TW201903686A - Wafer manufacturing management method and wafer manufacturing management system - Google Patents

Abstract

A wafer manufacturing management method is provided. The wafer manufacturing management method includes following steps: according to a plurality of yield indicators, determining a plurality of processing data generated by the wafer products passing a plurality of processing equipment, to determine whether select at least one of the processing data to an abnormal processing data; integrating the process data to accumulate a plurality of cost data of the wafer product sequentially passing the processing equipment; according to the abnormal production data and the cost data, determining a cost of loss data corresponding to the cost data; and integrating the abnormal processing data and the cost of loss data to generate a warning data. In addition, a wafer manufacturing management system is also provided.

Description

Wafer manufacturing management method and wafer manufacturing management system

The invention relates to a management method and a management system, and in particular to a wafer manufacturing management method and a wafer manufacturing management system.

With the development of technology, as the development of various electronic products becomes more and more diverse, the demand for various types of wafer products also increases. In the process of wafer manufacturing, the manufacturing of wafer products needs to be processed sequentially through multiple process equipment. However, when each different process equipment processes the wafer, it may cause part of the wafer to be damaged, resulting in a waste of process cost. Moreover, in the current wafer manufacturing technology, the process equipment may not be able to monitor the process yield in real time and automatically determine whether the process equipment is malfunctioning due to the all-weather and automated operation.

In this regard, in the field of wafer manufacturing, traditional process equipment management can only obtain process information, but cannot effectively monitor product yield and loss costs in real time. Therefore, how to efficiently monitor the process results and yields of wafer products passing through various process equipment during the manufacturing process, and can automatically determine the process yield rate of each process equipment to automatically provide loss costs to managers is currently Important subject.

The invention provides a wafer manufacturing management method and a wafer manufacturing management system, which can effectively monitor the process results and yield of wafer products passing through various process equipment during the manufacturing process, and can automatically judge and analyze the process yield of each process equipment Situation to provide loss costs.

The wafer manufacturing management method of the present invention includes the following steps: judging multiple process data generated after a wafer product passes through multiple process equipment according to multiple yield indicators, to determine whether to select at least one of these process data as Abnormal process data; integrate these process data to accumulate multiple cost data after the wafer products sequentially pass through these process equipment; determine the loss cost data corresponding to the abnormal process data based on the abnormal process data and these cost data; and integrate the abnormal process Data and loss cost data to generate warning data.

In an embodiment of the invention, the above-mentioned wafer manufacturing management method further includes: determining the main yield index of at least one of these yield indexes based on the yield average value and the yield standard deviation; and if these processes If the yield of at least one of the data is lower than the main yield index, then at least one of these process data is selected as the abnormal process data.

In an embodiment of the present invention, the above-mentioned wafer manufacturing management method further includes: establishing a control range with multiple judgment limits based on the yield average and yield standard deviation; and establishing judgment conditions by using these judgment limits, If the yield of at least one of the process data meets the judgment condition, then at least one of the process data is selected as abnormal process data.

In an embodiment of the present invention, the aforementioned regulatory range is six yield standard deviations.

In an embodiment of the invention, the above-mentioned wafer manufacturing management method further includes: determining at least one of these yield indicators according to the product type of the wafer product; and if the yield of at least one of the process data If it is lower than at least one of the yield indicators, then at least one of these process data is selected as abnormal process data.

In an embodiment of the present invention, the above-mentioned wafer manufacturing management method further includes: obtaining these process data through a manufacturing execution system, and each of these process data includes product specification data and defective mode data.

In an embodiment of the present invention, the above wafer manufacturing management method further includes: obtaining failure modes corresponding to the abnormal process data and affecting the analysis result according to the defective mode data corresponding to the abnormal process data.

In an embodiment of the invention, the above wafer manufacturing management method further includes: determining whether at least one defective product in the wafer product can be reused according to the defective mode data corresponding to the abnormal process data.

In an embodiment of the present invention, the above wafer manufacturing management method further includes: periodically generating warning data according to a preset time period, and outputting the warning data, wherein the warning data further includes product specification data of abnormal process data, Yield, yield index, and bad model data.

The wafer manufacturing management system of the present invention includes a manufacturing execution system (Manufacturing Execution System, MES) and a yield management system (Yield Management Solution, YMS). The manufacturing execution system is used to provide multiple process data generated after the wafer product passes through multiple process equipment. The yield management system is coupled to the manufacturing execution system. The yield management system is used to receive multiple yield indicators, multiple cost data, and multiple process data corresponding to these process equipment. The yield management system judges the process data generated after the wafer product passes through the process equipment according to the yield indicators, and decides whether to select at least one of the process data as abnormal process data. The yield management system integrates these process data to accumulate multiple cost data after the wafer products pass through these process devices in sequence. The yield management system determines the loss cost data corresponding to the abnormal process data according to one of the cost data corresponding to the abnormal process data. The yield management system integrates abnormal process data and loss cost data to generate warning data.

Based on the above, the wafer manufacturing management method and wafer manufacturing management system of the present invention can collect process data generated by a wafer product after passing through multiple process equipments through the manufacturing execution system, and determine the abnormal yield rate by the yield management system Process data and integrate cost data to generate loss cost data. Therefore, the wafer manufacturing management method and wafer manufacturing management system of the present invention can provide loss cost data in real time, so that managers can efficiently perform process monitoring and management.

In order to make the above-mentioned features and advantages of the present invention more obvious and understandable, the embodiments are specifically described below and described in detail in conjunction with the accompanying drawings.

In order to make the content of the present invention easier to understand, the following specific embodiments are taken as examples on which the present invention can indeed be implemented. In addition, wherever possible, elements / components / steps using the same reference numbers in the drawings and embodiments represent the same or similar components.

FIG. 1 is a block diagram of a wafer manufacturing management system according to an embodiment of the invention. The wafer manufacturing management system 100 includes a manufacturing execution system 110 (Manufacturing Execution System, MES), cost data 111, yield index 112, yield management system 120 (Yield Management Solution, YMS), and warning data 130. In this embodiment, the manufacturing execution system 110 can manage and integrate the process data 113 and the defect mode data 114, where the process data 113 includes process equipment data 1131 and product specification data 1132. In addition, the wafer manufacturing management system 100 inputs the relevant data parameters of the manufacturing execution system 110 (Manufacturing Execution System, MES), cost data 111, and yield index 112 to the yield management system 120. In this embodiment, when the yield of the wafer product is abnormal, or a certain process equipment is abnormal, the yield management system 120 can generate the warning data 130 according to the above-mentioned relevant data parameters, and can be real-time or according to the preset Time period to provide managers.

In one embodiment, the wafer manufacturing management system 100 can output the warning data 130 to a display device or an external electronic device through wireless or wired transmission, but the invention is not limited thereto. For example, the wafer manufacturing management system 100 may provide the warning data 130 to the administrator's email mailbox or handheld electronic device in the form of text messages, emails, electronic reports, or the like. In other words, the wafer manufacturing management system 100 of this embodiment can provide the warning data 130 to the manager in real time or at a preset time period, so that the manager can efficiently monitor and manage the process.

In this embodiment, the wafer manufacturing management system 100 can be configured in an electronic device and implemented in the form of software programs or hardware circuits. In this embodiment, the electronic device may be, for example, a computing device such as a file server, database server, application server, workstation or personal computer. And, the electronic device may include a storage device and a processor. The storage device may be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory (Flash memory), or the like Element or a combination of the above elements.

In this embodiment, the storage device may use storage cost data 111, yield index 112, process equipment data 1131, product specification data 1132, and bad mode data 114. In addition, the storage device can also be used to store the abnormal judgment data module 121, the loss cost calculation module 122, and the failure mode and effects analysis module 123 (FMEA) in the yield management system 120.

In this embodiment, the processor is, for example, a central processing unit (Central Processing Unit, CPU), or other programmable general-purpose or special-purpose microprocessor (Microprocessor), digital signal processor (Digital Signal Processor, DSP), programmable controller, application specific integrated circuits (ASIC), programmable logic device (Programmable Logic Device, PLD) or other similar devices or a combination of these devices. The electronic device equipped with the wafer manufacturing management system 100 can be coupled to a plurality of process equipment to receive and establish various data described in various embodiments of the present invention. In addition, the wafer manufacturing management system 100 can execute the above-mentioned various module functions by the processor to implement various data operations, integration, judgment, and analysis described in the embodiments of the present invention.

In this embodiment, the manufacturing execution system 110 may be connected to a plurality of process equipment, and used to record a plurality of process data 113 after the wafer products are processed through the process of these process equipment. These process data 113 may include process equipment data 1131 and product specification data 1132, respectively. In this embodiment, the process equipment data 1131 can represent the machine data of each process equipment, and the product specification data 1132 can be, for example, the name of the wafer product and wafer characteristics. For example, please refer to Table 1 below. Table 1

In the present embodiment, the process data 113 may, for example, present the content shown in Table 1. In the above Table 1, the process equipment information 1131 may include, for example, station name, end time of the station, leaflet number, etc., and the like. The product specification data 1132 may include, for example, a product number, a product name, a crystal orientation, a resistance range, a slice thickness, etc., and the like. That is to say, the above Table 1 can be used to represent the relevant equipment information and product specifications after the polishing sheet passes through the process equipment classified by flatness. However, the data content of the above Table 1 can be determined according to different product types or process requirements, or set by the administrator, and the present invention is not limited to this.

In this embodiment, the wafer manufacturing management system 100 may receive a plurality of cost data 111 and a plurality of yield indicators 112 from the outside or input through a manager. In this embodiment, the cost data 111 is used to represent the cost required for the wafer product to pass through the process equipment, such as equipment cost, material cost, staff time, etc. The present invention is not limited. In addition, these yield indicators 112 are used to determine whether the yield of wafer products after passing through these process equipment is abnormal. In this embodiment, the bad mode data 114 may represent the data of the cause of the wafer damage that may occur after the wafer products pass through these process equipment, the number of scraps, and whether they can be reused. For example, please refer to Table 2 below. Table 2

In this embodiment, the bad mode data 114 can be integrated with the process equipment data 1131 to present the data content shown in Table 2, for example. In the above Table 2, the process equipment information 1131 may include, for example, station name, station number, leaflet number, etc., and the like. The bad pattern data 114 may include, for example, data on the cause of the defect, the number of scraps, and whether it can be reused. However, the data content of the above Table 2 can be determined according to different product types or process requirements, or set by the administrator. The present invention is not limited to this.

In this embodiment, the yield management system 120 may include an abnormal data judgment module 121, a loss cost calculation module 122, and a failure mode and impact analysis module 123. The abnormality determination module 121 can be used to determine whether the yield of the wafer product after passing through these process equipment is lower than the corresponding yield index 112 according to the yield index 112, and take out the abnormal process data as abnormal process data.

In this embodiment, the yield index 112 may include a primary yield index and a secondary yield index. In this embodiment, the main yield index of the yield index 112 may be determined by calculating the average yield and the standard deviation of the yield based on the process history of the process equipment, for example. For example, please refer to the following formula (1) and formula (2). …………(1) …………(2)

In the above formula (1), μ is the average yield, X ₁ , X ₂ , and X _N are the historical yields of a certain time interval (such as 3 months, half a year, etc.) of the process equipment, where N is greater than zero Positive integer. In the above formula (2), σ is the standard deviation yield, X _i is the historical yield process equipment. Therefore, in this embodiment, the wafer manufacturing management system 100 may set μ-3σ (yield average value-3 yield standard deviations) as the main yield index. If the yield of a certain process data is lower than the corresponding main yield index (μ-3σ), then this process data is selected as an abnormal process data. It should be noted that the yield index 112 of different process equipment is different. The yield index 112 of each process equipment can be determined according to the respective process historical yield.

However, in an embodiment, the secondary yield index of the yield index 112 may be based on the above-described yield average μ and yield standard deviation σ to establish a control range with multiple judgment limits, and by these judgment limits Establish judgment conditions. For example, please refer to Table 3 below. table 3

In the above Table 3, the wafer manufacturing management system 100 may establish a control range with multiple determination limits, where the determination limits may be, for example, the first zone to the sixth zone. In other words, the control range in Table 3 is six standard deviations of yield. The wafer manufacturing management system 100 can set a judgment condition, for example, at least two of the process data of three consecutive production batches of a certain process equipment are lower than the sixth zone or the lower limit of the control, or five consecutive processes At least four of the yields in the data are lower than the fifth zone or the sixth zone or the lower control limit, then the yield management system 120 determines that these continuous process data are abnormal process data. Alternatively, the wafer manufacturing management system 100 may set a judgment condition such that the yield rate in the continuous six process data of a certain process equipment continues to decrease, and the yield management system 120 determines that the continuous process data is abnormal process data.

However, in an embodiment, the yield index 112 can also be determined according to the product type of the wafer product. For example, if the current wafer product is a by-product, the yield index of the current wafer product in the process equipment can refer to the yield index of another wafer product belonging to the main product in the process equipment, and is not limited to The definition method of the above yield index.

In this embodiment, the loss cost calculation module 122 can be used to calculate the loss cost data corresponding to the abnormal process data. In this embodiment, the loss cost calculation module 122 can integrate these process data, and accumulate multiple cost data 111 after the wafer product sequentially passes through these process equipment. The loss cost calculation module 122 can determine the loss cost data corresponding to the abnormal process data according to the cost data 111 and the bad pattern data 114. In this embodiment, the cost data 111 may be, for example, the manufacturing cost of a single wafer product passing through a single-station process equipment, and the bad mode data 114 may be, for example, the number of defective products. For example, please refer to Table 4 below. Table 4

The above Table 4 can be used to represent the results of the process data integrated by the loss cost calculation module 122, the single-chip cost data of each single station, and the number of defective products. In this embodiment, if the yield of the wafer product at the flatness classification site is lower than the yield index, the loss cost calculation module 122 may integrate the process data as shown in Table 4 above, and the cumulative wafer product order Multiple cost data after passing through these process equipments. For example, please refer to Table 5 below. table 5

The above Table 5 can be used to represent the cumulative results of multiple cost data after a single wafer of the wafer product accumulated by the loss cost calculation module 122 passes through these process equipments in sequence. In this embodiment, the loss cost calculation module 122 can count the number of defective pieces in Table 4 and the cumulative cost data of the single piece in Table 5 to calculate the loss cost data of abnormal process data (for example, 10 × 89.1 = 891 (NT )). In other words, the loss cost calculation module 122 may determine the loss cost data corresponding to the abnormal process data according to the abnormal process data and the cost data of each individual station. For example, please refer to Table 6 below. Table 6

The loss cost data of the embodiments of the present invention can be presented as shown in Table 6, for example. In Table 6 above, the loss cost data may include, for example, loss cost, reusable loss cost, and non-reusable loss cost. However, the content of the data in Table 6 above can be determined according to different product types or process requirements, or set by the administrator. The present invention is not limited to this. In this embodiment, the yield management system 120 can determine whether the wafer products belonging to the defective products can be reused (reworkable or non-reworkable) through the defective pattern data 114. For example, if the thickness of the defective wafer product is too thick, the resistance is too high / low, or the wafer is contaminated, it is a type that can be reused or processed through other processes to become a good product. low. However, if the wafer flatness of the defective wafer product is poor, the line marks appear on the wafer surface, the wafer thickness is too thin, or the radial resistivity of the wafer is not uniform, it is either irreusable or cannot be processed through other processes After becoming a good product type, the cost of loss may be higher. Please refer to Table 7 below. Table 7

In this embodiment, the defect mode data 114 may include data on whether defective wafers can be reused as shown in Table 7, and the yield management system 120 of this embodiment calculates the loss cost data as shown in Table 6 above The yield management system 120 can determine that the type of defective wafer is defective wafer flatness. According to Table 7, it is determined that the defective wafer is a type that cannot be reused, and the defective wafer in Table 6 is determined to be unavailable. Lost cost of reuse.

In this embodiment, the failure mode and impact analysis module 123 can search for corresponding potential causes and prevention mechanisms recorded in the database according to the bad mode. For example, please refer to Table 8 below. Table 8

In this embodiment, the failure mode and impact analysis module 123 can generate failure analysis data as shown in Table 8 above. The failure analysis data in Table 8 above may include, for example, potential failure modes, potential failure effects, severity, potential causes, preventive control, and detection control. In this embodiment, the yield management system 120 can output failure analysis data to the manager, so that the manager can efficiently troubleshoot the process equipment. However, the data content of the above Table 8 can be determined according to different product types or process requirements, or set by the administrator, and the present invention is not limited to this.

In this embodiment, the yield management system 120 can generate warning data 130 according to the calculation and analysis results of the abnormal data module 121, the loss cost calculation module 122, and the failure mode and impact analysis module 123, wherein the warning data 130 can be It includes at least one of product specification data, yield, yield index, and bad mode data about abnormal process data. For example, the warning data 130 may include data such as Table 1, Table 2, Table 6, and Table 8 described above. Moreover, in this embodiment, the yield management system 120 may periodically generate and output warning data 130 to the manager according to a preset time period (for example, every six hours, daily report, weekly report, etc.).

FIG. 2 is a block diagram of a plurality of process devices according to an embodiment of the present invention. Refer to FIGS. 1 and 2. For example, in FIG. 2, the wafer product 200 may be processed through multiple stages of multiple process equipment, such as slicing equipment 310, smoothing equipment 320, etching equipment 330, and polishing equipment 340, for example. Moreover, these process equipments can generate process data 113_1 ~ 113_4, respectively. The manufacturing execution system 110 can manage these process data 113_1 ~ 113_4, and provide these process data 113_1 ~ 113_4 and a plurality of corresponding cost data 111, a plurality of corresponding yield index 112, a plurality of corresponding bad mode data 114 to yield Management system 120.

In this embodiment, when the yield management system 120 determines that the yield of the wafer product 200 passes through the leveling device 320 is abnormal by determining the abnormal data module 121, the yield management system 120 uses the loss cost calculation module 122 Calculate the lost cost of the wafer product 200 after passing through the leveling device 320. In addition, the yield management system 120 can also generate the failure analysis data as shown in Table 8 above through the failure mode and impact analysis module 123. Therefore, in this embodiment, the yield management system 120 may output warning data 130 that includes at least loss cost data and failure analysis data to the manager. In this regard, after receiving the warning data 130 including at least loss cost data and failure analysis data, the administrator may perform troubleshooting or equipment adjustment on the grinding equipment 320. In other words, the manager can efficiently manage these process equipments through the wafer manufacturing management system 100.

3 is a flowchart of a wafer manufacturing management method according to an embodiment of the invention.

Referring to FIGS. 1 and 3, the wafer manufacturing management method of this embodiment can be applied to at least the wafer manufacturing management system 100 of FIG. 1. In step S410, the yield management system 120 can separately determine a plurality of process data generated by at least one wafer product passing through a plurality of process equipment according to a plurality of yield indicators to determine whether to select at least one of the process data It is abnormal process data. In step S420, the yield management system 120 may integrate the process data to accumulate multiple cost data after the wafer products pass through the process equipment in sequence. In step S430, the yield management system 120 may determine the loss cost data corresponding to the abnormal process data according to the abnormal process data and the cost data. In step S440, the yield management system 120 may integrate abnormal process data and loss cost data to generate warning data. Therefore, the wafer manufacturing management method of this embodiment can effectively monitor the process results and yields of wafer products passing through each process equipment during the manufacturing process, and can automatically determine the yield rate of each process equipment to provide real-time warning data. However, the relevant system features and data processing methods for implementing the wafer manufacturing management method of this embodiment can be referred to the descriptions of the above embodiments of FIGS. 1 to 2 to obtain sufficient teaching, suggestions, and implementation instructions, and thus will not be repeated.

In summary, the wafer manufacturing management method and wafer manufacturing management system of the present invention can collect process data generated by a wafer product after passing through multiple process equipments through the manufacturing execution system, and execute multiple molds through the yield management system To determine the process data of abnormal yield, the yield management system can integrate multiple cost data and abnormal process data to generate cost information about the loss of wafer products in these process equipment, and the yield management system also Failure analysis data can be generated by executing failure modes and impact analysis modules. Therefore, the wafer manufacturing management method and the wafer manufacturing management system of the present invention can not only provide loss cost data in real time, but also provide failure analysis data related to process equipment with abnormal yields, so that managers can efficiently perform Wafer process monitoring and process management.

Although the present invention has been disclosed as above with examples, it is not intended to limit the present invention. Any person with ordinary knowledge in the technical field can make some changes and modifications without departing from the spirit and scope of the present invention. The scope of protection of the present invention shall be subject to the scope defined in the appended patent application.

100‧‧‧ Wafer Manufacturing Management System

110‧‧‧Manufacturing execution system

111‧‧‧ Cost information

112‧‧‧ Yield index

1131‧‧‧Process equipment information

1132‧‧‧Product specifications

114‧‧‧ Bad Mode Information

120‧‧‧ Yield Management System

121‧‧‧Judge Abnormal Information Module

122‧‧‧ Loss cost calculation module

130‧‧‧Warning information

200‧‧‧ Wafer products

310‧‧‧Slicing equipment

320‧‧‧Smoothing equipment

330‧‧‧Etching equipment

340‧‧‧Polishing equipment

113, 113_1, 113_2, 113_3, 113_4, 113_5‧‧‧ process information

123‧‧‧ Failure Mode and Impact Analysis Module

S410, S420, S430, S440

FIG. 1 is a block diagram of a wafer manufacturing management system according to an embodiment of the invention. FIG. 2 is a block diagram of a plurality of process equipment according to an embodiment of the invention. 3 is a flowchart of a wafer manufacturing management method according to an embodiment of the invention.

Claims (10)

A wafer manufacturing management method includes: judging a plurality of process data generated after at least one wafer product passes through a plurality of process equipment according to a plurality of yield indicators, to decide whether to select at least one of the process data is An abnormal process data; integrate the process data to accumulate multiple cost data after the wafer product passes through the process equipment in sequence; based on the abnormal process data and the cost data, determine a corresponding to the abnormal process data Lost cost data; and integrate the abnormal process data and the lost cost data to generate a warning data. The wafer manufacturing management method as described in item 1 of the patent application scope further includes: determining a main yield index of at least one of the yield indexes based on a yield average and a yield standard deviation; And if at least one of the process data has a yield rate lower than the main yield index, then at least one of the process data is selected as the abnormal process data. The wafer manufacturing management method as described in item 1 of the patent application scope further includes: establishing a control range with multiple judgment limits based on a yield average and a yield standard deviation; and by these judgment limits A judgment condition is established, and if a yield of at least one of the process data meets the judgment condition, at least one of the process data is selected as the abnormal process data. The wafer manufacturing management method as described in item 1 of the patent application scope, wherein the control scope is six yield standard deviations. The wafer manufacturing management method as described in item 1 of the patent application scope further includes: determining at least one of the yield indicators according to a product type of the at least one wafer product; and if at least the process data One of the yield rates is lower than at least one of the yield index, and at least one of the process data is selected as the abnormal process data. The wafer manufacturing management method as described in item 1 of the patent application scope further includes: obtaining the manufacturing process data through a manufacturing execution system, and the manufacturing process data respectively include a product specification data and a defective mode data. The wafer manufacturing management method as described in item 6 of the patent application scope further includes: obtaining a failure analysis data corresponding to the abnormal process data according to the defective mode data corresponding to the abnormal process data. The wafer manufacturing management method described in item 6 of the patent application scope further includes: judging whether at least one defective product in the at least one wafer product can be reused according to the defective mode data corresponding to the abnormal process data. The wafer manufacturing management method as described in item 1 of the patent application scope further includes: periodically generating and outputting the warning data according to a preset time period, wherein the warning data further includes a product specification of the abnormal process data Data, a yield rate, a yield index, and a bad pattern data. A wafer manufacturing management system includes: a manufacturing execution system for providing multiple process data generated by at least one wafer product passing through multiple process equipment; and a yield management system coupled to the manufacturing execution system, Used to receive a plurality of yield indicators, a plurality of cost data and a plurality of process data corresponding to the process equipment, wherein the yield management system respectively judges that the at least one wafer product passes the processes according to the yield indicators The process data generated after the equipment determines whether to select at least one of the process data as an abnormal process data, wherein the yield management system integrates the process data to accumulate the at least one wafer product according to Multiple cost data after passing through the process equipment in sequence, wherein the yield management system determines a loss cost data corresponding to the abnormal process data according to one of the cost data corresponding to the abnormal process data, and the The yield management system integrates the abnormal process data and the loss cost data to generate a warning data.