CN1254846C - Quick warning system and method for continuous abnormal faults - Google Patents
Quick warning system and method for continuous abnormal faults Download PDFInfo
- Publication number
- CN1254846C CN1254846C CN 02148965 CN02148965A CN1254846C CN 1254846 C CN1254846 C CN 1254846C CN 02148965 CN02148965 CN 02148965 CN 02148965 A CN02148965 A CN 02148965A CN 1254846 C CN1254846 C CN 1254846C
- Authority
- CN
- China
- Prior art keywords
- defective
- fault
- detection data
- defects
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- General Factory Administration (AREA)
Abstract
The present invention relates to a quick warning system for continuous abnormal faults and a method thereof, which comprises a tendency chart for fault detection data, a fault alarm report form and an Email system, wherein the tendency chart for fault detection data is manufactured by using a fault data statistical system for obtaining the fault detection data through a statistical fault scanning station; the fault alarm report form is formed in a mode that a fault data analysis system analyzes the tendency chart for fault detection data in fixed time and records continuous abnormal fault events; the Email system sends the fault alarm report form in setting time to inform an engineer in charge. The engineer who receives the notification makes use of fault distribution figures of known events on a fault scanning figure data basis to compare a fault flaw distribution figure obtained by the fault scanning station, which distinguishes that whether the continuous abnormal events are the known events. Then, the engineer determines processing strategies.
Description
Technical field
The present invention is the caution system and the method for relevant a kind of wafer factory, particularly about continuously unusual quick caution system of defective and method in a kind of manufacture of semiconductor.
Background of invention
In the process that semiconductor crystal wafer is made, for the defective that produces because of board or processing procedure factor, generally speaking after finishing, certain some processing procedure needs to be provided with the defects detection station to detect its defects count and to analyze kind, its flow process as shown in Figure 1, at first step S10 does defective scanning (defectscan) to product, and then whether the defects count of step S11 statistics product exceeds the upper limit, if yes, then carry out step S13, differ to give and comprehend otherwise carry out step S12.The product that by operator defects count is exceeded the upper limit at step S13 is done defective and is checked (defect review); then will check result notification at step S14 operator is responsible for the engineer, shuts down according to checking the relevant board director of result notification step S15 engineer then.For the engineer, must wait until on-line operation person classify defect kind good and number of computations after, obtain notice and just can learn to single incident or have continuous anomalous event to take place, so work pattern easily because of artificial careless mistake postpone that anomalous event warns ageing.
In Taiwan patent announcement numbers 331650, mode and equipment at check and analysis on the line of the relevant defective of semiconductor factory, off-line defects detection are improved, it is the general flow process and the practice, and how the follow-up practice when not disclosing the generation of defective incident reduces the product loss that is caused because of the defective incident if reaching.
Therefore, a kind ofly can warn continuously unusual defective incident fast with the defective caution system and the method that reduce yield loss and be to be the institute Ji.
Summary of the invention
One of purpose of the present invention is to propose a kind of defective caution system and method, can warn continuously unusual defective incident fast, to reduce the loss that is caused the product yield because of continuously unusual defective.
The object of the present invention is achieved like this: a kind of continuously unusual quick caution system of defective comprises:
One defective scanning movement is to scan the product defects distribution scenario and to produce a defects detection data and make a defect map;
One defective statistics system adding up the defects count of this defects detection data, and makes a defects detection data tendency chart according to the priority of time;
One defective analysis system at this defects detection data tendency chart of set time inner analysis, and writes down the data that continuous two batches or above product defects quantity exceed the upper limit, to make defective caution form;
One e-mail system sees through the networking at preset time and transmits this defective caution form to being responsible for the engineer; And
One defective scanning patter data bank is stored many known event defect map, in order to compare the defect map of this defective scanning movement made.
A kind of continuously unusual quick alarming method for power of defective comprises the following steps:
The defective of scanning like product is also made a defect map and a defects detection data;
Add up the defects count of this defects detection data and make a defects detection data tendency chart according to time order and function;
Analyze this defects detection data tendency chart, write down continuous two batches or two batches of above defects count and exceed the data of formulating the upper limit, and according to this data make one defective caution form; And
Use an e-mail system to send this defective caution form automatically to responsible engineer.
More comprise and use a defective scanning patter data bank to compare the step of this defect map.
Comprise when this defect map when being own county magistrate's part the relevant board director of notice shut down.
Comprise that more notifying operator to do defective checks when the non-known event of this defect map, shut down according to checking the relevant board director of result notification again.
According to the present invention, a kind of defective caution system uses the obtained defects detection data of a defective statistics system statistics defective scanning movement, and make a defects detection data tendency chart, in the set time, one defective analysis network analysis defects detection data tendency chart, when the continuous product of the same type more than two batches or two batches of the defects count of a certain defective scanning movement exceeds in limited time, defective analysis system just writes down this data and makes defective caution form, and utilize e-mail system to send this defective caution form automatically to responsible engineer in the time of setting, the engineer is after obtaining this notice, utilize the defect distribution figure of known event in the defective scanning patter data bank to compare the continuous anomalous event defect distribution figure that the defective scanning movement is scanned, to differentiate whether this continuous anomalous event is known event.
Characteristics of the present invention are to allow the engineer finish one step of preceding morning that defective checks at operator and learn whether have continuous anomalous event to take place; can preferentially carry out defective by operator for the anomalous event that can't judge reason checks; if known event then engineer needn't wait until that the result that defective is checked comes out; just can directly notify relevant board director to shut down, so can reduce the loss that causes the product yield because of continuous defect.
Description of drawings
Fig. 1 is the flow chart of known defect inspection method;
Fig. 2 is the calcspar of defective caution system of the present invention;
Fig. 3 is a defects detection data tendency chart, and transverse axis is the time, and the longitudinal axis is a quantity;
Fig. 4 is the defect distribution figure on wafer; And
Fig. 5 is the flow chart of defect inspection method of the present invention.
The figure number explanation
10 production lines, 20 defective scanning movements
40 defective analysis systems of 30 defective statistics systems
50 e-mail systems 60 are responsible for engineer's computer
70 defective scanning patter data bank
80～88 product defects quantity exceed the punctuate of formulating the upper limit
Embodiment
The present invention is further described in more detail below in conjunction with drawings and the specific embodiments.
Fig. 2 is the calcspar that defective caution system according to the present invention is applied in the testing product defective, many production lines 10 are sent to predetermined defective scanning movement 20 according to different products, the situation that each defective scanning movement 20 scanning product defects distributes produces a defects detection data and a defect map, the quantity of defective and make a defects detection data tendency chart in the defective statistics system 30 statistical shortcomings detection data according to the priority of product processing procedure time, whether as shown in Figure 3, the engineer can monitor according to this figure has the anomalous event that causes because of defects count is too high to take place on the line.In the set time, defective analysis system 40 analyzing defects detect the data tendency chart, as continuous more than two batches or two batches product of the same type exceed going up in limited time of formulation, for example, in the 3rd figure shown in the punctuate 82,84,86 and 88, the data that defective analysis system 40 makes record exceed the upper limit of formulation is continuously warned form to be made into defective, shown in following table one.
Pr01 and Pr02 are two kinds of products that the continuous defect anomalous event is arranged of expression in the table one, and wherein Pr01 is to be example with Fig. 3, and form can find out that product P r01 has four unit of cargos to exceed the upper limit continuously in the defects count of defective scanning movement 863357 thus.E-mail system 50 is warned defective form automatically at preset time and is sent to responsible engineer's computer 60 to notify the engineer; the engineer receives that the defect map of utilizing known defect incident in the defective scanning patter data bank 70 behind the form compares whether the defect map of this continuously unusual defective is known event to differentiate; if the relevant board director of known then notice shuts down; if the unknown is then notified operator preferentially continuous anomalous event to be carried out defective and is checked; the engineer shuts down according to the relevant board director of the result notification of checking again; and with this continuously unusual defect distribution figure and check the result and be stored to defective scanning patter data bank 70, but so that fast processing and need not wait until that defective is checked and finish when again similar events as taking place.This system can allow the engineer finish defective at operator and learn continuously unusual defective incident and priority treatment before checking ahead of time, so reduce because of continuously unusually defective cause the loss of product yield.
Table one
Defect Scan Alarm REPORT
＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝ id Msc9046
| ROUTE | PROD | OPER | LOT_ID | Iteml | Iteml | Iteml |
| Route01 Route01 Route01 Route01 | Pr01 Pr01 Pr01 Pr01 | 863357 863357 863357 863357 | lot00001 lot00002 lot00003 lot00004 | DD_WEB DD_WEB DD_WEB DD_WEB | ||
| Route02 Route02 | Pr02 Pr02 | 860597 860597 | lot00005 lot00006 | DC_ML1 DC_ML1 | DD_ML1 DD_ML1 |
Result of Data(Today 12:00 To Today 17:00)
Order by 1)Route_ID 2)Oper_No 3)Reprt_Date 4)Reprt_Time
Fig. 3 is the defects detection data of the same defective scanning movement 20 of defective statistics system 30 statistics and according to the defects detection data tendency chart of the time order and function made of product processing procedure, as shown in the figure, it has the defects count of five batches of products 80,82,84,86 and 88 to exceed the qualification upper limit, 80 is single batch of incident, so defective caution form is record not, and 82,84,86 and 88 be continuous incident, thus the data of these four batches of continuous incidents of defective caution accounting logging, as shown in Table 1.
Fig. 4 is the defect distribution figure of being detected when wafer being done defective scanning, and wherein many stains are the part of defective.The distribution pattern of defective has feature, can represent that a certain particular event causes, and the engineer can judge it is which kind of anomalous event according to this defect map.
Fig. 5 is the flow chart according to defect inspection method of the present invention, and at first step S10 does defective scanning to product, and then whether the defects count of step S11 statistics product exceeds the upper limit, if yes, then carries out step S16, ignores otherwise carry out step S12.Step S16 promptly judge whether have continuous more than two batches or two batches the defects count of product of the same type exceed the upper limit, if yes, then carry out step S17, otherwise carry out step S13.The product that by operator defects count is exceeded the upper limit at step S13 is done defective and is checked, and then will check result notification at step S14 operator is responsible for the engineer, shuts down according to the relevant board director of the result notification of checking step S15 engineer then.Step S17 utilizes Email to send to responsible engineer automatically the product information of continuous defect incident, then judges by the engineer whether continuous anomalous event is known event at step S18, if not, then carry out step S19, otherwise carry out step S15.Preferentially continuous anomalous event is done defective at step S19 by operator and check, then carry out step S14 to S15 more in regular turn.Early a step learns whether have continuous anomalous event to take place to utilize this method; can preferentially do defective by operator for the anomalous event that can't judge reason checks; if known event then can directly notify relevant board director to shut down, so can reduce the loss of wafer yield.
Claims (4)
1, a kind of continuously unusual quick caution system of defective comprises:
One defective scanning movement is to scan the product defects distribution scenario and to produce a defects detection data and make a defect map;
One defective statistics system adding up the defects count of this defects detection data, and makes a defects detection data tendency chart according to the priority of time;
One defective analysis system at this defects detection data tendency chart of set time inner analysis, and writes down the data that continuous two batches or above product defects quantity exceed the upper limit, to make defective caution form;
One e-mail system sees through the networking at preset time and transmits this defective caution form to being responsible for the engineer; And
One defective scanning patter data bank is stored many known event defect map, in order to compare the defect map of this defective scanning movement made.
2, a kind of continuously unusual quick alarming method for power of defective comprises the following steps:
The defective of scanning like product is also made a defect map and a defects detection data;
Add up the defects count of this defects detection data and make a defects detection data tendency chart according to time order and function;
Analyze this defects detection data tendency chart, write down continuous two batches or two batches of above defects count and exceed the data of formulating the upper limit, and according to this data make one defective caution form; And
Use an e-mail system to send this defective caution form automatically to responsible engineer;
This method more comprises uses a defective scanning patter data bank to compare the step of this defect map.
3, the continuously unusual quick alarming method for power of defective as claimed in claim 2 is characterized in that, comprise when the relevant board director shutdown of notice during for known event of this defect map.
4, the continuously unusual quick alarming method for power of defective as claimed in claim 2 is characterized in that, comprises that more notifying operator to do defective when the non-known event of this defect map checks, and shuts down according to checking the relevant board director of result notification again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 02148965 CN1254846C (en) | 2002-11-14 | 2002-11-14 | Quick warning system and method for continuous abnormal faults |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 02148965 CN1254846C (en) | 2002-11-14 | 2002-11-14 | Quick warning system and method for continuous abnormal faults |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1501434A CN1501434A (en) | 2004-06-02 |
| CN1254846C true CN1254846C (en) | 2006-05-03 |
Family
ID=34233421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 02148965 Expired - Fee Related CN1254846C (en) | 2002-11-14 | 2002-11-14 | Quick warning system and method for continuous abnormal faults |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1254846C (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103823408B (en) * | 2012-11-16 | 2016-12-21 | 无锡华润上华科技有限公司 | Semiconductor equipment board quality control method and system |
| CN103559566B (en) * | 2013-06-04 | 2017-07-07 | 上海华力微电子有限公司 | A kind of work dispatching method of Defect Scanning output control process work bench |
| CN108986422A (en) * | 2018-07-25 | 2018-12-11 | 惠科股份有限公司 | Real-time alert news notifies system and method |
| CN109406533A (en) * | 2018-10-25 | 2019-03-01 | 北京阿丘机器人科技有限公司 | A kind of detection system and method for surface defects of products |
| CN112232012B (en) * | 2019-06-27 | 2022-04-26 | 长鑫存储技术有限公司 | Semiconductor process analysis system, semiconductor process analysis method, and computer-readable storage medium |
| CN110907135A (en) * | 2019-11-14 | 2020-03-24 | 深圳市华星光电半导体显示技术有限公司 | Control method and device of manufacturing equipment |
-
2002
- 2002-11-14 CN CN 02148965 patent/CN1254846C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1501434A (en) | 2004-06-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102324677B1 (en) | Dynamic care area for defect detection | |
| US8229584B2 (en) | Abnormality detection system and method of detecting abnormality | |
| US7386418B2 (en) | Yield analysis method | |
| KR101183133B1 (en) | System and method for haze control in semiconductor processes | |
| US20030060916A1 (en) | Automatic production quality control method and system | |
| US6701204B1 (en) | System and method for finding defective tools in a semiconductor fabrication facility | |
| JP2002149222A (en) | Managing quality method and system for production line for product | |
| US20080204739A1 (en) | Process Excursion Detection | |
| CN112801497B (en) | Abnormality detection method and device | |
| CN1254846C (en) | Quick warning system and method for continuous abnormal faults | |
| JP3888938B2 (en) | Chip quality judgment method, chip quality judgment program, marking mechanism using the same, and wafer abnormality analysis method | |
| CN116777433A (en) | Industrial production line equipment operation and maintenance management system based on data analysis | |
| CN116177146A (en) | Conveying belt detection system and method based on AI and laser monitoring technology | |
| CN102637617B (en) | Wafer quality detecting system and wafer quality detection method | |
| CN115206823A (en) | Method and system for monitoring product interval influenced by machine abnormality | |
| US20200011736A1 (en) | Method for detecting variation value | |
| US20040133294A1 (en) | Defect alarm system and method | |
| CN111598491B (en) | Data monitoring method applied to AOI detection and electronic equipment | |
| US6792386B2 (en) | Method and system for statistical comparison of a plurality of testers | |
| JPH09191032A (en) | Monitoring of process abnormality and apparatus thereof | |
| CN102709206B (en) | Automatic defect scanning and casual inspecting method and automatic defect scanning and casual inspecting device for controlling abnormality during silicon wafer production | |
| US20220026307A1 (en) | Reticle defect inspection method and system | |
| CN113837983B (en) | Wafer defect analysis method, system, equipment and medium | |
| TWI220288B (en) | Method of defect control | |
| JPH11176899A (en) | Method and system for alarming defect |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20060503 Termination date: 20191114 |