter including a time until processing of a product is started may be included. If comprised in this way, when several products from which an identification number mutually differs are processed in parallel with the same manufacturing apparatus, the influence which the process start interval in the manufacturing apparatus has on a yield can be investigated.

  In the process management method according to the first aspect, the step of calculating a process parameter includes a plurality of manufacturing processes when each of two or more manufacturing processes among the plurality of manufacturing processes includes the same manufacturing apparatus. Furthermore, a step of calculating a process parameter including the number of times that the product of interest is processed by the same manufacturing apparatus may be included. If comprised in this way, the influence which the frequency | count of processing (processing frequency) with the same manufacturing apparatus has on a yield can be investigated.

  In the process management method according to the first aspect, the step of calculating the process parameter is performed in a predetermined manufacturing apparatus under the same processing condition as the product of interest before the product of interest is processed. A step of calculating a process parameter including the number of times to perform If comprised in this way, the influence which the frequency | count of a process on the same process conditions has on a yield can be investigated.

  In the process management method according to the first aspect, preferably, after the step of determining a process parameter having a large contribution to the yield, the manufacturing process or the manufacturing apparatus corresponding to the process parameter having a large contribution to the yield. The method further includes a step of feeding back the cause of the yield reduction. If comprised in this way, the fall of a yield can be improved easily.

  A process management system according to a second aspect of the present invention is a process management system for managing a manufacturing process of a product manufactured by being processed in a plurality of manufacturing processes each including at least one manufacturing apparatus. Processing history information collecting means for collecting processing history information every time, calculation means for calculating process parameters when processing products based on processing history information collected for each manufacturing apparatus, and for calculating yield Inspection information collecting means for collecting inspection information, and determination means for discriminating process parameters having a large contribution to the yield by performing analysis using the process parameters and the inspection information.

  In the process management system according to the second aspect, as described above, the processing history information collecting means for collecting the processing history information for each manufacturing apparatus and the processing history information collected for each manufacturing apparatus are processed. By providing the calculation means for calculating the process parameters when performing the above, detailed process parameters such as (1) to (5) described above can be easily obtained. As a result, in the second aspect, by analyzing using the process parameters and inspection information (yield) as described in (1) to (5) above, process parameters having a large contribution to the yield are determined. Thus, the cause of the yield reduction can be specified with high accuracy.

  The process management system according to the second aspect preferably further includes feedback means for feeding back the cause of the yield decrease to the manufacturing process or the manufacturing apparatus corresponding to the process parameter having a large contribution to the yield. If comprised in this way, the fall of a yield can be improved easily.

  As described above, according to the present invention, it is possible to obtain a process management method and a process management system capable of specifying the cause of yield reduction with high accuracy.

(First embodiment)
FIG. 1 is a diagram for explaining the overall configuration of a process management system according to a first embodiment of the present invention. FIG. 2 is a diagram for explaining details of the manufacturing process management apparatus included in the process management system according to the first embodiment shown in FIG. 1, and FIG. 3 is a process according to the first embodiment shown in FIG. It is a figure for demonstrating the detail of the yield analysis apparatus contained in a management system. First, with reference to FIGS. 1-3, the structure of the process management system by 1st Embodiment is demonstrated.

  As shown in FIG. 1, the process management system 1 according to the first embodiment includes a manufacturing process management apparatus 2, a test process management apparatus 3, and a yield analysis apparatus 4. Each device (manufacturing process management device 2, test process management device 3, and yield analysis device 4) constituting this process management system 1 is connected to each other via a data communication path 5. In addition, a plurality of manufacturing apparatuses 6 are connected to the manufacturing process management apparatus 2 via data communication paths, and a plurality of testers 7 are connected to the test process management apparatus 3 via data communication paths. Yes. Here, the plurality of manufacturing apparatuses 6 according to the first embodiment have a function of manufacturing a wafer (a “product” of the present invention) in which a plurality of chips are formed. A wafer in which a plurality of chips are fabricated is manufactured by being processed in a plurality of manufacturing processes each including at least one manufacturing apparatus 6. In addition, the plurality of testers 7 have a function of determining non-defective products and defective products by inspecting chips formed on the wafer.

  As shown in FIG. 2, the manufacturing process management apparatus 2 includes a communication path connection unit 11, a control unit 12, a process management program storage unit 13, a temporary storage unit 14, a read / write unit 15, and a display. Means 16 and database 17 are included. The communication path connection unit 11 has a function of connecting the manufacturing process management device 2 and the data communication path 5. The control unit 12 has a function of controlling the operation of the manufacturing process management apparatus 2 based on the process management program stored in the process management program storage unit 13.

  Further, in the first embodiment, the control means 12 of the manufacturing process management device 2 collects processing history information for each manufacturing device 6 (see FIG. 1), and the processing history information collected for each manufacturing device 6 is collected. Based on this, it has a function of calculating process parameters when processing a wafer. Here, the processing history information is, for example, a wafer identification number, a processing time, a processing condition, and the like. In addition, the process parameters of the first embodiment are the processing of a target wafer in another manufacturing process (manufacturing apparatus 6) after the processing of the target wafer in a predetermined manufacturing process (manufacturing apparatus 6) is completed. Is the time until is started. That is, the process parameters of the first embodiment are calculated based on the wafer identification number and the processing time among the plurality of processing history information collected for each manufacturing apparatus 6. The control means 12 is an example of the “processing history information collection means” and “calculation means” in the present invention.

  In addition, the temporary storage unit 14 of the manufacturing process management apparatus 2 has a function of temporarily storing the process parameters calculated by the control unit 12. The read / write unit 15 has a function of reading information on process management stored in the external storage medium 18 via the interface 19. Further, the reading / writing means 15 has a function of writing information relating to process management to the external storage medium 18 via the interface 19. The display means 16 has a function of displaying information related to process management. The database 17 includes a processing time database 17a, a first processing number database 17b, a second processing number database 17c, a maintenance information database 17d, and a limit value database 17e. The time as the process parameter calculated by the control means 12 is stored in the processing time database 17a. Further, a personal computer 20 is connected to the manufacturing process management apparatus 2 via a data communication path 5.

  As shown in FIG. 1, the test process management device 3 has a function of collecting inspection information of chips inspected by the tester 7 for each wafer. Here, in the first embodiment, the yield for each wafer is calculated from the inspection information. The inspection information (the yield for each wafer) is stored in the database 21 inside the test process management device 3. The test process management device 3 is an example of the “inspection information collection unit” in the present invention.

  Further, as shown in FIG. 3, the yield analysis apparatus 4 includes a communication path connection unit 31, an analysis unit 32, an information arrangement unit 33, a feedback unit 34, a temporary storage unit 35, a display unit 36, a database. 37. The communication path connection unit 31 has a function of connecting the yield analysis device 4 and the data communication path 5.

  Here, in the first embodiment, the analysis means 32 of the yield analysis apparatus 4 includes the process parameters stored in the manufacturing process management apparatus 2 (see FIG. 1) and the inspection information (wafer) stored in the test process management apparatus 3. Yield), and a multiple regression analysis is performed using the collected process parameters and inspection information (yield for each wafer). Specifically, the multiple regression analysis by the analysis unit 32 is performed using process parameters as explanatory variables and inspection information (yield for each wafer) as an objective variable. That is, in the first embodiment, after the processing of the wafer of interest in a predetermined manufacturing process (manufacturing apparatus 6) is completed, the processing of the wafer of interest in another manufacturing process (manufacturing apparatus 6) is started. The time until is an explanatory variable. Thereby, in the analysis means 32, the process parameter with a large contribution to a yield is specified. The analyzing means 32 is an example of the “discriminating means” in the present invention.

  The information organizing unit 33 of the yield analyzing apparatus 4 has a function of organizing the information (contribution to the yield) analyzed by the analyzing unit 32 based on the magnitude of the contribution to the yield. Further, the feedback means 34 has a function of feeding back the cause of the yield reduction to the manufacturing process (manufacturing apparatus 6) corresponding to the process parameter having a large contribution to the yield. The feedback by the feedback means 34 is performed via the manufacturing process management device 2 (see FIG. 1).

  Further, the temporary storage means 35 of the yield analysis device 4 has a function of temporarily storing the information organized by the information organizing means 33. The display means 36 has a function of displaying information related to yield. The database 37 includes an in-device database 37a and a limit value database 37b. The information organized by the information organizing means 33 is stored in the in-device database 37a. As shown in FIG. 1, the yield analysis device 4 is connected to an output device 38 for outputting information related to the yield.

  FIG. 4 is a flowchart for explaining the operation (process management method) of the process management system according to the first embodiment of the present invention. FIG. 5 is a diagram for explaining process parameters calculated by the process management system according to the first embodiment of the present invention. 6 to 9 are diagrams for explaining the multiple regression analysis performed in the process management system according to the first embodiment of the present invention. Next, the operation (process management method) of the process management system according to the first embodiment will be described with reference to FIGS.

  In the first embodiment, first, in step S1 of FIG. 4, processing history information is collected for each manufacturing apparatus 6 (see FIG. 1) by the control means 12 (see FIG. 2) of the manufacturing process management apparatus 2.

  Next, in step S <b> 2 of FIG. 4, process parameters are calculated based on the processing history information (wafer identification number and processing time) collected for each manufacturing apparatus 6. Specifically, in the first embodiment, after the processing of the wafer of interest in a predetermined manufacturing process (manufacturing apparatus 6) is completed, the processing of the wafer of interest in another manufacturing process (manufacturing apparatus 6) is performed. The time until the start (hereinafter referred to as the time between manufacturing processes) is calculated. Below, with reference to FIG. 5, the time between the manufacturing processes as a process parameter of 1st Embodiment is demonstrated. Note that identification numbers A to D are assigned to the wafer shown in FIG. In FIG. 5, the wafer is processed in the order of the manufacturing process M1, the manufacturing processes M2, M3,. Further, DB1 to DBn in FIG. 5 represent databases in which processing history information collected for each manufacturing apparatus 6 is stored.

  When attention is paid to the wafer A, the time between manufacturing steps as process parameters of the first embodiment includes T1 to T4 in FIG. Specifically, the times (process parameters) T1 and T2 between the manufacturing steps are the following after the processing of the wafer A (the wafer of interest) in the manufacturing steps M1 and M2 (predetermined manufacturing steps) is completed. This is the time until the processing of the wafer A (target wafer) in the manufacturing steps M2 and M3 (other manufacturing steps) is started. In addition, the time (process parameter) T3 between the manufacturing steps is equal to the third manufacturing step M3 (others) after the processing of the wafer A (target wafer) in the manufacturing step M1 (predetermined manufacturing step) is completed. This is the time until the processing of wafer A (the wafer of interest) in the manufacturing process) is started. The time (process parameter) T4 between the manufacturing processes is the fourth and subsequent manufacturing processes Mn (others) after the processing of the wafer A (target wafer) in the manufacturing process M1 (predetermined manufacturing process) is completed. This is the time until the processing of the wafer A (the wafer of interest) in (the manufacturing process) is started. That is, in the first embodiment, the other manufacturing process is not limited to the manufacturing process subsequent to the predetermined manufacturing process, and may be a manufacturing process after the predetermined manufacturing process.

  In addition, calculation of the time between manufacturing processes as an above-described process parameter is performed about all the wafers containing wafer AD. Moreover, the calculation of the time between manufacturing steps as the process parameters described above may be performed over all the manufacturing steps or only for a specific manufacturing step. Further, the time between manufacturing processes as the calculated process parameter is stored in the processing time database 17a (see FIG. 2) of the manufacturing process management apparatus 2.

  Next, in step S3 of FIG. 4, the test process management device 3 (see FIG. 1) collects the inspection information of the chips inspected by the tester 7 (see FIG. 1) for each wafer. Note that the yield for each wafer is calculated from the inspection information collected at this time.

  Next, in step S4 of FIG. 4, process parameters (time between manufacturing processes) are collected from the manufacturing process management apparatus 2 by the analysis means 32 (see FIG. 3) of the yield analysis apparatus 4, and inspection information (for each wafer) is collected. ) Is collected from the test process management device 3. Thereafter, the analysis means 32 of the yield analysis apparatus 4 associates process parameters (time between manufacturing processes) with inspection information (yield for each wafer). FIG. 6 shows a part of the result of associating process parameters (time between manufacturing steps) and inspection information (yield for each wafer). Note that the process parameters P1 to Pn in FIG. 6 are times between manufacturing steps calculated in step S3.

  Next, in step S5 of FIG. 4, the analysis means 32 of the yield analysis device 4 uses the inspection information (yield for each wafer) as an objective variable for all wafers including the wafers A to D, and process parameters (between manufacturing steps). The partial correlation coefficient and F value (dispersion ratio) are calculated by performing multiple regression analysis using P1 to Pn as explanatory variables. Here, as shown in FIG. 6, when focusing on the wafer A, the objective variable is 90.4, and the explanatory variables are 38.3 (P1), 90.8 (P2),... .9 (Pn-1) and 68.1 (Pn). In the multiple regression analysis, the partial correlation coefficient and the F value indicate the degree of contribution of the explanatory variable (process parameter) to the objective variable (inspection information). That is, it can be said that the greater the partial correlation coefficient and the absolute value of the F value, the greater the contribution to the yield. Thereafter, the information organizing unit 33 (see FIG. 3) of the yield analysis apparatus 4 ranks the process parameters in descending order of the contribution to the yield with reference to the F value. FIG. 7 shows the result of ranking the process parameters. From the result of FIG. 7, process parameters having a large contribution to the yield are determined.

  Note that the above-described operation of step S5 (determination of process parameters having a large contribution to the yield) is performed without specifying a wafer with a low yield and a wafer with a high yield.

  Next, in step S6 of FIG. 4, the analysis unit 32 of the yield analysis apparatus 4 determines whether or not there is a wafer with a low yield. When it is determined that there is no wafer with a low yield, the process returns to step S1 in FIG. On the other hand, if it is determined that there is a wafer with a low yield, the process proceeds to step S7 in FIG.

  Next, in step S7 of FIG. 4, the feedback means 34 (see FIG. 3) of the yield analyzer 4 feeds back the cause of the yield reduction to the manufacturing process corresponding to the process parameter having a large contribution to the yield. .

  Here, FIG. 8 shows the relationship between the process parameter (time between manufacturing steps) having the highest degree of contribution to the yield and the yield, and FIG. 9 shows the magnitude of the contribution to the yield. The relationship between the second-order process parameter (time between manufacturing steps) and the yield is shown. Referring to FIG. 8, in the manufacturing process corresponding to the process parameter having the highest degree of contribution to the yield, the time between the manufacturing processes is less than T10 (limit value). It can be said. Further, referring to FIG. 9, in the manufacturing process corresponding to the process parameter having the second largest contribution to the yield, the time between the manufacturing processes exceeds T20 (limit value). It can be said that. The above limit values are temporarily stored in the limit value database 37b (see FIG. 3) of the yield analysis device 4. The determination of the yield level is made based on a preset reference yield.

  In the feedback operation, the limit value of the time between the manufacturing processes described above is transmitted to the manufacturing process management apparatus 2. The limit value transmitted to the manufacturing process management device 2 is stored in the limit value database 17e (see FIG. 2). Thereby, the manufacturing process management device 2 adjusts the time between the manufacturing processes based on the limit value of the time between the manufacturing processes for the manufacturing process corresponding to the process parameter having a large contribution to the yield.

  In the first embodiment, as described above, the processing history information is collected for each manufacturing apparatus 6, and the process parameters for processing the wafer are calculated based on the processing history information collected for each manufacturing apparatus 6. This makes it easy to obtain detailed process parameters (the time from the completion of the processing of the wafer of interest in a predetermined manufacturing process until the processing of the wafer of interest in another manufacturing process is started). Can do. Thereby, in 1st Embodiment, the cause of a yield fall can be specified with high accuracy.

  In the first embodiment, as described above, the process parameter (explanatory variable) is the process of processing the wafer of interest in another manufacturing process after the processing of the wafer of interest in a predetermined manufacturing process is completed. By using the time until the start, the influence of the wafer leaving time on the yield can be examined.

  In the first embodiment, as described above, by providing the feedback means 34 in the yield analysis apparatus 4, it is possible to easily improve the yield reduction.

  In the first embodiment, since many process parameters (explanatory variables) are used when specifying the cause of the yield reduction, the cause of the yield reduction can be specified with higher accuracy.

(Second Embodiment)
FIG. 10 is a diagram for explaining process parameters calculated by the process management system according to the second embodiment of the present invention. 11 and 12 are diagrams for explaining the multiple regression analysis performed in the process management system according to the second embodiment of the present invention. Next, the operation (process management method) of the process management system according to the second embodiment will be described with reference to FIGS. The configuration of the process management system according to the second embodiment is the same as the configuration of the process management system according to the first embodiment shown in FIGS. 1 to 3, and the operation flow of the process management system according to the second embodiment is as follows. This is the same as the operation flow of the process management system of the first embodiment shown in FIG.

  In the second embodiment, first, in step S1 of FIG. 4, processing history information is collected for each manufacturing apparatus 6 by the control means 12 of the manufacturing process management apparatus 2 as in the first embodiment.

  Next, in step S <b> 2 of FIG. 4, process parameters are calculated based on the processing history information (wafer identification number and processing time) collected for each manufacturing apparatus 6. Specifically, in the second embodiment, processing of the target wafer is started after processing of another wafer to be processed before the target wafer is processed in the predetermined manufacturing apparatus 6. Time (hereinafter, idling time) is calculated. The idling time as a process parameter according to the second embodiment will be described below with reference to FIG. Note that identification numbers A to D are assigned to the wafer shown in FIG. In FIG. 10, the wafer is processed in the order of the manufacturing process M1, the manufacturing processes M2, M3,. Further, DB1 to DBn in FIG. 10 represent databases in which processing history information collected for each manufacturing apparatus 6 is stored.

  When the wafer A is focused on, the idling time as the process parameter of the second embodiment includes T11, T12, and T13 in FIG. Specifically, the idling time (process parameter) T11 is determined after the processing of the wafer C (processing of another wafer) is completed in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process M2. This is the time until the processing (processing of the target wafer) is started. In addition, the idling time (process parameter) T12 is determined after the processing of the wafer C (processing of another wafer) is completed in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process M3. This is the time until the processing of the target wafer) starts. In addition, the idling time (process parameter) T13 is determined after the processing of the wafer C (processing of another wafer) is completed in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process Mn. This is the time until the processing of the target wafer) starts.

  The calculation of the idling time as the process parameter described above is performed for all wafers including wafers A to D. Further, the calculation of the idling time as the process parameter described above may be performed over all the manufacturing steps or only for a specific manufacturing step. The calculated idling time as a process parameter is stored in the processing time database 17a of the manufacturing process management device 2.

  Next, in step S3 of FIG. 4, the test process management device 3 collects inspection information of chips inspected by the tester 7 for each wafer. Note that the yield for each wafer is calculated from the inspection information collected at this time.

  Next, in step S4 of FIG. 4, process parameters (idling time) are collected from the manufacturing process management apparatus 2 by the analysis means 32 of the yield analysis apparatus 4, and inspection information (yield for each wafer) is collected from the test process management apparatus. Collect from 3. Thereafter, the analysis unit 32 of the yield analysis device 4 associates the process parameter (idling time) with the inspection information (yield for each wafer). FIG. 11 shows a part of the result of associating process parameters (idling time) and inspection information (yield for each wafer). Note that the process parameters P11 to Pn in FIG. 11 are the idling time calculated in step S3.

  Next, in step S5 of FIG. 4, the analysis means 32 of the yield analysis apparatus 4 uses the inspection information (yield for each wafer) as an objective variable for all wafers including the wafers A to D, and process parameters (idling time). A partial correlation coefficient and an F value (dispersion ratio) are calculated by performing multiple regression analysis using P11 to Pn as explanatory variables. Here, as shown in FIG. 11, when attention is paid to wafer A, the objective variable is 96.2, and the explanatory variables are 6.25 (P11), 13.58 (P12),... .72 (Pn-1) and 5.15 (Pn). Thereafter, the information organizing unit 33 of the yield analysis device 4 ranks the process parameters in descending order of the contribution to the yield with the F value as a reference. FIG. 12 shows the result of ranking the process parameters. From the results shown in FIG. 12, process parameters having a large contribution to the yield are determined.

  Next, in step S6 of FIG. 4, the analysis unit 32 of the yield analysis apparatus 4 determines whether or not there is a wafer with a low yield. When it is determined that there is no wafer with a low yield, the process returns to step S1 in FIG. On the other hand, if it is determined that there is a wafer with a low yield, the process proceeds to step S7 in FIG.

  Next, in step S7 of FIG. 4, the cause of the yield reduction is fed back to the manufacturing apparatus 6 corresponding to the process parameter having a large contribution to the yield by the feedback means 34 of the yield analysis apparatus 4. Specifically, in the same manner as in the first embodiment, the limit value of the idling time at which the yield decreases is calculated, and the limit value of the idling time is transmitted to the manufacturing process management device 2. Thereby, the idling time is adjusted by the manufacturing process management device 2 based on the limit value of the idling time for the manufacturing device 6 corresponding to the process parameter having a large contribution to the yield.

  In the second embodiment, as described above, the processing history information is collected for each manufacturing apparatus 6, and the process parameters for processing the wafer are calculated based on the processing history information collected for each manufacturing apparatus 6. As a result, the detailed process parameters (from the processing of another wafer to be processed before the target wafer is processed in the predetermined manufacturing apparatus 6 until the processing of the target wafer is started) Time) can be easily obtained. Thereby, in 2nd Embodiment, the cause of a yield fall can be specified with high accuracy.

  In the second embodiment, as described above, as a process parameter (explanatory variable), after processing of another wafer processed before the wafer of interest is processed in a predetermined manufacturing apparatus 6 is completed. By using the time until the processing of the target wafer is started, it is possible to examine the influence of the idle time (idling time) of the manufacturing apparatus 6 on the yield.

  The remaining effects of the second embodiment are similar to those of the aforementioned first embodiment.

(Third embodiment)
FIG. 13 is a diagram for explaining process parameters calculated by the process management system according to the third embodiment of the present invention. Next, an operation (process management method) of the process management system according to the third embodiment will be described with reference to FIG. The configuration of the process management system according to the third embodiment is the same as the configuration of the process management system according to the first embodiment shown in FIGS. 1 to 3, and the operation flow of the process management system according to the third embodiment is as follows. This is the same as the operation flow of the process management system of the first embodiment shown in FIG.

  In the third embodiment, first, in step S1 of FIG. 4, processing history information is collected for each manufacturing apparatus 6 by the control means 12 of the manufacturing process management apparatus 2 as in the first embodiment.

  Next, in step S <b> 2 of FIG. 4, process parameters are calculated based on the processing history information (wafer identification number and processing time) collected for each manufacturing apparatus 6. Specifically, in the third embodiment, in a predetermined manufacturing apparatus 6, processing of a target wafer is started after processing of another wafer to be processed is started before the target wafer is processed. Time (hereinafter, processing start interval) is calculated. Hereinafter, with reference to FIG. 13, a processing start interval as a process parameter according to the third embodiment will be described. Incidentally, identification numbers A to D are given to the wafer shown in FIG. In FIG. 13, the wafer is processed in the order of the manufacturing process M1, the manufacturing processes M2, M3,. Further, DB1 to DBn in FIG. 13 represent databases in which processing history information collected for each manufacturing apparatus 6 is stored.

  When attention is paid to the wafer A, the processing start interval as a process parameter of the third embodiment includes T21 to T23 in FIG. Specifically, the processing start interval (process parameter) T21 is determined after the processing of the wafer C (processing of another wafer) is started in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process M2. This is the time until the processing of A (processing of the target wafer) is started. Further, the processing start interval (process parameter) T22 is the processing of the wafer A after the processing of the wafer C (processing of another wafer) is started in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process M3. This is the time until the processing of the target wafer is started. Further, the processing start interval (process parameter) T23 is the processing of the wafer A after the processing of the wafer C (processing of another wafer) is started in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process Mn. This is the time until the processing of the target wafer is started.

  The calculation of the processing start interval as the process parameter described above is performed for all wafers including wafers A to D. Further, the calculation of the processing start interval as the process parameter described above may be performed over all the manufacturing steps, or may be performed only for a specific manufacturing step. The calculated processing start interval as the process parameter is stored in the processing time database 17a of the manufacturing process management apparatus 2.

  Next, in step S3 of FIG. 4, the test process management device 3 collects inspection information of chips inspected by the tester 7 for each wafer. Note that the yield for each wafer is calculated from the inspection information collected at this time.

  Next, in step S4 of FIG. 4, process parameters (processing start intervals) are collected from the manufacturing process management apparatus 2 by the analysis means 32 of the yield analysis apparatus 4, and inspection information (yield for each wafer) is managed in the test process. Collect from device 3. Thereafter, the analysis unit 32 of the yield analysis apparatus 4 associates the process parameter (processing start interval) with the inspection information (yield for each wafer).

  Next, in step S5 of FIG. 4, the analysis means 32 of the yield analysis device 4 uses the inspection information (yield for each wafer) as an objective variable for all wafers including the wafers A to D, and process parameters (processing start interval). ) As an explanatory variable, a partial correlation coefficient and an F value (dispersion ratio) are calculated. Thereafter, the information organizing unit 33 of the yield analysis device 4 ranks the process parameters in descending order of the contribution to the yield with the F value as a reference. Thereby, a process parameter having a large contribution to the yield is determined.

  The subsequent operations are the same as in the first embodiment.

  In the third embodiment, as described above, the processing history information is collected for each manufacturing apparatus 6, and the process parameters for processing the wafer are calculated based on the processing history information collected for each manufacturing apparatus 6. As a result, detailed process parameters (from the start of processing of another wafer to be processed before the target wafer is processed in a predetermined manufacturing apparatus 6 to the start of processing of the target wafer are started. Time) can be easily obtained. Thereby, in 3rd Embodiment, the cause of a yield fall can be specified with high accuracy.

  In the third embodiment, as described above, as a process parameter (explanatory variable), a predetermined manufacturing apparatus 6 starts processing another wafer that is processed before the target wafer is processed. When a plurality of wafers having different identification numbers are processed in parallel by the same manufacturing apparatus 6 by using the time until the processing of the target wafer is started, the processing start interval in the manufacturing apparatus 6 is You can investigate the impact on yield.

  The remaining effects of the third embodiment are similar to those of the aforementioned first embodiment.

(Fourth embodiment)
FIG. 14 is a diagram for explaining process parameters calculated by the process management system according to the fourth embodiment of the present invention. Next, the operation (process management method) of the process management system according to the fourth embodiment will be described with reference to FIG. In addition, the structure of the process management system by 4th Embodiment is the same as that of the process management system of 1st Embodiment shown in FIGS. 1-3, and the operation | movement flow of the process management system by 4th Embodiment is as follows. This is the same as the operation flow of the process management system of the first embodiment shown in FIG.

  In the fourth embodiment, first, in step S <b> 1 of FIG. 4, processing history information is collected for each manufacturing apparatus 6 by the control means 12 of the manufacturing process management apparatus 2 as in the first embodiment.

  Next, in step S <b> 2 of FIG. 4, process parameters are calculated based on the processing history information (wafer identification number) of the manufacturing apparatus 6 collected for each manufacturing apparatus 6. Specifically, in the fourth embodiment, the number of times that the wafer of interest is processed by the same manufacturing apparatus 6 over a plurality of manufacturing steps (hereinafter referred to as the number of processing times) is calculated. Hereinafter, with reference to FIG. 14, the number of processes as a process parameter according to the fourth embodiment will be described. Incidentally, identification numbers A to D are assigned to the wafer shown in FIG. In addition, apparatus numbers 6a to 6d are assigned to the manufacturing apparatus 6 illustrated in FIG. In FIG. 14, the wafer is processed in the order of the manufacturing process M1, the manufacturing processes M2, M3,.

  When the wafer B is focused on, the number of processes as process parameters in the fourth embodiment is four. That is, the wafer B (target wafer) is processed four times by the manufacturing apparatus 6b (the same manufacturing apparatus) over a plurality of manufacturing steps M1 to Mn.

  Note that the calculation of the number of processes as the process parameter is performed for all wafers including the wafers A to D. Further, the calculation of the number of processes as the process parameter described above may be performed over all the manufacturing steps, or may be performed only for a specific plurality of manufacturing steps. Further, the calculated processing number as the process parameter is stored in the first processing number database 17b (see FIG. 2) of the manufacturing process management apparatus 2.

  Next, in step S3 of FIG. 4, the test process management device 3 collects inspection information of chips inspected by the tester 7 for each wafer. Note that the yield for each wafer is calculated from the inspection information collected at this time.

  Next, in step S4 of FIG. 4, the analysis means 32 of the yield analysis device 4 collects process parameters (number of processes) from the manufacturing process management device 2, and also inspects inspection information (yield for each wafer) in the test process management device. Collect from 3. Thereafter, the analysis means 32 of the yield analysis apparatus 4 associates process parameters (number of processes) with inspection information (yield for each wafer).

  Next, in step S5 of FIG. 4, the analysis means 32 of the yield analysis apparatus 4 uses the inspection information (yield for each wafer) as an objective variable for all wafers including the wafers A to D, and process parameters (number of processes). Is used as an explanatory variable to calculate a partial correlation coefficient and an F value (variance ratio). Thereafter, the information organizing unit 33 of the yield analysis apparatus 4 ranks the process parameters in descending order of the contribution to the yield with reference to the F value. Thereby, a process parameter having a large contribution to the yield is determined.

  The subsequent operations are the same as in the first embodiment.

  In the fourth embodiment, as described above, the processing history information is collected for each manufacturing apparatus 6, and the process parameters for processing the wafer are calculated based on the processing history information collected for each manufacturing apparatus 6. By doing this, it is possible to easily obtain detailed process parameters (the number of times a target wafer is processed by the same manufacturing apparatus 6 over a plurality of manufacturing steps). Thereby, in 4th Embodiment, the cause of a yield fall can be specified with high accuracy.

  In the fourth embodiment, as described above, the process parameter (explanatory variable) is the same by using the number of times that the wafer of interest is processed by the same manufacturing apparatus 6 over a plurality of manufacturing steps. The influence of the number of times of processing (processing frequency) in the manufacturing apparatus 6 on the yield can be examined.

  The remaining effects of the fourth embodiment are similar to those of the aforementioned first embodiment.

(Fifth embodiment)
FIG. 15 is a diagram for explaining process parameters calculated by the process management system according to the fifth embodiment of the present invention. Next, the operation (process management method) of the process management system according to the fifth embodiment will be described with reference to FIG. The configuration of the process management system according to the fifth embodiment is the same as the configuration of the process management system according to the first embodiment shown in FIGS. 1 to 3, and the operation flow of the process management system according to the fifth embodiment is as follows. This is the same as the operation flow of the process management system of the first embodiment shown in FIG.

  In the fifth embodiment, first, in step S <b> 1 of FIG. 4, processing history information is collected for each manufacturing apparatus 6 by the control means 12 of the manufacturing process management apparatus 2 as in the first embodiment.

  Next, in step S <b> 2 of FIG. 4, process parameters are calculated based on the processing history information (wafer identification number and processing condition) collected for each manufacturing apparatus 6. Specifically, in the fifth embodiment, before a target wafer is processed in a predetermined manufacturing apparatus 6, the number of times other wafers are processed under the same processing conditions as the target wafer (hereinafter referred to as the number of processing times). Calculated). In the following, with reference to FIG. 15, the number of processes as a process parameter of the fifth embodiment will be described. Note that identification numbers A to C are assigned to the wafer shown in FIG. In FIG. 15, the wafer is processed in the order of the manufacturing process M1, the manufacturing process M2,.

  When the wafer A is focused on, the number of processes as process parameters in the fifth embodiment is 9, 4, and 1 times. That is, in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process M1, the wafer A (target wafer) is processed tenth after the processing conditions are switched. In addition, in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process M2, the wafer A (target wafer) is processed fifth after the processing conditions are switched. In addition, in the manufacturing apparatus 6 (predetermined manufacturing apparatus) included in the manufacturing process Mn, the wafer A (target wafer) is processed second after the processing conditions are switched.

  The calculation of the number of processes as the process parameter described above is performed for all wafers including the wafers A to C. Further, the calculation of the number of processes as the process parameter described above may be performed over all the manufacturing steps, or may be performed only for a specific manufacturing step. Further, the calculated processing number as the process parameter is stored in the second processing number database 17c (see FIG. 2) of the manufacturing process management apparatus 2.

  Next, in step S3 of FIG. 4, the test process management device 3 collects inspection information of chips inspected by the tester 7 for each wafer. Note that the yield for each wafer is calculated from the inspection information collected at this time.

  Next, in step S4 of FIG. 4, the analysis means 32 of the yield analysis device 4 collects process parameters (the number of processing times) from the manufacturing process management device 2 and the inspection information (yield for each wafer) is a test process management device. Collect from 3. Thereafter, the analysis means 32 of the yield analysis apparatus 4 associates process parameters (number of processes) with inspection information (yield for each wafer).

  Next, in step S5 of FIG. 4, the analysis means 32 of the yield analysis apparatus 4 uses the inspection information (yield for each wafer) as an objective variable for all wafers including the wafers A to C, and process parameters (number of processes). Is used as an explanatory variable to calculate a partial correlation coefficient and an F value (variance ratio). Thereafter, the information organizing means 33 of the yield analysis device 4 ranks the process parameters in descending order of the contribution to the yield on the basis of the F value. Thereby, a process parameter having a large contribution to the yield is determined.

  The subsequent operations are the same as in the first embodiment.

  In the fifth embodiment, processing history information is collected for each manufacturing apparatus 6 as described above, and process parameters for processing a wafer are calculated based on the processing history information collected for each manufacturing apparatus 6. By doing this, it is possible to easily obtain detailed process parameters (the number of times other wafers are processed under the same processing conditions as the target wafer before the target wafer is processed in the predetermined manufacturing apparatus 6). it can. Thereby, in 5th Embodiment, the cause of a yield fall can be specified with high accuracy.

  In the fifth embodiment, as described above, the process parameters (explanatory variables) are the same processing conditions as the wafers to which other wafers are focused before the wafers to be focused are processed in a predetermined manufacturing apparatus 6. By using the number of times processed in (1), it is possible to examine the influence of the number of times of processing under the same processing condition on the yield.

  The remaining effects of the fifth embodiment are similar to those of the aforementioned first embodiment.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above description of the embodiment but by the scope of claims for patent, and all modifications within the meaning and scope equivalent to the scope of claims for patent are included.

  For example, in the above-described embodiment, an example in which the present invention is applied to a process management system that manages a manufacturing process of a wafer in which chips are fabricated is shown. However, the present invention is not limited thereto, and a wafer in which chips are fabricated. The present invention can also be applied to a process management system that manages manufacturing processes of products other than the above.

  In the above embodiment, the yield is managed for each wafer. However, the present invention is not limited to this, and the yield may be managed for each lot including a plurality of wafers.

  In the above embodiment, the process parameter is calculated by the manufacturing process management device. However, the present invention is not limited to this, and the process parameter may be calculated by a yield analysis device.

It is a figure for demonstrating the whole structure of the process management system by 1st Embodiment of this invention. It is a figure for demonstrating the detail of the manufacturing process management apparatus contained in the process management system by 1st Embodiment shown in FIG. It is a figure for demonstrating the detail of the yield analysis apparatus contained in the process management system by 1st Embodiment shown in FIG. It is a flowchart for demonstrating operation | movement (process management method) of the process management system by 1st Embodiment of this invention. It is a figure for demonstrating the process parameter calculated with the process management system by 1st Embodiment of this invention. It is a figure for demonstrating the multiple regression analysis performed with the process management system by 1st Embodiment of this invention. It is a figure for demonstrating the multiple regression analysis performed with the process management system by 1st Embodiment of this invention. It is a figure for demonstrating the multiple regression analysis performed with the process management system by 1st Embodiment of this invention. It is a figure for demonstrating the multiple regression analysis performed with the process management system by 1st Embodiment of this invention. It is a figure for demonstrating the process parameter calculated with the process management system by 2nd Embodiment of this invention. It is a figure for demonstrating the multiple regression analysis performed with the process management system by 2nd Embodiment of this invention. It is a figure for demonstrating the multiple regression analysis performed with the process management system by 2nd Embodiment of this invention. It is a figure for demonstrating the process parameter calculated with the process management system by 3rd Embodiment of this invention. It is a figure for demonstrating the process parameter calculated with the process management system by 4th Embodiment of this invention. It is a figure for demonstrating the process parameter calculated with the process management system by 5th Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Process management system 2 Manufacturing process management apparatus 3 Test process management apparatus (inspection information collection means)
4 Yield analysis device 12 Control means (processing history information collection means, calculation means)
32 Analysis means (discrimination means)
34 Feedback means

Claims (9)

A process management method for managing a manufacturing process of a product manufactured by being processed in a plurality of manufacturing processes each including at least one manufacturing apparatus,
Collecting processing history information for each manufacturing apparatus;
Calculating process parameters when processing the product based on the processing history information collected for each manufacturing apparatus;
Collecting inspection information for calculating yield;
And a step of discriminating a process parameter having a large contribution to the yield by performing an analysis using the process parameter and the inspection information. Calculating the process parameter comprises:
Including a step of calculating a process parameter including a time from when processing of the product of interest in a predetermined manufacturing process is completed until processing of the product of interest in another manufacturing process is started. The process management method according to claim 1. Calculating the process parameter comprises:
A step of calculating a process parameter including a time until processing of the product of interest is started after processing of another product processed before the product of interest is processed in a predetermined manufacturing apparatus; The process management method according to claim 1, comprising: Calculating the process parameter comprises:
A step of calculating a process parameter including a time from the start of processing of another product processed before the target product is processed in a predetermined manufacturing apparatus to the start of processing of the target product; The process management method according to claim 1, comprising: Calculating the process parameter comprises:
When each of two or more manufacturing processes among the plurality of manufacturing processes includes the same manufacturing apparatus, the product of interest is processed by the same manufacturing apparatus over the plurality of manufacturing processes. The process management method according to claim 1, further comprising a step of calculating a process parameter including the number of times. Calculating the process parameter comprises:
A predetermined manufacturing apparatus includes a step of calculating a process parameter including a number of times that another product is processed under the same processing conditions as the target product before the target product is processed. Item 6. The process management method according to Item 1. After determining a process parameter having a large contribution to the yield,
The process according to any one of claims 1 to 6, further comprising a step of feeding back a cause of yield reduction to a manufacturing process or a manufacturing apparatus corresponding to a process parameter having a large contribution to the yield. Management method. A process management system for managing a manufacturing process of a product manufactured by being processed in a plurality of manufacturing processes each including at least one manufacturing apparatus,
Processing history information collecting means for collecting processing history information for each manufacturing apparatus;
Based on the processing history information collected for each of the manufacturing devices, calculating means for calculating a process parameter when processing the product,
Inspection information collection means for collecting inspection information for calculating yield;
A process management system comprising: a discriminating unit that discriminates a process parameter having a large contribution to the yield by performing an analysis using the process parameter and the inspection information.   9. The process management system according to claim 8, further comprising feedback means for feeding back a cause of yield reduction to a manufacturing process or a manufacturing apparatus corresponding to a process parameter having a large contribution to the yield.

Images