CN114429311B - Dynamic monitoring method and system for semiconductor manufacturing process - Google Patents
Dynamic monitoring method and system for semiconductor manufacturing process Download PDFInfo
- Publication number
- CN114429311B CN114429311B CN202210143870.9A CN202210143870A CN114429311B CN 114429311 B CN114429311 B CN 114429311B CN 202210143870 A CN202210143870 A CN 202210143870A CN 114429311 B CN114429311 B CN 114429311B
- Authority
- CN
- China
- Prior art keywords
- manufacturing
- abnormal
- monitoring
- manufacturing process
- determining
- 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.)
- Active
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 304
- 238000012544 monitoring process Methods 0.000 title claims abstract description 263
- 239000004065 semiconductor Substances 0.000 title claims abstract description 110
- 238000000034 method Methods 0.000 title claims abstract description 109
- 230000002159 abnormal effect Effects 0.000 claims abstract description 183
- 238000004458 analytical method Methods 0.000 claims abstract description 23
- 238000007405 data analysis Methods 0.000 claims abstract description 9
- 230000008569 process Effects 0.000 claims description 47
- 239000011159 matrix material Substances 0.000 claims description 31
- 206010000117 Abnormal behaviour Diseases 0.000 claims description 19
- 238000009826 distribution Methods 0.000 claims description 18
- 238000002372 labelling Methods 0.000 claims description 13
- 230000005856 abnormality Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 description 8
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 239000000047 product Substances 0.000 description 3
- 239000013078 crystal Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000007924 injection Substances 0.000 description 2
- 238000002347 injection Methods 0.000 description 2
- 239000007769 metal material Substances 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 230000008021 deposition Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000005468 ion implantation Methods 0.000 description 1
- 238000001459 lithography Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 229920002120 photoresistant polymer Polymers 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 239000011265 semifinished product Substances 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- General Factory Administration (AREA)
Abstract
The invention provides a dynamic monitoring method and a system for a semiconductor manufacturing process, wherein the method comprises the following steps: determining a semiconductor to be manufactured, calling the manufacturing process of the semiconductor to be manufactured from a manufacturing process database, and acquiring manufacturing indexes in the manufacturing process; performing type analysis on historical manufacturing data related to a semiconductor to be manufactured according to the manufacturing indexes to obtain a historical data set of each manufacturing index, and determining manufacturing abnormal events corresponding to the manufacturing indexes on the basis of a data analysis model; determining a through flow line corresponding to the abnormal index in the manufacturing flow, determining an abnormal thread section in the through flow line based on the manufacturing abnormal event, and constructing an abnormal monitoring set of the abnormal thread section; and constructing a dynamic monitoring set of the semiconductor to be manufactured by the abnormal monitoring set and the normal monitoring set, thereby realizing the dynamic monitoring of the semiconductor to be manufactured in the manufacturing process. And determining an abnormal thread section through abnormal analysis, and setting a monitoring set to realize effective dynamic monitoring.
Description
Technical Field
The present invention relates to the field of dynamic monitoring technologies, and in particular, to a dynamic monitoring method and system for a semiconductor manufacturing process.
Background
With the wide application of semiconductors in various industries, the quality requirements of semiconductors are higher and higher, and the conventional semiconductor manufacturing is generally completed by manufacturing semiconductors, however, in the semiconductor manufacturing process, the semiconductor manufacturing is generally performed only according to a set program code, and only after the semiconductor manufacturing is finally manufactured, whether the semiconductor manufacturing is complete or not can be known, which undoubtedly results in the reduction of product quality and the reduction of production efficiency;
or when the product is half manufactured, the semiconductor is manually controlled to stop, and the semi-finished product is manually checked, so that the checking workload is increased.
Therefore, the present invention provides a dynamic monitoring method and system for semiconductor manufacturing process.
Disclosure of Invention
The invention provides a dynamic monitoring method for a semiconductor manufacturing process, which is used for determining an abnormal thread section through abnormal analysis and setting a monitoring set to ensure the timeliness and pertinence of monitoring, thereby realizing effective dynamic monitoring, improving the product quality and improving the manufacturing efficiency.
The invention provides a dynamic monitoring method for a semiconductor manufacturing process, which comprises the following steps:
step 1: determining a semiconductor to be manufactured, calling a manufacturing process of the semiconductor to be manufactured from a manufacturing process database, and acquiring manufacturing indexes in the manufacturing process;
step 2: performing type analysis on historical manufacturing data related to the semiconductor to be manufactured according to the manufacturing indexes to obtain a historical data set of each manufacturing index, determining whether definable abnormal indexes exist or not based on a data analysis model, and if yes, determining manufacturing abnormal events corresponding to the abnormal indexes;
and step 3: determining a through flow line of a corresponding abnormal index in the manufacturing flow, determining an abnormal thread section in the through flow line based on the manufacturing abnormal event, and constructing an abnormal monitoring set of the abnormal thread section;
and 4, step 4: acquiring a normal monitoring set of the manufacturing process based on the manufacturing process database;
and 5: and constructing the abnormal monitoring set and the normal monitoring set into a dynamic monitoring set of the semiconductor to be manufactured, and realizing dynamic monitoring of the semiconductor to be manufactured in the manufacturing process according to the dynamic monitoring set.
In one possible implementation manner, determining a semiconductor to be manufactured, retrieving a manufacturing process of the semiconductor to be manufactured from a manufacturing process database, and acquiring a manufacturing index in the manufacturing process, includes:
acquiring a manufacturing instruction input by a target user, and matching a semiconductor to be manufactured based on an instruction semiconductor database;
acquiring a unique identifier of the semiconductor to be manufactured, and acquiring a manufacturing process matched with the unique identifier from the manufacturing process database;
and extracting the manufacturing index from the manufacturing flow.
In one possible implementation manner, performing type analysis on the historical manufacturing data related to the semiconductor to be manufactured according to the manufacturing index to obtain a historical data set of each manufacturing index includes:
acquiring an index parameter of each manufacturing index, and taking the index parameter as a specified object;
and traversing the historical manufacturing data, acquiring the historical data of each specified object in each manufacturing index, and constructing to obtain a historical data set of each manufacturing index.
In one possible implementation, determining a manufacturing anomaly event corresponding to the anomaly indicator includes:
constructing the same parameter vector of the same parameter in the manufacturing process of different batches based on the historical data set of the corresponding abnormal index, and further constructing a parameter matrix F of the corresponding abnormal index;
wherein A is 11 Representing the co-parameter vector of the 1 st parameter in the abnormal index in the 1 st batch manufacturing process; a. The 1n1 Representing the co-parameter vector of the 1 st parameter in the abnormal index in the manufacturing process of the n1 th batch; a. The n21 Representing the co-parameter vector of the n2 th parameter in the abnormal index in the 1 st batch manufacturing process; a. The n2n1 Representing the co-parameter vector of the n2 th parameter in the abnormal index in the manufacturing process of the n1 th batch; n1 represents a total parameter corresponding to the abnormality index; n2 represents the total batch;
respectively carrying out subtraction processing on the parameter matrix F and the standard matrix B to obtain a difference matrix C;
respectively inputting the column elements in the difference matrix C into an element anomaly analysis model to obtain first anomaly information of different parameters in the same batch, and setting a first anomaly label;
respectively inputting the row elements in the difference matrix C into an element anomaly analysis model to obtain second anomaly information of the same parameter in different batches, and setting a second anomaly label;
acquiring a first sequence of the first abnormal information, determining abnormal dispersed distribution, and acquiring a second sequence of the second abnormal information, determining abnormal concentrated distribution;
and determining the satisfied abnormal event conditions based on the label setting result and the distribution result, and further obtaining the manufacturing abnormal event corresponding to the manufacturing index.
In one possible implementation, determining a through-flow line of a corresponding anomaly indicator in the manufacturing flow includes:
traversing in a standard index set corresponding to each manufacturing process based on the abnormal indexes, and screening to obtain a first sub-process with consistent indexes;
determining the process influence distribution of the abnormal indexes based on a preset process rule to obtain a second sub-process;
performing first labeling on overlapped sub-processes in the first sub-process and the second sub-process, performing second labeling on the remaining single sub-processes of the first sub-process, and performing third labeling on the remaining single sub-processes of the second sub-process;
and obtaining a through flow line based on the labeling result.
In one possible implementation, determining an exception thread segment in the through-flow line based on the manufacturing exception event, and constructing an exception monitoring set of the exception thread segment includes:
acquiring an abnormal behavior process of the abnormal manufacturing event based on a through flow line, and determining an initial point and a termination point of the abnormal behavior process;
determining a first sub-item included in the abnormal behavior process;
determining complete sub-items and incomplete sub-items in the first sub-items, and meanwhile, obtaining a project which needs to be monitored and a project which can be monitored according to the manufacturing importance of the complete sub-items based on the whole manufacturing process and the manufacturing importance of the incomplete sub-items based on the whole manufacturing process;
wherein, Y i Representing the manufacturing importance of the ith complete sub-item; z j Representing the manufacturing importance of the jth incomplete sub-item; s. the i Complete item information representing the ith complete sub-item; y represents flow information of the manufacturing flow; t is i Representing a complete manufacturing time of the ith complete sub-item; t is zong A flow manufacturing time representing a manufacturing flow;information representing the ith complete sub-item-time adjustment coefficient; z is a linear or branched member j Item information currently contained representing a jth incomplete sub-item; t is a unit of j Represents the manufacturing time currently contained for the jth incomplete sub-item; t is J Representing a complete manufacturing time for the jth incomplete sub-item; />Information representing the jth incomplete sub-item-time adjustment factor;
when importance of production Y i When the number of the corresponding complete sub-items is larger than the first preset value, the corresponding complete sub-items are regarded as the items which need to be monitored;
when importance of manufacture Z j When the number of the corresponding incomplete sub-items is larger than a second preset value, the corresponding incomplete sub-items are regarded as items which need to be monitored;
otherwise, the remaining complete sub-items and the remaining incomplete sub-items are regarded as monitorable items;
determining a first main abnormal sequence of the project which needs to be monitored and is based on the abnormal behavior process, determining a maximum time period corresponding to the first main abnormal sequence, setting the maximum time period as a first thread segment, simultaneously determining a second main abnormal sequence of the project which can be monitored and is based on the abnormal behavior process, determining a middle time point of the second main abnormal sequence, and regarding the second main abnormal sequence as a monitorable point;
constructing an abnormal thread section corresponding to the initial point and the end point, calibrating the first thread section and the monitorable point on the abnormal thread section, and adding a secondary monitoring event;
and obtaining an abnormal monitoring set based on the secondary monitoring sections corresponding to all the secondary monitoring events and the second monitoring points.
In one possible implementation manner, obtaining the normal monitoring set of the manufacturing process based on the manufacturing process database includes:
matching the monitoring points and the monitoring sections of each sub-item in the manufacturing process in a normal operation state based on the manufacturing process database;
and constructing to obtain a normal monitoring set according to all the monitoring points and the monitoring sections.
In a possible implementation manner, according to the dynamic monitoring set, dynamic monitoring of the semiconductor to be manufactured in the manufacturing process is implemented, and the dynamic monitoring includes:
constructing a flow line of a manufacturing flow of the semiconductor to be manufactured;
configuring a secondary monitoring section and a secondary monitoring point in an abnormal monitoring set at a first position on the flow line;
configuring a normal monitoring section and a normal monitoring point in a normal monitoring set at a second position on the flow line;
if the first position is overlapped with the second position, reserving the second monitoring section and the second monitoring point configured on the overlapped position, and deleting the normal monitoring section and the normal monitoring point configured on the overlapped position to obtain a process monitoring line;
setting a corresponding section monitoring trigger window to a first monitoring section of the process monitoring line, and issuing a corresponding first monitoring task according to the to-be-executed project attribute and the to-be-monitored important attribute corresponding to the first monitoring section;
setting a corresponding point trigger response window to a first monitoring point of the process monitoring line, and issuing a corresponding second monitoring task according to the item execution type of the first monitoring point and the point attribute of the first monitoring point;
identifying the monitoring numbers in the first monitoring task and the second monitoring task and the combination characteristics of the monitoring marks, and determining corresponding monitoring modes;
when the manufacturing progress of the semiconductor to be manufactured reaches the entry point of the corresponding exit setting window, calling a corresponding monitoring mode for dynamic monitoring;
wherein the window includes a trigger entry point and a trigger exit point.
The invention provides a dynamic monitoring system for a semiconductor manufacturing process, which comprises:
the first determining module is used for determining a semiconductor to be manufactured, calling the manufacturing process of the semiconductor to be manufactured from a manufacturing process database and acquiring the manufacturing index in the manufacturing process;
the second determining module is used for performing type analysis on historical manufacturing data related to the semiconductor to be manufactured according to the manufacturing indexes to obtain a historical data set of each manufacturing index, determining whether definable abnormal indexes exist or not based on the data analysis model, and if yes, determining manufacturing abnormal events corresponding to the abnormal indexes;
a third determining module, configured to determine a through flow line in the manufacturing flow corresponding to the abnormal indicator, determine an abnormal thread segment in the through flow line based on the manufacturing abnormal event, and construct an abnormal monitoring set of the abnormal thread segment;
the acquisition module is used for acquiring a normal monitoring set of the manufacturing process based on the manufacturing process database;
and the dynamic monitoring module is used for constructing the dynamic monitoring set of the semiconductor to be manufactured by the abnormal monitoring set and the normal monitoring set and realizing dynamic monitoring of the semiconductor to be manufactured in the manufacturing process according to the dynamic monitoring set.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a flow chart of a method for dynamic monitoring of a semiconductor manufacturing process in accordance with an embodiment of the present invention;
FIG. 2 is a block diagram of a dynamic monitoring system for a semiconductor manufacturing process in accordance with an embodiment of the present invention;
fig. 3 is a structural diagram of a maximum time period in the embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
The present invention provides a dynamic monitoring method for semiconductor manufacturing process, as shown in fig. 1, comprising:
step 1: determining a semiconductor to be manufactured, calling a manufacturing process of the semiconductor to be manufactured from a manufacturing process database, and acquiring manufacturing indexes in the manufacturing process;
step 2: performing type analysis on historical manufacturing data related to the semiconductor to be manufactured according to the manufacturing indexes to obtain a historical data set of each manufacturing index, determining whether definable abnormal indexes exist or not based on a data analysis model, and if yes, determining manufacturing abnormal events corresponding to the abnormal indexes;
and step 3: determining a through flow line of a corresponding abnormal index in the manufacturing flow, determining an abnormal thread section in the through flow line based on the manufacturing abnormal event, and constructing an abnormal monitoring set of the abnormal thread section;
and 4, step 4: acquiring a normal monitoring set of the manufacturing process based on the manufacturing process database;
and 5: and constructing a dynamic monitoring set of the semiconductor to be manufactured by using the abnormal monitoring set and the normal monitoring set, and realizing dynamic monitoring of the semiconductor to be manufactured in the manufacturing process according to the dynamic monitoring set.
In this embodiment, the semiconductor to be manufactured is, for example, a semiconductor to be manufactured, and the semiconductor to be manufactured includes a manufacturing process, for example, including: deposition, photoresist coating, exposure, computational lithography, baking and development, etching, metrology and inspection, ion implantation, packaging, and the like.
Since the manufacturing processes or steps of different semiconductors may be different, the manufacturing process of the semiconductor to be manufactured is obtained, and the manufacturing process includes many operations (sub-processes or items), the manufacturing indexes are directly obtained by a preset process, and the process and the indexes are already available, and the semiconductor to be manufactured is an existing semiconductor and is used for monitoring the manufacturing process of the existing semiconductor.
In this embodiment, a historical data set of each manufacturing index is obtained, for example, the manufacturing index 1 is obtained, data of one year in history is analyzed to determine a cause of an abnormality, at this time, it may be regarded as an abnormal event, and a flow time period caused by the cause is determined, that is, an abnormal thread segment.
In this embodiment, an analysis of the historical data set is used to determine whether the indicator can be defined as an abnormal indicator, and if so, a through flow line is obtained.
In this embodiment, for example: the manufacturing process includes items 1, 2, 3, and 4 in execution order, where the anomaly index is 01, and appears in items 1 and 2, where items 1 and 2 are considered as a through flow line, and if they appear in items 1 and 3, where items 1 and 3 are considered as a through flow line.
In this embodiment, the normal monitoring set is a monitoring point that has been set for each sub-process in advance according to the process attribute of the manufacturing process, which is to effectively monitor the manufacturing process, ensure effective manufacturing of the semiconductor, and improve the quality effect.
In this embodiment, the dynamic monitoring set is a set including abnormal monitoring and normal monitoring.
In this embodiment, the manufacturing process is monitored based on the monitoring time points of the dynamic monitoring set, that is, each monitoring point is bound to a corresponding manufacturing step, that is, when the sub-process is reached, monitoring is performed, so that monitoring loss is reduced, and monitoring integrity of abnormal points is improved.
For example, a monitoring point b2 is provided at the manufacturing point b1 in the item a, for example, the manufacturing is started from the manufacturing point b1 within 0.2 second, and at this time, the monitoring point b2 is triggered to monitor the manufacturing process of the manufacturing point b1, that is, to ensure timely monitoring.
In this embodiment, the data analysis model is obtained by training a sample of the indexes corresponding to the various processes, different historical data corresponding to the indexes, and the definition result of whether the data analysis model can be defined as an abnormal index.
The beneficial effects of the above technical scheme are: through exception analysis, an exception thread section is determined, and a monitoring set is set, so that the timeliness and pertinence of monitoring are guaranteed, effective dynamic monitoring is further realized, the quality of a produced semiconductor is improved, and the manufacturing efficiency is improved.
The invention provides a dynamic monitoring method for a semiconductor manufacturing process, which is used for determining a semiconductor to be manufactured, calling the manufacturing process of the semiconductor to be manufactured from a manufacturing process database and acquiring manufacturing indexes in the manufacturing process, and comprises the following steps:
acquiring a manufacturing instruction input by a target user, and matching a semiconductor to be manufactured based on an instruction semiconductor database;
acquiring a unique identifier of the semiconductor to be manufactured, and acquiring a manufacturing process matched with the unique identifier from the manufacturing process database;
and extracting the manufacturing index from the manufacturing flow.
In this embodiment, the manufacturing instruction, for example, is to manufacture the semiconductor 1, at this time, the manufacturing semiconductor 1 is determined, the unique identifier of the semiconductor is preset, and the instruction semiconductor database includes: commands, semiconductor information.
In this embodiment, the manufacturing process database includes the manufacturing process and the indexes of the corresponding items, which are preset.
The beneficial effects of the above technical scheme are: the semiconductor is automatically matched based on the instruction, so that the flow and the index are automatically matched, and an effective basis is provided for the subsequent judgment of whether the index is normal or not.
The invention provides a dynamic monitoring method for a semiconductor manufacturing process, which determines a manufacturing abnormal event corresponding to an abnormal index, and comprises the following steps:
constructing the same parameter vector of the same parameter in the manufacturing process of different batches based on the historical data set of the corresponding abnormal index, and further constructing a parameter matrix F of the corresponding abnormal index;
wherein A is 11 Representing the co-parameter vector of the 1 st parameter in the abnormal index in the 1 st batch manufacturing process; a. The 1n1 Representing the co-parameter vector of the 1 st parameter in the abnormal index in the manufacturing process of the n1 th batch; a. The n21 Representing the co-parameter vector of the n2 th parameter in the abnormal index in the 1 st batch manufacturing process; a. The n2n1 Representing the co-parameter vector of the n2 th parameter in the abnormal index in the manufacturing process of the n1 th batch; n1 represents a total parameter corresponding to the abnormality index; n2 represents the total batch;
respectively carrying out subtraction processing on the parameter matrix F and the standard matrix B to obtain a difference matrix C;
respectively inputting the column elements in the difference matrix C into an element anomaly analysis model to obtain first anomaly information of different parameters in the same batch, and setting a first anomaly label;
respectively inputting the row elements in the difference matrix C into an element anomaly analysis model to obtain second anomaly information of the same parameter in different batches, and setting a second anomaly label;
acquiring a first sequence of the first abnormal information, determining abnormal dispersed distribution, and acquiring a second sequence of the second abnormal information, determining abnormal concentrated distribution;
and determining the satisfied abnormal event conditions based on the label setting result and the distribution result, and further obtaining the manufacturing abnormal event corresponding to the manufacturing index.
In this embodiment, the historical data set may be data of the same parameter in different batches of manufacturing processes, for example, after the index is defined as an abnormal index, all data of the abnormal index is obtained from the historical manufacturing data, for example, the historical manufacturing data includes data a1 of a first batch related to the index, data a2 of a second batch related to the index, data a3 of a third batch related to the index, and the like, at this time, a1, a2, a3, and the like may be regarded as the historical data set of the abnormal index, and further, the parameter sets of different batches may be constructed according to these parameters.
In this embodiment, the difference matrix may be obtained by comparing the standard matrix B with the parameter matrix F, which is obtained when there is no abnormality in the manufacturing process of each lot.
In this embodiment, the element anomaly analysis model is obtained by training using the difference parameters as elements and anomaly information corresponding to different difference parameter combinations as samples.
In this embodiment, the first anomaly information is for anomalies occurring in the same batch, the second anomaly information is for anomalies occurring in the same parameter, and the first anomaly information and the second anomaly information are for different anomalies, so that a more accurate anomaly event can be obtained by combining the two modes.
In this embodiment, the first abnormal label and the second abnormal label are provided for better discrimination.
In this embodiment, the first sequence is for the same batch, so that the distributed distribution is obtained, and the abnormal factors can be obtained in a wider range, and the second sequence is for the same parameter, so that the targeted abnormal factors can be obtained conveniently, and the two sequences are collected to obtain more comprehensive factors.
In this embodiment, for example, factor 1, factor 2, tag 1, and tag 2 are obtained, and at this time, the conditions that factor 1, factor 2, tag 1, and tag 2 satisfy are obtained, and the conditions are obtained based on a condition database (including conditions corresponding to various tag setting results and distribution results and abnormal events).
In this embodiment, for example, the abnormal event is due to the absence of metal material in the semiconductor, which affects the fabrication of the semiconductor.
The beneficial effects of the above technical scheme are: the method comprises the steps of obtaining a historical data set, constructing a parameter matrix, comparing the parameter matrix with a standard to obtain a difference matrix, analyzing rows and columns of the difference matrix respectively to obtain different abnormal information, determining abnormal events according to abnormal distribution, improving the rationality of subsequent determination and monitoring sets, and indirectly ensuring the reliability of semiconductor manufacturing.
The invention provides a dynamic monitoring method for a semiconductor manufacturing process, which determines a through process line corresponding to an abnormal index in the manufacturing process, and comprises the following steps:
traversing the standard index set corresponding to each manufacturing process based on the abnormal index, and screening to obtain a first sub-process with consistent indexes;
determining the process influence distribution of the abnormal indexes based on a preset process rule to obtain a second sub-process;
performing first labeling on overlapped sub-processes in the first sub-process and the second sub-process, performing second labeling on the remaining single sub-processes of the first sub-process, and performing third labeling on the remaining single sub-processes of the second sub-process;
and obtaining a through flow line based on the labeling result.
In this embodiment, for example: the first sub-process comprises: 1. 2, 4; the second sub-process comprises: 1. 3, 5; the overlap sub-process is: 1, the second noted sub-process is 2, 4, and the third noted sub-process is 3 and 5.
The preset flow rule is preset to determine the influence of different indexes on other sub-flows except the sub-flow containing the index itself.
The beneficial effects of the above technical scheme are: the overlapped flow and the independent flow are determined by acquiring the flows in two modes, so that flow lines with different labeling results are obtained, and subsequent reasonable analysis on the abnormity is ensured.
The invention provides a dynamic monitoring method for a semiconductor manufacturing process, which determines an abnormal thread segment in a through process line based on a manufacturing abnormal event and constructs an abnormal monitoring set of the abnormal thread segment, and comprises the following steps:
acquiring an abnormal behavior process of the abnormal manufacturing event based on a through flow line, and determining an initial point and a termination point of the abnormal behavior process;
determining a first sub-item included in the abnormal behavior process;
determining complete sub-items and incomplete sub-items in the first sub-items, and meanwhile, obtaining a project which needs to be monitored and a project which can be monitored according to the manufacturing importance of the complete sub-items based on the whole manufacturing process and the manufacturing importance of the incomplete sub-items based on the whole manufacturing process;
wherein Y is i Representing the manufacturing importance of the ith complete sub-item; z is a linear or branched member j Representing the manufacturing importance of the jth incomplete sub-item; s. the i Complete item information representing the ith complete sub-item; y represents process information of the manufacturing process; t is i Representing a complete manufacturing time of the ith complete sub-item; t is zong A flow manufacturing time representing a manufacturing flow;information-time adjustment coefficients representing the ith complete sub-item; z is a linear or branched member j Item information currently contained representing a jth incomplete sub-item; t is j Represents the currently involved manufacturing time of the jth incomplete sub-item; t is J Representing a complete manufacturing time of a jth incomplete sub-item; />Information representing the jth incomplete sub-item-time adjustment factor;
when importance of production Y i When the number of the corresponding complete sub-items is larger than the first preset value, the corresponding complete sub-items are regarded as the items which need to be monitored;
when importance of manufacture Z j When the value is larger than the second preset value, the corresponding incomplete sub-project is regarded as a project which needs to be monitored;
otherwise, the remaining complete sub-items and the remaining incomplete sub-items are regarded as monitorable items;
determining a first main abnormal sequence of the project which needs to be monitored and is based on the abnormal behavior process, determining a maximum time period corresponding to the first main abnormal sequence, setting the maximum time period as a first thread segment, simultaneously determining a second main abnormal sequence of the project which can be monitored and is based on the abnormal behavior process, determining a middle time point of each second main abnormal sequence, and regarding the second main abnormal sequence as a monitorable point;
constructing an abnormal thread section corresponding to the initial point and the termination point, calibrating the first thread section and the monitorable point on the abnormal thread section, and adding a secondary monitoring event;
and obtaining an abnormal monitoring set based on the secondary monitoring sections corresponding to all the secondary monitoring events and the second monitoring points.
In this embodiment, for example, sub-flow 1 and sub-flow 2 are included in the through-flow line, and sub-items 1 and 2 in sub-flow 1 and sub-items 13 and 12 in sub-flow 2 constitute an abnormal behavior process, and therefore, the initial point and the end point of the process can be determined. And the first sub-item may be a sub-item including: sub-items 1, 2, sub-items 13, 12, and the order may also be the execution order of the abnormal behavior process;
and in the process of executing each sub-item, the incomplete sub-items which may be the sub-item 1 and the sub-item 12 exist, that is, the first time point and the last time point of the problem occurring in the process are determined, if the sub-item 1 corresponding to the first time point is executed for a long time, the execution is the incomplete sub-item, and the complete judgment on the sub-item 12 is similar to the above.
In this embodiment, it is assumed that the execution times of the sub-items 1, 2 and 13, 12 are the same, but the time for which the sub-item 1 and the sub-item 12 remain in the abnormal behavior process is only half, and at this time, the sub-item 1 and the sub-item 12 are regarded as incomplete sub-items.
In this embodiment, the first preset value and the second preset value are preset.
In this embodiment, the monitoring item is, for example, a sub-item 1, at this time, an abnormal sequence of the sub-item 1 is determined, and main abnormal sequences d1, d2, and d3 are determined from the abnormal sequence to determine the time period, as shown in fig. 3, d1, d2, and d3 are located on a sequence p corresponding to the sub-item 1, and the main abnormal sequences d1, d2, and d3 have corresponding time periods, p1, p2, and p3, respectively, at this time, the obtained p0 is the maximum time period.
In this embodiment, sub-item 2 may be monitored, for example, the second main sequence is u1, and the middle time point of the corresponding time period of the sequence is obtained.
In this embodiment, for example: the finally obtained abnormal thread segments are r 1-r 2, and the thread segments are provided with first thread segments r 11-r 12 which can monitor points r13 and r14 and can be used as a monitoring event corresponding to an abnormal manufacturing event.
In this embodiment, all the abnormal manufacturing events have a corresponding monitoring event, and a monitoring set is obtained.
The beneficial effects of the above technical scheme are: the setting of the monitoring time of the sub-items with different properties is determined by analyzing the sub-items contained in the abnormal behavior process, so that the monitoring events of different abnormal events are obtained, the effectiveness and the reliability of a monitoring set are guaranteed, and the production efficiency is indirectly guaranteed.
The invention provides a dynamic monitoring method for a semiconductor manufacturing process, which is used for acquiring a normal monitoring set of the manufacturing process based on a manufacturing process database and comprises the following steps:
matching the monitoring points and the monitoring sections of each sub-item in the manufacturing process in a normal operation state based on the manufacturing process database;
and constructing to obtain a normal monitoring set according to all the monitoring points and the monitoring sections.
The beneficial effects of the above technical scheme are: by determining the normal monitoring set, a basis is provided for the subsequent monitoring set conveniently.
The invention provides a dynamic monitoring method for a semiconductor manufacturing process, which realizes dynamic monitoring of a semiconductor to be manufactured in a manufacturing process according to a dynamic monitoring set, and comprises the following steps:
constructing a flow line of a manufacturing flow of the semiconductor to be manufactured;
configuring a secondary monitoring section and a secondary monitoring point in an abnormal monitoring set at a first position on the flow line;
configuring a normal monitoring section and a normal monitoring point in a normal monitoring set at a second position on the flow line;
if the first position is overlapped with the second position, the second monitoring section and the second monitoring point configured on the overlapped position are reserved, and the normal monitoring section and the normal monitoring point configured on the overlapped position are deleted to obtain a process monitoring line;
setting a corresponding section monitoring trigger window to a first monitoring section of the process monitoring line, and issuing a corresponding first monitoring task according to the to-be-executed item attribute and the to-be-monitored important attribute corresponding to the first monitoring section;
setting a corresponding point trigger response window to a first monitoring point of the process monitoring line, and issuing a corresponding second monitoring task according to the project execution type of the first monitoring point and the point attribute of the first monitoring point;
identifying the monitoring numbers and the combination characteristics of the monitoring marks in the first monitoring task and the second monitoring task, and determining corresponding monitoring modes;
when the manufacturing progress of the semiconductor to be manufactured reaches the entry point of the corresponding exit setting window, calling a corresponding monitoring mode to perform dynamic monitoring;
wherein the window includes a trigger entry point and a trigger exit point.
In this embodiment, all sub-processes of the manufacturing process are configured, and the process lines of all sub-processes are acquired, and each sub-process is executed one by one, and there is no parallel execution.
In this embodiment, the first monitoring segment may have both the secondary monitoring segment and the normal monitoring segment, and the first monitoring point may have both the secondary monitoring point and the normal monitoring point.
In this embodiment, the setting of the window is to set a trigger condition for the monitoring segment or the monitoring point, so that the monitoring can be performed in time when the point or the segment is reached.
And each first monitoring section and each first monitoring point are matched with the corresponding execution operation behaviors one by one.
In this embodiment, the to-be-executed item attribute refers to an item attribute of a sub item in a sub flow involved in the monitoring segment, such as injection crystal, and the like, and the to-be-monitored important attribute is related to the flow execution importance and the abnormal frequency of the monitoring segment.
Item execution types such as: the types of crystal injection, metal material setting, pin setting, etc., and the point attribute refers to an item attribute of the corresponding sub item.
In this embodiment, the first monitoring task and the second monitoring task refer to different monitoring modes and are obtained by combining feature recognition.
The beneficial effects of the above technical scheme are: the non-overlapping of monitoring places and the completeness of monitoring layout are ensured by determining the process monitoring line, different monitoring tasks are issued, so that the monitoring can be conveniently carried out according to different modes, the monitoring efficiency is improved, the effective monitoring is ensured, and the efficient manufacturing is realized.
The present invention provides a dynamic monitoring system for semiconductor manufacturing process, as shown in fig. 2, comprising:
the first determining module is used for determining a semiconductor to be manufactured, calling the manufacturing process of the semiconductor to be manufactured from a manufacturing process database and acquiring the manufacturing index in the manufacturing process;
the second determining module is used for performing type analysis on historical manufacturing data related to the semiconductor to be manufactured according to the manufacturing indexes to obtain a historical data set of each manufacturing index, determining whether definable abnormal indexes exist or not based on the data analysis model, and if yes, determining manufacturing abnormal events corresponding to the abnormal indexes;
a third determining module, configured to determine a through flow line in the manufacturing flow corresponding to the abnormal indicator, determine an abnormal thread segment in the through flow line based on the manufacturing abnormal event, and construct an abnormal monitoring set of the abnormal thread segment;
the acquisition module is used for acquiring a normal monitoring set of the manufacturing process based on the manufacturing process database;
and the dynamic monitoring module is used for constructing the dynamic monitoring set of the semiconductor to be manufactured by the abnormal monitoring set and the normal monitoring set and realizing dynamic monitoring of the semiconductor to be manufactured in the manufacturing process according to the dynamic monitoring set.
The beneficial effects of the above technical scheme are: through exception analysis, an exception thread section is determined, and a monitoring set is set, so that timeliness and pertinence of monitoring are guaranteed, effective dynamic monitoring is further achieved, semiconductor manufacturing qualification is improved, and manufacturing efficiency is improved.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (8)
1. A method for dynamic monitoring of a semiconductor manufacturing process, comprising:
step 1: determining a semiconductor to be manufactured, calling a manufacturing process of the semiconductor to be manufactured from a manufacturing process database, and acquiring manufacturing indexes in the manufacturing process;
step 2: analyzing the type of the historical manufacturing data related to the semiconductor to be manufactured according to the manufacturing indexes to obtain a historical data set of each manufacturing index, determining whether definable abnormal indexes exist or not based on a data analysis model, and if yes, determining manufacturing abnormal events corresponding to the abnormal indexes;
and 3, step 3: determining a through flow line of a corresponding abnormal index in the manufacturing flow, determining an abnormal thread section in the through flow line based on the manufacturing abnormal event, and constructing an abnormal monitoring set of the abnormal thread section;
and 4, step 4: acquiring a normal monitoring set of the manufacturing process based on the manufacturing process database;
and 5: constructing the abnormal monitoring set and the normal monitoring set into a dynamic monitoring set of the semiconductor to be manufactured, and realizing dynamic monitoring of the semiconductor to be manufactured in the manufacturing process according to the dynamic monitoring set;
wherein determining the manufacturing exception event corresponding to the exception indicator includes:
constructing the same parameter vector of the same parameter in the manufacturing process of different batches based on the historical data set corresponding to the abnormal index, and further constructing a parameter matrix F corresponding to the abnormal index;
wherein,A 11 Representing the co-parameter vector of the 1 st parameter in the abnormal index in the 1 st batch manufacturing process; a. The 1n1 Representing the co-parameter vector of the 1 st parameter in the abnormal index in the manufacturing process of the n1 st batch; a. The n21 Representing the co-parameter vector of the n2 th parameter in the abnormal index in the 1 st batch manufacturing process; a. The n2n1 Representing the co-parameter vector of the n2 th parameter in the abnormal index in the manufacturing process of the n1 th batch; n2 represents a total parameter corresponding to the abnormality index; n1 represents the total batch;
performing subtraction processing on the parameter matrix F and the standard matrix B to obtain a difference matrix C;
respectively inputting the column elements in the difference matrix C into an element anomaly analysis model to obtain first anomaly information of different parameters in the same batch, and setting a first anomaly label;
respectively inputting the row elements in the difference matrix C into an element anomaly analysis model to obtain second anomaly information of the same parameter in different batches, and setting a second anomaly label;
acquiring a first sequence of the first abnormal information, determining abnormal dispersed distribution, and acquiring a second sequence of the second abnormal information, determining abnormal concentrated distribution;
and determining the satisfied abnormal event conditions based on the label setting result and the distribution result, and further obtaining the manufacturing abnormal event corresponding to the manufacturing index.
2. The dynamic monitoring method of claim 1, wherein determining a semiconductor to be manufactured, retrieving a manufacturing process of the semiconductor to be manufactured from a manufacturing process database, and obtaining a manufacturing index in the manufacturing process comprises:
acquiring a manufacturing instruction input by a target user, and matching a semiconductor to be manufactured based on an instruction semiconductor database;
acquiring a unique identifier of the semiconductor to be manufactured, and acquiring a manufacturing process matched with the unique identifier from the manufacturing process database;
and extracting the manufacturing index from the manufacturing flow.
3. The dynamic monitoring method of claim 1, wherein performing a type analysis on historical manufacturing data associated with the semiconductor to be manufactured according to manufacturing criteria to obtain a historical data set for each manufacturing criteria comprises:
acquiring an index parameter of each manufacturing index, and taking the index parameter as a specified object;
and traversing the historical manufacturing data, acquiring the historical data of each specified object in each manufacturing index, and constructing to obtain a historical data set of each manufacturing index.
4. The dynamic monitoring method of claim 1, wherein determining a through-flow line in the manufacturing flow for a corresponding anomaly indicator comprises:
traversing in a standard index set corresponding to each manufacturing process based on the abnormal indexes, and screening to obtain a first sub-process with consistent indexes;
determining the process influence distribution of the abnormal indexes based on a preset process rule to obtain a second sub-process;
performing first labeling on overlapped sub-processes in the first sub-process and the second sub-process, performing second labeling on the remaining single sub-processes of the first sub-process, and performing third labeling on the remaining single sub-processes of the second sub-process;
and obtaining a through flow line based on the labeling result.
5. The dynamic monitoring method of claim 1, wherein determining an exception thread segment in the through-flow line based on the manufacturing exception event and constructing an exception monitoring set of the exception thread segment comprises:
acquiring an abnormal behavior process of the manufacturing abnormal event based on a through flow line, and determining an initial point and a termination point of the abnormal behavior process;
determining a first sub-item included in the abnormal behavior process;
determining complete sub-items and incomplete sub-items in the first sub-items, and meanwhile, obtaining a project which needs to be monitored and a project which can be monitored according to the manufacturing importance of the complete sub-items based on the whole manufacturing process and the manufacturing importance of the incomplete sub-items based on the whole manufacturing process;
wherein Y is i Representing the manufacturing importance of the ith complete sub-item; z j Representing the manufacturing importance of the jth incomplete sub-item; s. the i Complete item information representing the ith complete sub-item; y represents process information of the manufacturing process; t is a unit of i Representing the complete manufacturing time of the ith complete sub-item; t is zong A flow manufacturing time representing a manufacturing flow; is a direct change i Information-time adjustment coefficients representing the ith complete sub-item; r j Item information currently contained representing a jth incomplete sub-item; t is j Represents the manufacturing time currently contained for the jth incomplete sub-item; t is J Representing a complete manufacturing time for the jth incomplete sub-item; is a direct change j Information representing the jth incomplete sub-item-time adjustment factor;
when importance of production Y i When the number of the corresponding complete sub-items is larger than a first preset value, the corresponding complete sub-items are regarded as items which need to be monitored;
when importance of manufacture Z j When the value is larger than the second preset value, the corresponding incomplete sub-project is regarded as a project which needs to be monitored;
otherwise, the remaining complete sub-items and the remaining incomplete sub-items are regarded as monitorable items;
determining a first main abnormal sequence of the project which needs to be monitored and is based on the abnormal behavior process, determining a maximum time period corresponding to the first main abnormal sequence, setting the maximum time period as a first thread segment, simultaneously determining a second main abnormal sequence of the project which can be monitored and is based on the abnormal behavior process, determining a middle time point of the second main abnormal sequence, and regarding the second main abnormal sequence as a monitorable point;
constructing an abnormal thread section corresponding to the initial point and the termination point, calibrating the first thread section and the monitorable point on the abnormal thread section, and adding a secondary monitoring event;
and obtaining an abnormal monitoring set based on the secondary monitoring sections corresponding to all the secondary monitoring events and the second monitoring points.
6. The dynamic monitoring method of claim 1, wherein obtaining a normal monitoring set for the manufacturing process based on the manufacturing process database comprises:
matching the monitoring points and the monitoring sections of each sub-item in the manufacturing process in a normal operation state based on the manufacturing process database;
and constructing to obtain a normal monitoring set according to all the monitoring points and the monitoring sections.
7. The dynamic monitoring method of claim 1, wherein the dynamic monitoring of the semiconductor to be manufactured in the manufacturing process according to the dynamic monitoring set comprises:
constructing a flow line of a manufacturing flow of the semiconductor to be manufactured;
configuring a secondary monitoring section and a second monitoring point in an abnormal monitoring set at a first position on the flow line;
configuring a normal monitoring section and a normal monitoring point in a normal monitoring set at a second position on the flow line;
if the first position is overlapped with the second position, the secondary monitoring section and the second monitoring point configured on the overlapped position are reserved, and the normal monitoring section and the normal monitoring point configured on the overlapped position are deleted to obtain a process monitoring line;
setting a corresponding section monitoring trigger window to a first monitoring section of the process monitoring line, and issuing a corresponding first monitoring task according to the to-be-executed item attribute and the to-be-monitored important attribute corresponding to the first monitoring section;
setting a corresponding point trigger response window to a first monitoring point of the process monitoring line, and issuing a corresponding second monitoring task according to the item execution type of the first monitoring point and the point attribute of the first monitoring point;
identifying the monitoring numbers in the first monitoring task and the second monitoring task and the combination characteristics of the monitoring marks, and determining corresponding monitoring modes;
when the manufacturing progress of the semiconductor to be manufactured reaches the entry point of the corresponding exit setting window, calling a corresponding monitoring mode to perform dynamic monitoring;
wherein the window includes a trigger entry point and a trigger exit point.
8. A dynamic monitoring system for a semiconductor manufacturing process, comprising:
the device comprises a first determining module, a second determining module and a control module, wherein the first determining module is used for determining a semiconductor to be manufactured, calling the manufacturing process of the semiconductor to be manufactured from a manufacturing process database and acquiring the manufacturing index in the manufacturing process;
the second determining module is used for performing type analysis on historical manufacturing data related to the semiconductor to be manufactured according to the manufacturing indexes to obtain a historical data set of each manufacturing index, determining whether definable abnormal indexes exist or not based on the data analysis model, and if yes, determining manufacturing abnormal events corresponding to the abnormal indexes;
a third determining module, configured to determine a through flow line of the corresponding abnormal indicator in the manufacturing flow, determine an abnormal thread segment in the through flow line based on the manufacturing abnormal event, and construct an abnormal monitoring set of the abnormal thread segment;
the acquisition module is used for acquiring a normal monitoring set of the manufacturing process based on the manufacturing process database;
the dynamic monitoring module is used for constructing the abnormal monitoring set and the normal monitoring set into a dynamic monitoring set of the semiconductor to be manufactured and realizing dynamic monitoring of the semiconductor to be manufactured in the manufacturing process according to the dynamic monitoring set;
wherein determining the manufacturing exception event corresponding to the exception indicator includes:
constructing the same parameter vector of the same parameter in the manufacturing process of different batches based on the historical data set corresponding to the abnormal index, and further constructing a parameter matrix F corresponding to the abnormal index;
wherein A is 11 Representing the co-parameter vector of the 1 st parameter in the abnormal index in the 1 st batch manufacturing process; a. The 1n1 Representing the co-parameter vector of the 1 st parameter in the abnormal index in the manufacturing process of the n1 th batch; a. The n21 Representing the co-parameter vector of the n2 th parameter in the abnormal index in the 1 st batch manufacturing process; a. The n2n1 Representing the co-parameter vector of the n2 th parameter in the abnormal index in the manufacturing process of the n1 th batch; n2 represents a total parameter corresponding to the abnormality index; n1 represents the total batch;
performing subtraction processing on the parameter matrix F and the standard matrix B to obtain a difference matrix C;
respectively inputting the column elements in the difference matrix C into an element anomaly analysis model to obtain first anomaly information of different parameters in the same batch, and setting a first anomaly label;
respectively inputting the row elements in the difference matrix C into an element anomaly analysis model to obtain second anomaly information of the same parameter in different batches, and setting a second anomaly label;
acquiring a first sequence of the first abnormal information, determining abnormal dispersed distribution, and acquiring a second sequence of the second abnormal information, determining abnormal concentrated distribution;
and determining the satisfied abnormal event conditions based on the label setting result and the distribution result, and further obtaining the manufacturing abnormal event corresponding to the manufacturing index.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210143870.9A CN114429311B (en) | 2022-02-17 | 2022-02-17 | Dynamic monitoring method and system for semiconductor manufacturing process |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210143870.9A CN114429311B (en) | 2022-02-17 | 2022-02-17 | Dynamic monitoring method and system for semiconductor manufacturing process |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114429311A CN114429311A (en) | 2022-05-03 |
CN114429311B true CN114429311B (en) | 2023-04-07 |
Family
ID=81312894
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210143870.9A Active CN114429311B (en) | 2022-02-17 | 2022-02-17 | Dynamic monitoring method and system for semiconductor manufacturing process |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114429311B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115796058B (en) * | 2023-02-06 | 2023-04-14 | 广州志橙半导体有限公司 | Equipment data analysis method and system of CVD equipment |
CN116705603B (en) * | 2023-08-04 | 2023-10-10 | 北京市天润中电高压电子有限公司 | Manufacturing method of semiconductor pulse high-voltage diode |
CN116759311B (en) * | 2023-08-16 | 2023-11-14 | 北京市天润中电高压电子有限公司 | Manufacturing method of semiconductor avalanche high-voltage diode |
CN117075522B (en) * | 2023-09-27 | 2024-03-29 | 金麒麟新能源股份有限公司 | Remote centralized operation control method and system for energy system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102361014A (en) * | 2011-10-20 | 2012-02-22 | 上海大学 | State monitoring and fault diagnosis method for large-scale semiconductor manufacture process |
CN113656255A (en) * | 2021-08-26 | 2021-11-16 | 韩玉兰 | Operation abnormity judgment method based on chip operation data |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4882239B2 (en) * | 2005-02-16 | 2012-02-22 | 東京エレクトロン株式会社 | Semiconductor manufacturing apparatus, computer program, and storage medium |
CN109684179B (en) * | 2018-09-03 | 2022-05-17 | 平安科技(深圳)有限公司 | Early warning method, device, equipment and storage medium for system fault |
CN116719698A (en) * | 2019-05-06 | 2023-09-08 | 创新先进技术有限公司 | Identification method and device for index abnormality reasons |
FR3098938B1 (en) * | 2019-07-15 | 2022-01-07 | Bull Sas | Method and device for determining an anomaly prediction performance index value in a computer infrastructure from performance indicator values |
US11404331B2 (en) * | 2020-06-29 | 2022-08-02 | Vanguard International Semiconductor Corporation | System and method for determining cause of abnormality in semiconductor manufacturing processes |
-
2022
- 2022-02-17 CN CN202210143870.9A patent/CN114429311B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102361014A (en) * | 2011-10-20 | 2012-02-22 | 上海大学 | State monitoring and fault diagnosis method for large-scale semiconductor manufacture process |
CN113656255A (en) * | 2021-08-26 | 2021-11-16 | 韩玉兰 | Operation abnormity judgment method based on chip operation data |
Also Published As
Publication number | Publication date |
---|---|
CN114429311A (en) | 2022-05-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114429311B (en) | Dynamic monitoring method and system for semiconductor manufacturing process | |
EP1252556B1 (en) | Method for automatic fault tree synthesis | |
US8594826B2 (en) | Method and system for evaluating a machine tool operating characteristics | |
US20050159835A1 (en) | Device for and method of creating a model for determining relationship between process and quality | |
US11487996B2 (en) | Real-time predictive maintenance of hardware components using a stacked deep learning architecture on time-variant parameters combined with a dense neural network supplied with exogeneous static outputs | |
CN113609790B (en) | Product virtual measuring method, system, device and medium | |
CN114551271A (en) | Method and device for monitoring machine operation condition, storage medium and electronic equipment | |
JP2009258890A (en) | Influence factor specifying device | |
Perelló et al. | Random diffusion and leverage effect in financial markets | |
TW201824100A (en) | Product quality prediction method for mass customization | |
US6551847B2 (en) | Inspection analyzing apparatus and semiconductor device | |
US20170176985A1 (en) | Method for predicting end of line quality of assembled product | |
US20100131091A1 (en) | A method and relative device for the management of technological recipe information to aid in defining process flows, in particular for the development and production of micro-and nanotechnology devices in cleanroom laboratories | |
CN112950908B (en) | Data monitoring and early warning method, system, medium and electronic terminal | |
US7007038B1 (en) | Defect management database for managing manufacturing quality information | |
JP2001067109A (en) | Method and device for managing quality and recording medium recording quality management prograqm | |
CN113868948A (en) | User-oriented dynamic threshold model training system and method | |
CN113835947A (en) | Method and system for determining abnormality reason based on abnormality identification result | |
CN115689126A (en) | Data processing method, device, equipment and medium | |
JP4918343B2 (en) | Production management system, production index display method | |
CN115526431A (en) | Data analysis method based on industrial Internet and related device | |
US20080125883A1 (en) | Method and apparatus for consolidating process control | |
CN111199419B (en) | Stock abnormal transaction identification method and system | |
US10268562B1 (en) | Advanced manufacturing insight system for semiconductor application | |
CN111311086A (en) | Capacity monitoring method and device and computer readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |