CN116414780A - A data management system and management method - Google Patents

Abstract

The application provides a data management system and a management method, wherein the method comprises the following steps: acquiring a process formula file of at least one production machine end from the production machine end; setting a first identification mark for identifying each process recipe file; decomposing each process recipe file into a plurality of step-level data units, and setting a second identification mark for identifying the plurality of step-level data units for the plurality of step-level data units; and decomposing each step-level data unit into a plurality of process recipe parameters, and setting a third identification mark for identifying the plurality of process recipe parameters for the plurality of process recipe parameters. The application provides a data management system and a management method, which can realize file-level operation and parameter-level operation and realize accurate management of all process recipe parameters in a process recipe file.

Description

A data management system and management method

technical field

The present application relates to the field of semiconductor technology, and in particular to a data management system and a management method.

Background technique

Process recipe parameters are the parameters contained in the process recipe file, and are the most critical configuration information in the production process. Strengthening the management of process formula parameters is the only way to ensure production quality. The management requirements for process formula parameters mainly include: file-level operation, which requires complete file content without redundancy, correct order of internal parameters, needs to be able to view, call, transfer, status management, version management, and needs to ensure the integrity of the file. Deletions and omissions, the setting value and sequence of internal parameters have a direct impact on production quality, and it is necessary to be able to compare multiple files; parameter-level operations require horizontal operations across files and vertical operations across versions, and the ability to view the details of process recipes Setting values can change the retrospective of historical records, summarize the impact on output quality, group process recipes according to purposes, and summarize and manage parameter settings in different process recipes. Parameter-level information is the basis for the realization of big data functions.

The current process recipe parameter management function only supports file-level operations, and parameter-level operations need to be accessed individually file by file. Coordination and generalization management work requires a lot of manual work to achieve the generalization work.

Therefore, it is necessary to provide a more effective and reliable technical solution.

Contents of the invention

The application provides a data management system and a management method, which can take into account both file-level operations and parameter-level operations, and realize precise management of all process formula parameters in process formula files.

One aspect of the present application provides a data management system, including: a memory for storing process recipe files obtained from the production machine; an automatic equipment control unit, which performs the following steps when working: from at least one production The machine side obtains the process recipe file of the production machine side; sets a first identification mark for identifying the process recipe file for each of the process recipe files; decomposes each of the process recipe files into several step-level data unit, and set a second identification mark for identifying the plurality of step-level data units for the plurality of step-level data units, wherein the first identification mark and the second identification mark are jointly used to identify the step level data unit; decompose each of the step-level data units into a number of process recipe parameters, and set a third identification mark for identifying the number of process recipe parameters for the number of process recipe parameters, wherein the first The identification mark, the second identification mark and the third identification mark are jointly used to identify the process recipe parameter.

In some embodiments of the present application, when the automatic equipment control unit works, it also: establishes file-level attribute information for the process recipe file, and the file-level attribute information includes editable attributes and non-editable attributes, wherein the The editable attributes include status and scope of application, and the non-editable attributes include name, production machine to which it belongs, current version and update time.

In some embodiments of the present application, when the device automatic control unit works, it also: establishes step-level attribute information for the step-level data unit, and the step-level attribute information includes position information and type information.

In some embodiments of the present application, when the equipment automatic control unit works, it also: establishes parameter attribute information for the process recipe parameters, and the parameter attribute information includes position information and identifiable information.

In some embodiments of the present application, the first identification mark includes the information of the production machine side; the second identification mark includes the step information of the step-level data unit; the third identification mark includes the The name of the process recipe parameter.

In some embodiments of the present application, the automatic control unit of the equipment further includes: an automatic identification model established through artificial intelligence deep learning, the automatic identification model is used to identify the step-level data unit to which the process formula parameters belong and to identify The process recipe file to which the step-level data unit belongs.

In some embodiments of the present application, the process recipe file is a folder, the step-level data unit is a subfolder after the folder is opened, and the process recipe parameters are located in the subfolder.

In some embodiments of the present application, the process recipe file is a hyperlink, the step-level data unit is a sub-hyperlink expanded from the hyperlink, and the process recipe parameter is located in the sub-hyperlink.

In some embodiments of the present application, the process recipe file is a table, the step-level data unit is a subtable of the table, and the process recipe parameters are located in the subtable.

Another aspect of the present application also provides a data management method, including: performing the following steps through the equipment automatic control unit: obtaining the process recipe file of the production machine end from at least one production machine end; Setting a first identification mark for identifying the process recipe file; decomposing each of the process recipe files into several step-level data units, and setting the several step-level data units for identifying the several step-level data The second identification mark of the unit, wherein the first identification mark and the second identification mark are jointly used to identify the step-level data unit; each of the step-level data units is decomposed into several process recipe parameters, and Setting a third identification mark for identifying the plurality of process recipe parameters for the several process recipe parameters, wherein the first identification mark, the second identification mark and the third identification mark are jointly used to identify the The process formula parameters described above.

In some embodiments of the present application, the management method further includes: establishing file-level attribute information for the process recipe file, the file-level attribute information including editable attributes and non-editable attributes, wherein the editable attribute Including status and scope of application, the non-editable attributes include name, production machine to which it belongs, current version and update time.

In some embodiments of the present application, the management method further includes: establishing step-level attribute information for the step-level data unit, where the step-level attribute information includes location information and type information.

In some embodiments of the present application, the management method further includes: establishing parameter attribute information for the process recipe parameters, where the parameter attribute information includes location information and identifiable information.

In some embodiments of the present application, the first identification mark includes the information of the production machine side; the second identification mark includes the step information of the step-level data unit; the third identification mark includes the The name of the process recipe parameter.

In some embodiments of the present application, the management method further includes: establishing an automatic identification model for the automatic control unit of the equipment through deep learning of artificial intelligence, and the automatic identification model is used to identify the step level to which the process formula parameter belongs The data unit identifies the process recipe file to which the step-level data unit belongs.

In some embodiments of the present application, the process recipe file is a folder, the step-level data unit is a subfolder after the folder is opened, and the process recipe parameters are located in the subfolder.

In some embodiments of the present application, the process recipe file is a hyperlink, the step-level data unit is a sub-hyperlink expanded from the hyperlink, and the process recipe parameter is located in the sub-hyperlink.

In some embodiments of the present application, the process recipe file is a table, the step-level data unit is a subtable of the table, and the process recipe parameters are located in the subtable.

The application provides a data management system and a management method, which can take into account both file-level operations and parameter-level operations, and realize precise management of all process formula parameters in process formula files.

Description of drawings

The following figures describe in detail exemplary embodiments disclosed in this application. Wherein the same reference numerals denote similar structures in the several views of the drawings. Those of ordinary skill in the art will understand that these embodiments are non-limiting, exemplary embodiments, and that the accompanying drawings are for illustration and description purposes only, and are not intended to limit the scope of the application, and other embodiments are also possible Complete the invention intention among the application likewise. It should be understood that the drawings are not drawn to scale.

in:

Fig. 1 is a schematic structural diagram of the data management system described in the embodiment of the present application;

FIG. 2 is a flow chart of the data management method described in the embodiment of the present application.

Detailed ways

The following description provides specific application scenarios and requirements of the application, with the purpose of enabling those skilled in the art to manufacture and use the contents of the application. Various local modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments and embodiments without departing from the spirit and scope of the application. application. Thus, the application is not limited to the embodiments shown, but is to be accorded the widest scope consistent with the claims.

The technical solution of the present invention will be described in detail below in conjunction with the embodiments and the accompanying drawings.

The current process recipe files are mainly stored on the production machine side. When it is necessary to manage (such as modify, call, view, etc.) the process recipe file, in order to reduce the manual transcription work at the production machine end, the process recipe file at the production machine end is generally transcribed into a system file as a whole through the equipment automatic control unit. The process formula management system is available for users to view, focusing on solving the issue of remote visualization.

However, different equipment manufacturers or even different models and versions of the same manufacturer have very different process recipe structures, and the system files transcribed from this also have different structures, and there is no standard structure. The service functions in the current process formula management system are all based on the above. It mainly focuses on file-level operations of system files and lacks parameter-level functions. Individual parameter-level functions are only identified by parameter path information, lacking necessary identification information, making the parameter-level display insufficient and completely mere formality. The overall lack of inductive statistics at the parameter level, the data granularity is too large, and there is no foundation for big data.

In view of the above problems, this application provides a data management system and management method, which accurately transcribes the process recipe file at the production machine end according to the structure of file level-step level-parameter level, and sets a separate identification for each level of data Marking can take into account both file-level operations and parameter-level operations, and realize precise management of all process recipe parameters in the process recipe file.

FIG. 1 is a schematic structural diagram of a data management system described in an embodiment of the present application. The data management system described in the embodiment of the present application will be described in detail below with reference to the accompanying drawings.

The embodiment of the present application provides a data management system, as shown in FIG. 1 , including: a memory 130 for storing the process recipe file obtained from the production machine terminal 110; an automatic equipment control unit 120, the automatic equipment control unit 120 Perform the following steps during work: obtain the process recipe file of the production machine end 110 from at least one production machine end 110; set the first identification mark for identifying the process recipe file for each of the process recipe files; One said process recipe file is decomposed into several step-level data units, and a second identification mark for identifying said several step-level data units is set for said several step-level data units, wherein said first identification mark and said The second identification mark is commonly used to identify the step-level data unit; each of the step-level data units is decomposed into a number of process recipe parameters, and the number of process recipe parameters is set to identify the number of process recipe parameters The third identification mark, wherein, the first identification mark, the second identification mark and the third identification mark are jointly used to identify the process recipe parameter.

In some embodiments of the present application, the production tool end 110 may be any production tool, for example, a photolithography tool, a CVD tool, a cleaning tool, and the like. In order to facilitate the description of the technical solution of the present application, the embodiment of the present application only uses the photolithography machine as an example for illustration, but it should be noted that, in practice, the production machine end 110 can be any production machine, or Multiple production machines. The following only uses the photolithography machine as an example for illustration.

The production machine end 110 is, for example, a lithography machine, and the number of the lithography machine is, for example, A (there are usually multiple machines of each type in the factory, and one of the lithography machines numbered A is used here as demonstration). The lithography machine A is used to execute the lithography process, and all process recipe parameters of the lithography process are set on the lithography machine A. The photolithography process includes, for example, step A (deposition of photoresist), step B (exposure and development), step C (etching) and step D (removal of photoresist). It should be noted that this is only a simple demonstration, and the actual steps of the photolithography process are more complicated and include more steps. The process recipe parameters of each step include, for example: step A (material of photoresist, amount of photoresist, deposition time, etc.), step B (type of mask plate, exposure time, amount of developer, etc.), step C (type of etchant, dosage of etchant, etching time, etc.), step D (ashing reactant and its dosage, etc.).

The storage method of the process formula parameters on the reproduction machine terminal 110 is irregular, and the naming of process formula parameters is also irregular. When the user needs to view or modify the process formula parameters, he needs to search for himself at the production machine terminal 110. There is no convenient retrieval method. low efficiency. Therefore, in the technical solution of the present application, the process recipe parameters of the production machine terminal 110 are transcribed into a regular process recipe file by the equipment automatic control unit 120 and stored in the memory 130 . The user can subsequently view, call and modify the process recipe file in the memory 130 .

Referring to FIG. 1 , the automatic equipment control unit 120 obtains the process recipe file of the production machine end 110 from at least one production machine end 110 when working.

First, the automatic equipment control unit 120 obtains (for example, by downloading) the process recipe file of the production machine terminal 110 from the production machine terminal 110 , and the process recipe file includes all process recipe parameters of the production machine terminal 110 . When there are multiple production machine terminals 110, multiple process recipe files corresponding to each production machine terminal 110 are obtained. For example, the process recipe file obtained from lithography machine A is a file, which contains all the process recipe parameters in lithography machine A; the process recipe file obtained from CVD machine is a file, and the file contains all process recipe parameters in the CVD machine; the process recipe file obtained from the cleaning machine is a file, which contains all the process recipe parameters in the cleaning machine.

There is no containment relationship between the multiple process recipe files, and the multiple process recipe files are arranged side by side. When the process recipe file is a folder, the multiple process recipe files are multiple parallel folders; when the process recipe file is a hyperlink, the multiple process recipe files are multiple parallel hyperlinks. link; when the process recipe file is a table, the multiple process recipe files are multiple parallel tables.

Then, the equipment automatic control unit 120 sets a first identification mark for identifying the process recipe file for each of the process recipe files. In the technical solution of the present application, different process recipe files are marked in a regular naming manner (first identification mark), so that users can directly use the first identification mark to find the corresponding process recipe file.

In some embodiments of the present application, the first identification mark includes the information of the production machine end 110 . Associating the first identification mark with the information of the production machine end 110 can facilitate the user to memorize the first identification mark and use the first identification mark. For example, the information of the production machine end 110 is, for example, the serial number of the production machine end 110 . Then when the production machine end 110 is a lithography machine A, the first identification mark is "lithography machine A"; when the production machine end 110 is a CVD machine, the first identification mark is "CVD machine"; when the production machine end 110 is a cleaning machine, the first identification mark is "cleaning machine". In this manner, it is easy for the user to memorize and use the first identification mark. When the user wants to search for the process recipe file of lithography machine A, he can use the first identification mark "lithography machine A" to search.

In some embodiments of the present application, when the automatic equipment control unit 120 works, it also: establishes file-level attribute information for the process recipe file, and the file-level attribute information includes editable attributes and non-editable attributes, wherein the The above editable attributes include configuration information such as status (available, disabled, applicable to a small range) and scope of application (specifically specifying the scope of application), which are used to implement management operations. The non-editable attributes include name, production machine, current version Natural attributes such as and update time are used to realize the operation of screening and selecting parts of a single file.

Next, the equipment automatic control unit 120 decomposes each of the process recipe files into several step-level data units, and sets a second identification for identifying the several step-level data units for the several step-level data units mark.

For example, the production machine end 110 is a lithography machine A, and the lithography process of the lithography machine A includes, for example: step A (deposition of photoresist), step B (exposure and development), step C (etching ) and step D (removal of photoresist). Then, the process recipe file of the lithography machine A (the corresponding first identification mark is "lithography machine A") is decomposed into four step-level data units of step A, step B, step C and step D. Each step-level data unit includes all process recipe parameters in this step. For example, the step-level data unit of step A includes all process recipe parameters required for the step of depositing photoresist, including the material of photoresist, the amount of photoresist used, and the deposition time.

The plurality of step-level data units are not in an inclusion relationship, and the plurality of step-level data units are arranged side by side. When the process recipe file is a folder, the multiple step-level data units are multiple parallel subfolders after the folder is opened; when the process recipe file is a hyperlink, the multiple The step-level data units are multiple parallel sub-hyperlinks after the hyperlink is expanded; when the process recipe file is a table, the multiple step-level data units are multiple sub-tables of the table.

In some embodiments of the present application, the second identification mark includes step information of the step-level data unit. Associating the second identification mark with the step information can facilitate the user to memorize the second identification mark and use the second identification mark. For example, the step information is, for example, the number of the step. Then when the step information is step A, the second identification mark is "step A"; when the step information is step B, the second identification mark is "step B"; when the step information When it is step C, the second identification mark is "step C"; when the step information is step D, the second identification mark is "step D". In this manner, it is easy for the user to memorize and use the second identification mark.

In the technical solution of the present application, different step-level data units are marked in a regular naming manner (second identification mark), and the first identification mark and the second identification mark are used together to identify a step-level data unit unit. For example, when the user wants to search for the process recipe file of step A in lithography machine A, the first identification mark and the second identification mark "lithography machine A and step A" can be used in series to query.

In some embodiments of the present application, when the device automatic control unit 120 is working, it also: establishes step-level attribute information for the step-level data unit, the step-level attribute information includes position information, type information and parameter set information, The combination of the position information and the type information is the unique identification basis of the step-level data unit. The position information reflects the position information of the current step in the process parameter file of the production machine; the type information reflects the commonality of the parameter set of the current step, and is the basis information for parameter classification; the parameter set is composed of multiple parameter information, Parameter names are unique within the current step.

Finally, the equipment automatic control unit 120 decomposes each of the step-level data units into several process recipe parameters, and sets a third identification mark for identifying the several process recipe parameters for the several process recipe parameters, wherein , the first identification mark, the second identification mark and the third identification mark are jointly used to identify the process recipe parameter.

For example, the production machine end 110 is a lithography machine A, and the lithography process of the lithography machine A includes step A (deposition of photoresist), and the process recipe parameters of the step A include, for example: photoresist The material, the amount of photoresist, the deposition time, etc. Then, the step-level data unit corresponding to step A in the process recipe file of the lithography machine A (the corresponding first identification mark is "lithography machine A") (the corresponding second identification mark is "step A") is decomposed into several process recipe parameters. Each technological recipe parameter includes the type, value and unit of the technological recipe parameter. For example, the process recipe parameters of deposition time include: type (time), value (10), and unit (second).

The several process recipe parameters are not in an inclusive relationship, and the several process recipe parameters are set side by side. When the process recipe file is a folder, the multiple step-level data units are multiple parallel subfolders after the folder is opened, and the several process recipe parameters are located in the subfolder; when When the process recipe file is a hyperlink, the multiple step-level data units are multiple parallel sub-hyperlinks after the hyperlink is expanded, and the several process recipe parameters are located in the sub-hyperlinks; when the When the process recipe file is a table, the multiple step-level data units are multiple sub-tables of the table, and the several process recipe parameters are located in the sub-tables.

In some embodiments of the present application, the third identification mark includes the name of the process recipe parameter. Associating the third identification mark with the name of the process recipe parameter can facilitate the user to memorize the third identification mark and use the third identification mark. For example, when the name of the process recipe parameter is photoresist material, the third identification mark is "material of photoresist"; when the name of the process recipe parameter is the amount of photoresist, the first The third identification mark is "amount of photoresist"; when the name of the process recipe parameter is deposition time, the third identification mark is "deposition time". In this way, it is easy for the user to memorize and use the third identification mark.

In the technical solution of the present application, different process formula parameters are marked in a regular naming manner (third identification mark), and the first identification mark, the second identification mark and the third identification mark are used together to identify a process recipe parameter. For example, when the user wants to find the process recipe parameter of the deposition time of step A in lithography machine A, use the first identification mark, the second identification mark and the third identification mark in series "lithography machine A and step A and Deposition time" can be queried.

In some embodiments of the present application, during operation, the automatic equipment control unit 120 further: establishes parameter attribute information for the process recipe parameters, and the parameter attribute information includes position information and identifiable information. The position information is the position information of the current process formula parameters in the process formula file on the production machine side; the identifiable information includes parameter name, parameter value, format, upper and lower limits and other auxiliary information visible in the system.

In some embodiments of the present application, the automatic control unit 120 of the equipment further includes: an automatic recognition model, established through artificial intelligence deep learning, the automatic recognition model is used to identify the step-level data unit to which the process recipe parameters belong and A process recipe file to which the step-level data unit belongs is identified. For example, collect big data information on the public network, that is, all production machines involved in the semiconductor process, the process flow and process steps corresponding to each production machine, the type of process recipe parameters for each process step, etc., and use these big data Carry out artificial intelligence training, establish a general automatic identification model that can identify the process steps to which the process formula parameters belong according to the type and value of the process formula parameters, and further identify the production machine to which the process steps belong; The production machine and the corresponding process steps and process formula parameters are fine-tuned and trained on the general automatic identification model to obtain a local automatic identification model suitable for local semiconductor factories. The local automatic identification model can identify the process step to which the process formula parameter belongs according to the type and value of the process formula parameter acquired by the equipment automatic control unit 120 and further identify the production machine to which the process step belongs.

The technical solution of the present application is automatically implemented by the automatic equipment control unit 120, which does not require manual participation. The local automatic identification model can automatically identify and help the automatic equipment control unit 120 to sort out the process recipe files, greatly improving Increased efficiency.

The data management system described in the technical solution of the present application can transcribe the process recipe file at the production machine end into the memory through the equipment automatic control unit. The transcribed process recipe file retains the compatibility of file-level operations. From the system level, the attributes of the file level (process recipe file) are divided into editable attributes and non-editable attributes. A general process recipe file structure is established: the process recipe file is composed of one or more step-level data units, each step-level data unit includes several process recipe parameters, and a process recipe parameter corresponds to the smallest information unit of the process recipe at the production machine end . The process formula file of the present application also expands the support of parameter-level operations. The process formula file with the same file structure can use the unified conversion function and the unique identification information of each layer of information (the first identification mark, the second identification mark and the third identification mark). Identification mark), can expand the file parameters and store each parameter separately, and can support identification to any parameter. The combination of the above expansion information and the file-level attributes can realize the inductive management of parameter bases across files.

This application provides a data management system, which provides a standard file structure after transcribing the process recipe file at the production machine end, and accurately decomposes the process recipe file into parameter-level data through fixed functions, which solves the problem of remote visualization of the process recipe file At the same time, accurate and efficient induction and arrangement of process formula parameters are realized. The technical solution of this application is compatible with the current old system, and the support for parameter-level operations has the characteristics of oriented to big data-level functions. It provides transcription functions, which can be freely converted between file-level operations and parameter-level operations, and can be called in real time as needed. That is, high precision and high efficiency.

This application provides a data management system, which can take into account both file-level operations and parameter-level operations, and realize precise management of all process formula parameters in process formula files.

FIG. 2 is a flow chart of the data management method described in the embodiment of the present application.

The embodiment of the present application also provides a data management method, as shown in FIG. 2 , including: performing the following steps through the equipment automatic control unit:

Step S1: Obtain the process formula file of the production machine end from at least one production machine end;

Step S2: setting a first identification mark for identifying the process recipe file for each of the process recipe files;

Step S3: decomposing each of the process recipe files into several step-level data units, and setting a second identification mark for identifying the several step-level data units for the several step-level data units;

Step S4: decomposing each of the step-level data units into several process recipe parameters, and setting a third identification mark for identifying the several process recipe parameters for the several process recipe parameters.

The data management method described in the embodiment of the present application will be described in detail below with reference to the accompanying drawings.

Referring to FIG. 2 , step S1 is to obtain (for example, download) the process recipe file of the production machine terminal 110 from at least one production machine terminal 110 .

The process recipe file includes all process recipe parameters of the production machine terminal 110 . When there are multiple production machine terminals 110, multiple process recipe files corresponding to each production machine terminal 110 are obtained. For example, the process recipe file obtained from lithography machine A is a file, which contains all the process recipe parameters in lithography machine A; the process recipe file obtained from CVD machine is a file, and the file contains all process recipe parameters in the CVD machine; the process recipe file obtained from the cleaning machine is a file, which contains all the process recipe parameters in the cleaning machine.

There is no containment relationship between the multiple process recipe files, and the multiple process recipe files are arranged side by side. When the process recipe file is a folder, the multiple process recipe files are multiple parallel folders; when the process recipe file is a hyperlink, the multiple process recipe files are multiple parallel hyperlinks. link; when the process recipe file is a table, the multiple process recipe files are multiple parallel tables.

Continuing to refer to FIG. 2 , in step S2, a first identification mark for identifying the process recipe file is set for each of the process recipe files.

In the technical solution of the present application, different process recipe files are marked in a regular naming manner (first identification mark), so that users can directly use the first identification mark to find the corresponding process recipe file.

In some embodiments of the present application, the first identification mark includes the information of the production machine end 110 . Associating the first identification mark with the information of the production machine end 110 can facilitate the user to memorize the first identification mark and use the first identification mark. For example, the information of the production machine end 110 is, for example, the serial number of the production machine end 110 . Then when the production machine end 110 is a lithography machine A, the first identification mark is "lithography machine A"; when the production machine end 110 is a CVD machine, the first identification mark is "CVD machine"; when the production machine end 110 is a cleaning machine, the first identification mark is "cleaning machine". In this manner, it is easy for the user to memorize and use the first identification mark. When the user wants to search for the process recipe file of lithography machine A, he can use the first identification mark "lithography machine A" to search.

In some embodiments of the present application, when the automatic equipment control unit 120 works, it also: establishes file-level attribute information for the process recipe file, and the file-level attribute information includes editable attributes and non-editable attributes, wherein the The above editable attributes include configuration information such as status (available, disabled, applicable to a small range) and scope of application (specifically specifying the scope of application), which are used to implement management operations. The non-editable attributes include name, production machine, current version Natural attributes such as and update time are used to realize the operation of screening and selecting parts of a single file.

Continue to refer to Figure 2, step S3, decompose each of the process recipe files into several step-level data units, and set the second identification for identifying the several step-level data units for the several step-level data units mark, wherein the first identification mark and the second identification mark are jointly used to identify the step-level data unit.

For example, the production machine end 110 is a lithography machine A, and the lithography process of the lithography machine A includes, for example: step A (deposition of photoresist), step B (exposure and development), step C (etching ) and step D (removal of photoresist). Then, the process recipe file of the lithography machine A (the corresponding first identification mark is "lithography machine A") is decomposed into four step-level data units of step A, step B, step C and step D. Each step-level data unit includes all process recipe parameters in this step. For example, the step-level data unit of step A includes all process recipe parameters required for the step of depositing photoresist, including the material of photoresist, the amount of photoresist used, and the deposition time.

The plurality of step-level data units are not in an inclusion relationship, and the plurality of step-level data units are arranged side by side. When the process recipe file is a folder, the multiple step-level data units are multiple parallel subfolders after the folder is opened; when the process recipe file is a hyperlink, the multiple The step-level data units are multiple parallel sub-hyperlinks after the hyperlink is expanded; when the process recipe file is a table, the multiple step-level data units are multiple sub-tables of the table.

In some embodiments of the present application, the second identification mark includes step information of the step-level data unit. Associating the second identification mark with the step information can facilitate the user to memorize the second identification mark and use the second identification mark. For example, the step information is, for example, the number of the step. Then when the step information is step A, the second identification mark is "step A"; when the step information is step B, the second identification mark is "step B"; when the step information When it is step C, the second identification mark is "step C"; when the step information is step D, the second identification mark is "step D". In this manner, it is easy for the user to memorize and use the second identification mark.

In the technical solution of the present application, different step-level data units are marked in a regular naming manner (second identification mark), and the first identification mark and the second identification mark are used together to identify a step-level data unit unit. For example, when the user wants to search for the process recipe file of step A in lithography machine A, the first identification mark and the second identification mark "lithography machine A and step A" can be used in series to query.

In some embodiments of the present application, the data management method further includes: establishing step-level attribute information for the step-level data unit, the step-level attribute information including position information, type information and parameter set information, the position The combination of information and type information is the only basis for identification of step-level data units. The position information reflects the position information of the current step in the process parameter file of the production machine; the type information reflects the commonality of the parameter set of the current step, and is the basis information for parameter classification; the parameter set is composed of multiple parameter information, Parameter names are unique within the current step.

Continuing to refer to Figure 2, step S4, decomposing each of the step-level data units into several process recipe parameters, and setting a third identification mark for identifying the several process recipe parameters for the several process recipe parameters, Wherein, the first identification mark, the second identification mark and the third identification mark are jointly used to identify the process recipe parameters.

For example, the production machine end 110 is a lithography machine A, and the lithography process of the lithography machine A includes step A (deposition of photoresist), and the process recipe parameters of the step A include, for example: photoresist The material, the amount of photoresist, the deposition time, etc. Then, the step-level data unit corresponding to step A in the process recipe file of the lithography machine A (the corresponding first identification mark is "lithography machine A") (the corresponding second identification mark is "step A") is decomposed into several process recipe parameters. Each technological recipe parameter includes the type, value and unit of the technological recipe parameter. For example, the process recipe parameters of deposition time include: type (time), value (10), and unit (second).

The several process recipe parameters are not in an inclusive relationship, and the several process recipe parameters are set side by side. When the process recipe file is a folder, the multiple step-level data units are multiple parallel subfolders after the folder is opened, and the several process recipe parameters are located in the subfolder; when When the process recipe file is a hyperlink, the multiple step-level data units are multiple parallel sub-hyperlinks after the hyperlink is expanded, and the several process recipe parameters are located in the sub-hyperlinks; when the When the process recipe file is a table, the multiple step-level data units are multiple sub-tables of the table, and the several process recipe parameters are located in the sub-tables.

In some embodiments of the present application, the third identification mark includes the name of the process recipe parameter. Associating the third identification mark with the name of the process recipe parameter can facilitate the user to memorize the third identification mark and use the third identification mark. For example, when the name of the process recipe parameter is photoresist material, the third identification mark is "material of photoresist"; when the name of the process recipe parameter is the amount of photoresist, the first The third identification mark is "amount of photoresist"; when the name of the process recipe parameter is deposition time, the third identification mark is "deposition time". In this way, it is easy for the user to memorize and use the third identification mark.

In the technical solution of the present application, different process formula parameters are marked in a regular naming manner (third identification mark), and the first identification mark, the second identification mark and the third identification mark are used together to identify a process recipe parameter. For example, when the user wants to find the process recipe parameter of the deposition time of step A in lithography machine A, use the first identification mark, the second identification mark and the third identification mark in series "lithography machine A and step A and Deposition time" can be queried.

In some embodiments of the present application, the data management method further includes: establishing parameter attribute information for the process recipe parameters, where the parameter attribute information includes location information and identifiable information. The position information is the position information of the current process formula parameters in the process formula file on the production machine side; the identifiable information includes parameter name, parameter value, format, upper and lower limits and other auxiliary information visible in the system.

In some embodiments of the present application, the data management method further includes: establishing an automatic identification model through artificial intelligence deep learning, and the automatic identification model is used to identify the step-level data unit to which the process formula parameter belongs and to identify the The process recipe file to which the step-level data unit belongs. For example, collect big data information on the public network, that is, all production machines involved in the semiconductor process, the process flow and process steps corresponding to each production machine, the type of process recipe parameters for each process step, etc., and use these big data Carry out artificial intelligence training, establish a general automatic identification model that can identify the process steps to which the process formula parameters belong according to the type and value of the process formula parameters, and further identify the production machine to which the process steps belong; The production machine and the corresponding process steps and process formula parameters are fine-tuned and trained on the general automatic identification model to obtain a local automatic identification model suitable for local semiconductor factories. The local automatic identification model can identify the process step to which the process formula parameter belongs according to the type and value of the process formula parameter acquired by the equipment automatic control unit 120 and further identify the production machine to which the process step belongs.

The technical solution of the present application can be automatically implemented by the automatic equipment control unit 120 without manual participation, and the local automatic identification model can automatically identify and help the automatic equipment control unit 120 to organize the process recipe files, Greatly improved efficiency.

The data management method described in the technical solution of the present application can transcribe the process recipe file at the production machine end into the memory. The transcribed process recipe file retains the compatibility of file-level operations. From the system level, the attributes of the file level (process recipe file) are divided into editable attributes and non-editable attributes. A general process recipe file structure is established: the process recipe file is composed of one or more step-level data units, each step-level data unit includes several process recipe parameters, and a process recipe parameter corresponds to the smallest information unit of the process recipe at the production machine end . The process formula file of the present application also expands the support of parameter-level operations. The process formula file with the same file structure can use the unified conversion function and the unique identification information of each layer of information (the first identification mark, the second identification mark and the third identification mark). Identification mark), can expand the file parameters and store each parameter separately, and can support identification to any parameter. The combination of the above expansion information and the file-level attributes can realize the inductive management of parameter bases across files.

This application provides a data management system and management method. It provides a standard file structure after transcribing the process formula file at the production machine end. The process formula file is accurately decomposed into parameter-level data through fixed functions, and the process formula is solved. While the files are visualized remotely, accurate and efficient induction and arrangement of process formula parameters are realized. The technical solution of this application is compatible with the current old system, and the support for parameter-level operations has the characteristics of oriented to big data-level functions. It provides transcription functions, which can be freely converted between file-level operations and parameter-level operations, and can be called in real time as needed. That is, high precision and high efficiency.

The application provides a data management system and a management method, which can take into account both file-level operations and parameter-level operations, and realize precise management of all process formula parameters in process formula files.

To sum up, after reading the content of this application, those skilled in the art can understand that the content of the foregoing application may be presented by way of example only, and may not be limiting. Although not explicitly stated herein, those skilled in the art will understand that this application is intended to cover various reasonable changes, improvements and modifications to the embodiments. These changes, improvements and modifications are within the spirit and scope of the exemplary embodiments of the present application.

It should be understood that the term "and/or" used in this embodiment includes any or all combinations of one or more of the associated listed items. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present.

Similarly, it will be understood that when an element such as a layer, region or substrate is referred to as being "on" another element, it can be directly on the other element or intervening elements may also be present. In contrast, the term "directly" means that there are no intervening elements. It should also be understood that the terms "comprising", "comprising", "comprising" or "comprising", when used in this application document, indicate the presence of the described features, integers, steps, operations, elements and/or components, But it does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or their groups.

It will also be understood that although the terms first, second, third etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. Thus, a first element in some embodiments could be termed a second element in other embodiments without departing from the teachings of the present application. The same reference numerals or the same reference signs denote the same elements throughout the specification.

Claims (18)

1. A data management system, characterized in that, comprising: Memory, used to store process recipe files obtained from the production machine; The equipment automatic control unit, the following steps are performed when the equipment automatic control unit works: Obtain the process recipe file of the production machine end from at least one production machine end; Setting a first identification mark for identifying the process recipe file for each of the process recipe files; Decomposing each of the process recipe files into several step-level data units, and setting a second identification mark for identifying the several step-level data units for the several step-level data units, wherein the first identification mark used together with said second identification mark to identify said step-level data unit; Decomposing each of the step-level data units into a number of process recipe parameters, and setting a third identification mark for identifying the number of process recipe parameters for the number of process recipe parameters, wherein the first identification mark, the The second identification mark and the third identification mark are jointly used to identify the process recipe parameters. 2. The data management system according to claim 1, wherein when the automatic equipment control unit works, it also: establishes file-level attribute information for the process recipe file, and the file-level attribute information includes editable attributes and Uneditable attributes, wherein the editable attributes include status and scope of application, and the uneditable attributes include name, production machine to which it belongs, current version and update time. 3. The data management system according to claim 1, wherein, when the automatic control unit of the equipment works, it also: establishes step-level attribute information for the step-level data unit, and the step-level attribute information includes position information and type information. 4. The data management system according to claim 1, wherein, when the equipment automatic control unit is working, it also: establishes parameter attribute information for the process recipe parameters, and the parameter attribute information includes position information and identifiable information . 5. The data management system according to claim 1, wherein the first identification mark includes information on the production machine side; the second identification mark includes step information of the step-level data unit; The third identification mark includes the name of the process recipe parameter. 6. The data management system according to claim 1, wherein the automatic control unit of the equipment further comprises: an automatic identification model, which is established through artificial intelligence deep learning, and the automatic identification model is used to identify the parameters of the process formula The step-level data unit to which it belongs and the process recipe file to which the step-level data unit belongs. 7. The data management system according to claim 1, wherein the process recipe file is a folder, the step-level data unit is a subfolder after the folder is opened, and the process recipe parameter is located in in the subfolder. 8. The data management system according to claim 1, wherein the process recipe file is a hyperlink, the step-level data unit is a sub-hyperlink after the hyperlink is expanded, and the process recipe parameter is located at in the child hyperlink. 9. The data management system according to claim 1, wherein the process recipe file is a table, the step-level data unit is a sub-table of the table, and the process recipe parameters are located in the sub-table . 10. A data management method, characterized in that, comprising: performing the following steps by an automatic equipment control unit: Obtain the process recipe file of the production machine end from at least one production machine end; Setting a first identification mark for identifying the process recipe file for each of the process recipe files; Decomposing each of the process recipe files into several step-level data units, and setting a second identification mark for identifying the several step-level data units for the several step-level data units, wherein the first identification mark used together with said second identification mark to identify said step-level data unit; Decomposing each of the step-level data units into a number of process recipe parameters, and setting a third identification mark for identifying the number of process recipe parameters for the number of process recipe parameters, wherein the first identification mark, the The second identification mark and the third identification mark are jointly used to identify the process recipe parameters. 11. The data management method according to claim 10, further comprising: establishing file-level attribute information for the process recipe file, the file-level attribute information including editable attributes and non-editable attributes, wherein the The editable attributes include status and scope of application, and the non-editable attributes include name, production machine to which it belongs, current version and update time. 12. The data management method according to claim 10, further comprising: establishing step-level attribute information for the step-level data unit, the step-level attribute information including position information and type information. 13. The data management method according to claim 10, further comprising: establishing parameter attribute information for the process recipe parameters, the parameter attribute information including location information and identifiable information. 14. The data management method according to claim 10, wherein the first identification mark includes information on the production machine side; the second identification mark includes step information of the step-level data unit; The third identification mark includes the name of the process recipe parameter. 15. The data management method according to claim 10, further comprising: establishing an automatic identification model for the automatic control unit of the equipment through artificial intelligence deep learning, and the automatic identification model is used to identify the parameters of the process recipe The step-level data unit to which it belongs and the process recipe file to which the step-level data unit belongs. 16. The data management method according to claim 10, wherein the process recipe file is a folder, the step-level data unit is a subfolder after the folder is opened, and the process recipe parameter is located in in the subfolder. 17. The data management method according to claim 10, wherein the process recipe file is a hyperlink, the step-level data unit is a sub-hyperlink after the hyperlink is expanded, and the process recipe parameter is located at in the child hyperlink. 18. The data management method according to claim 10, wherein the process recipe file is a table, the step-level data unit is a sub-table of the table, and the process recipe parameters are located in the sub-table .

Images