Lithography tool performance evaluation system based on key indicator algorithm and its application method

Abstract

The present invention provides a photolithography machine performance evaluation system based on a key indicator algorithm and an application method thereof, providing production source data of the photolithography machine, processing the production source data through a data processing module to obtain initial version data; obtaining multiple key indicators based on the initial version data; and developing a system based on multiple key indicators, including: performing interface design and its logical implementation; performing interface project design; creating a service program and implementing it; creating a host and creating an agent; and ranking the performance utilization among photolithography machines. The present invention designs a key indicator algorithm to better evaluate the utilization efficiency of the photolithography machine, and systematizes the algorithm design, predicts the incoming goods situation or the status of the machine based on historical and current data, and monitors the current running goods situation of the machine in real time, which is conducive to obtaining the machine with the best performance and obtaining the performance utilization ranking among the machines, so that the performance evaluation of the photolithography machine can be accurate.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and in particular to a photolithography machine performance evaluation system based on a key indicator algorithm and an application method thereof. Background Art

[0002] First, regarding the evaluation of the utilization rate of lithography machines, the usual practice is for technicians to check the idle status of each machine on a static report based on established historical data. This method can only remain at a rough evaluation. In fact, the actual time for LOTs to enter and exit the machine, the number of LOT carriers per card, the number of times the lithography board is switched, downtime, etc. are all factors that affect the evaluation of the machine's operating efficiency; at the same time, existing methods lack real-time monitoring of such data. For the technology-intensive semiconductor manufacturing industry, the performance evaluation of lithography machines should be more precise, more diversified, and more instantiated.

[0003] Secondly, there is currently no relevant research on comparative analysis of the performance utilization between lithography tools, which would be more scientific and accurate than the evaluation of the tools themselves. Summary of the invention

[0004] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a lithography machine performance evaluation system based on a key indicator algorithm and an application method thereof, so as to solve the problem that the evaluation method of lithography machine utilization in the prior art lacks real-time monitoring of such data and the performance evaluation of the lithography machine cannot be accurately achieved.

[0005] To achieve the above-mentioned and other related purposes, the present invention provides a photolithography tool performance evaluation system based on a key indicator algorithm, which at least includes:

[0006] Production source data of a photolithography machine; a data processing module for processing the production source data; initial version data formed by processing the production source data by the data processing module;

[0007] A plurality of key indicators obtained based on the initial version data;

[0008] A system based on the plurality of key indicators.

[0009] Preferably, the production source data includes at least: lot quantity, product name, product type, site information, operation step sequence number, product grade, machine entry time, machine exit time, operation start time, operation end time, historical operation time and current operation time.

[0010] Preferably, the multiple key indicators include at least: machine operation time, machine idle time, machine utilization efficiency, actual lot number, lot movement amount, number of continuous small lot operations, number of photolithography board switching times, abnormal lot, average photolithography board movement amount, and non-continuous operation lot.

[0011] Preferably, the machine operation time is the time within the factory date, the historical date of the lithography machine and the time when the lithography machine is running up to the current time.

[0012] Preferably, the machine idle time is the idle time of the lithography machine within the factory date, the historical date of the lithography machine and up to the current time.

[0013] Preferably, the machine usage efficiency is the proportion of time that the machine is in a running state when the product type is a product piece within the factory date or up to the current date.

[0014] Preferably, the actual lot quantity is the lot quantity of product sheets and engineering sheets under different machines within the factory date or up to the current date.

[0015] Preferably, the Lot movement amount is the ratio of the actual wafer movement amount generated under each machine to the actual Lot number, one step of wafer movement is one movement amount, and the actual wafer movement amount is the sum of each movement amount.

[0016] Preferably, the system based on the multiple key indicators at least includes: an interface, an interface item, a service program, a host and an agent.

[0017] Preferably, the interface includes a historical date option, a current date option, result display items based on the multiple key indicators, and drill-down prompt information based on the result display items.

[0018] Preferably, the interface project is a program library, which contains the multiple key indicators, the drilling information corresponding to the multiple key indicators, and the declaration of the service method of the system.

[0019] Preferably, the service program is an implementation library of the interface program, which is used to reference the interface program and pass in the multiple key indicators to specifically implement the interface program.

[0020] Preferably, the host is a console application created based on the Windows platform, which is used to reference the interface project and the service program, implement server and port configuration, and start on the server.

[0021] Preferably, the agent is used to add configuration to enable hosting and expose the production source data, and start the system.

[0022] The present invention also provides an application method of a photolithography machine performance evaluation system based on a key indicator algorithm, the method comprising at least the following steps:

[0023] Step 1: providing production source data of a lithography machine, and processing the production source data through a data processing module to obtain initial version data;

[0024] Step 2: obtaining multiple key indicators based on the initial version data;

[0025] Step 3: Develop the system based on the multiple key indicators, including: (1) designing the interface and its logical implementation; (2) designing the interface project; (3) creating a service program and implementing it; (4) creating a host; and (5) creating an agent.

[0026] Step 4: ranking the performance utilization of the lithography machines.

[0027] Preferably, the method in which the data processing module in step 1 processes the production source data is: extracting, converting, and loading the production source data from the source end to the destination end.

[0028] Preferably, the multiple key indicators in step 2 include: machine operation time, machine idle time, machine utilization efficiency, actual lot number, lot movement amount, number of continuous small lot operations, number of photolithography plate switching times, abnormal lot, average photolithography plate movement amount, and non-continuous operation lot.

[0029] Preferably, the machine operation time obtained based on the initial version data in step 2 is the machine operation time within the factory date, the historical date of the lithography machine and the time when the lithography machine is out of stock up to the current time, and the algorithm for calculating the machine operation time uptime is: i is a state parameter; σ is a state set, which at least includes: backup state, running state, recycling state, idle state, pause state, mid is the machine identifier; t and T are time periods.

[0030] Preferably, the idle time of the machine obtained based on the initial version data in step 2 is the idle time of the lithography machine within the factory date, the historical date of the lithography machine and the current time, and the algorithm of the idle time of the machine is: i is the state parameter; mid is the machine identifier; t and T are the time periods.

[0031] Preferably, the machine usage efficiency obtained based on the initial version data in step 2 is the proportion of time that the machine is in a running state within the factory date or up to the current date when the product type is a product piece, and the algorithm of the machine usage efficiency UTI is: Where ξ is a valid evaluation set, and its valid values ​​include at least: backup status, running status, and recycling status.

[0032] Preferably, the actual lot quantity obtained based on the initial version data in step 2 is the lot quantity of product sheets and engineering sheets under different machines within the factory date or up to the current date, and the algorithm for calculating the actual lot quantity lot_cnt is: The product type attributes of lot are engineering piece and product piece.

[0033] Preferably, the Lot movement amount obtained based on the initial version data in step 2 is the ratio of the actual wafer movement amount generated under each machine to the actual Lot quantity, one step of wafer movement is one movement amount, the actual wafer movement amount is the sum of each movement amount, and the algorithm of the Lot movement amount lotsize is: One step of movement of a wafer is called a step, and c is the number of lots.

[0034] Preferably, the algorithm for obtaining the number of consecutive small lot operations smallsize_cnt based on the initial version data in step 2 is: The number of wafers in each lot is less than or equal to 10, and the lots are continuous. If there are more than or equal to 3 consecutive lots, it is counted as 1 time; k is the lot number, cnt represents the count, and start is sorted by the exit time. is a set of continuous lots, k++ is greater than or equal to 3, and qty is the number of lots and less than 10.

[0035] Preferably, the number of photomask switching MASK_CNT obtained in step 2 based on the initial version data indicates that when the lithography machine operates different products, the required photomasks are different, and if the types of two consecutive photomasks are different, it is counted as 1 time, and the objective function is the total number of switching times within a cycle.

[0036] Preferably, the abnormal lot ABNORMAL_RT_CNT obtained based on the initial version data in step 2 is expressed as: The calculation rule of the running time runtime is: the lots under the machine are arranged in ascending order according to the time of entering the machine. When the time of entering the machine of a lot is greater than the time of leaving the machine of the previous lot, the runtime is the time of leaving the machine of the previous lot minus the time of entering the machine of this lot; when the time of entering the machine of a lot is less than the time of leaving the machine of the previous lot, the runtime is the time of leaving the machine of this lot minus the time of leaving the machine of the previous lot. The ABNORMAL_RT_CNT time is the number of lots with a runtime time greater than 20 minutes, and the objective function is the total number of abnormal lots defined within the cycle.

[0037] Preferably, the average movement amount AVG_MASK_MOVE of the photomask obtained in step 2 based on the initial version data is expressed as: the ratio of all machine movement amounts to MASK_CNT within the factory date or up to the current date, and the objective function is the ratio of the actual movement amount to MASK_CNT.

[0038] Preferably, the non-continuous operation Lot NO_CONTINUE_CNT obtained based on the initial version data in step 2 is expressed as follows: all lots entering the machine are arranged in order, and the difference between the exit time of the current lot and the entry time of the fourth lot in the continuous order is calculated. If the difference is greater than 15 seconds, it is counted as 1, and the target value is the total number of counts.

[0039] Preferably, in the interface design performed in step three, the display of the interface includes historical date options, current date options, result display items based on the multiple key indicators, and drill-down prompt information based on the result display items.

[0040] Preferably, the interface project in step three is a program library, and the program library contains the multiple key indicators, the drilling information corresponding to the multiple key indicators, and the declaration of the service method of the system.

[0041] Preferably, the method for creating and implementing the service program in step three is: creating the key indicator class library project, referencing the interface project, passing in the corresponding key indicator, and implementing the logic of the interface project.

[0042] Preferably, in step three, the host references the interface project, implements server and port configuration and starts on the server.

[0043] Preferably, the agent in step three is used to add configuration to implement hosting and publicizing of the production source data, and to start the system.

[0044] Preferably, the performance utilization ranking among the lithography tools in step 4 is aimed at selecting the best tool with the best performance, and the criterion layer is based on the designed 10 KPIs as criteria.

[0045] Preferably, the method for ranking the efficiency utilization of lithography machines in step 4 includes: 1. constructing a judgment matrix, wherein the criterion layer compares 10 different factors in pairs and calculates the ratio as a ij And put it into the matrix A with 10 rows and 10 columns to judge it as an inverse matrix or a consistency matrix; 2. Weight calculation and consistency check; 3. Candidate layer coefficient setting.

[0046] As described above, the lithography machine performance evaluation system based on the key indicator algorithm of the present invention and the application method thereof have the following beneficial effects: the present invention designs a key indicator algorithm to better evaluate the utilization efficiency of the lithography machine, and systematizes the algorithm design, predicts the machine arrival situation or machine status based on historical and current data, and monitors the current running situation of the machine in real time, which is conducive to obtaining the machine with the best performance and obtaining the performance utilization ranking among the machines, so as to make the performance evaluation of the lithography machine more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It shows the application method of the photolithography tool performance evaluation system based on the key indicator algorithm of the present invention. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0049] See also Figure 1 It should be noted that the illustrations provided in this embodiment are only used to illustrate the basic concept of the present invention in a schematic manner, and the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0050] The present invention provides a photolithography machine performance evaluation system based on a key indicator algorithm, which at least includes: production source data of the photolithography machine; a data processing module for processing the production source data; initial version data formed after the production source data is processed by the data processing module; multiple key indicators obtained based on the initial version data; and a system based on the multiple key indicators.

[0051] The present invention further comprises: the production source data in this embodiment includes at least: lot quantity, product name, product type, site information, operation step sequence number, product grade, machine entry time, machine exit time, operation start time, operation end time, historical operation time and current operation time.

[0052] The present invention further states that the multiple key indicators in this embodiment include at least: machine operation time, machine idle time, machine utilization efficiency, actual lot number, lot movement, number of continuous small lot operations, number of photolithography plate switching, abnormal lot, average movement of photolithography plates, and non-continuous operation lot. The present invention further states that the machine operation time in this embodiment is the time when the photolithography machine is out of stock within the factory date, the historical date of the photolithography machine, and up to the current time. The machine idle time in this embodiment is the idle time of the photolithography machine within the factory date, the historical date of the photolithography machine, and up to the current time. The machine utilization efficiency in this embodiment is the proportion of time that the product type is product pieces and the machine is in the out of stock state within the factory date or up to the current date. The actual lot number in this embodiment is the number of lots of product types of product pieces and engineering pieces under different machines within the factory date or up to the current date. The Lot movement amount in this embodiment is the ratio of the actual wafer movement amount generated by each machine to the actual Lot number, one step of wafer movement is one movement amount, and the actual wafer movement amount is the sum of each movement amount.

[0053] The present invention further provides that the system based on the multiple key indicators in this embodiment at least includes: an interface, an interface project, a service program, a host and an agent. The interface in this embodiment includes a historical date option, a current date option, a result display item based on the multiple key indicators, and a drill prompt information based on the result display item. The interface project is a program library, and the program library contains the multiple key indicators, the drill information corresponding to the multiple key indicators, and a declaration of the service method of the system. The service program is an implementation library of the interface program, which is used to reference the interface program and pass in the multiple key indicators to specifically implement the interface program. The host is a console application created based on the Windows platform, which is used to reference the interface project and the service program, implement server and port configuration, and start it on the server. The agent is used to add configuration to implement hosting and publicize the production source data, and start the system.

[0054] The present invention also provides an application method of the photolithography machine performance evaluation system based on the key indicator algorithm, such as Figure 1 As shown, Figure 1 The present invention shows an application method of the photolithography machine performance evaluation system based on the key indicator algorithm, which method at least includes the following steps:

[0055] Step 1: Provide the production source data of the lithography machine, and obtain the initial version data after processing the production source data through the data processing module; further, the method for the data processing module in step 1 of this embodiment to process the production source data is: extracting, converting, and loading the production source data from the source end to the destination end. The production source data in this embodiment at least includes: lot quantity, product name, product type, site information, operation step sequence number, product level, machine entry time, machine exit time, start operation time, end operation time, historical operation time, and current operation time. This step one performs data ETL work (the process of extracting, converting, and loading data from the source to the destination): the original data is processed according to the corresponding rules and requirements (the required data includes: Lot important attribute information and production information: quantity, product name, product type, site information, operation step sequence number, product grade, machine entry time, machine exit time, operation start time, operation end time, historical and current operation specific time, etc.); based on the background server, the source data is cleaned using database knowledge and complex data analysis, and the initial version of the data after processing is stored in the corresponding data warehouse.

[0056] Step 2, based on the initial version data, multiple key indicators are obtained; based on the initial version data in the database, the following key algorithm indicator design and platform design are performed with the help of tools such as ORACLE, PL / SQL and C#. The multiple key indicators in step 2 of this embodiment include: machine operation time, machine idle time, machine utilization efficiency, actual lot number, lot movement amount, number of continuous small lot operations, number of photolithography plate switching, abnormal lot, average photolithography plate movement amount, and non-continuous operation lot.

[0057] In step 2 of this embodiment, the machine operation time obtained based on the initial version data is within the factory date (for example, from 8:00 am to 8:00 am the next day is a factory date), the machine operation time within the historical date of the lithography machine and the time when the lithography machine is running up to the current time, and the algorithm for calculating the machine operation time uptime is: i is a state parameter; σ is a state set, which at least includes: backup state, running state, recycling state, idle state, hold state (pause); mid is the machine identifier; t and T are time periods.

[0058] In step 2 of this embodiment, the idle time of the machine obtained based on the initial version data is the idle time of the lithography machine within the factory date, the historical date of the lithography machine and the current time, and the algorithm of the idle time of the machine is: i is the state parameter; mid is the machine identifier; t and T are the time periods.

[0059] In step 2 of this embodiment, the machine usage efficiency obtained based on the initial version data is the proportion of time that the machine is in a running state within the factory date or up to the current date, and the algorithm of the machine usage efficiency UTI is: Where ξ is a valid evaluation set, and its valid values ​​include at least: backup status, running status, and recycling status.

[0060] The actual lot quantity obtained based on the initial version data in step 2 of this embodiment is the lot quantity of product sheets and engineering sheets in different machines within the factory date or up to the current date, and the algorithm for calculating the actual lot quantity lot_cnt is: The product type attributes of lot are Eng (engineering lot) and Prod (product lot).

[0061] In step 2 of this embodiment, the Lot movement amount obtained based on the initial version data is the ratio of the actual wafer movement amount generated by each machine to the actual Lot quantity. One wafer movement step is one movement amount. The actual wafer movement amount is the sum of each movement amount. The algorithm of the Lot movement amount lotsize is: One step of movement of a wafer is called a step, and c is the number of lots.

[0062] In step 2 of this embodiment, the algorithm for obtaining the number of consecutive small lot operations, smallsize_cnt, based on the initial version data is: The number of wafers in each lot is less than or equal to 10, and the lots are continuous. If there are more than or equal to 3 consecutive lots, it is counted as 1 time; k is the lot number, cnt represents the count, and start is sorted according to the jobout exit time. is a set of continuous lots, k++ is greater than or equal to 3, and qty is the number of lots and less than 10.

[0063] The number of photomask switching MASK_CNT obtained based on the initial version data in step 2 of this embodiment indicates that when the lithography machine operates different products, the required photomasks are different, and if the types of two consecutive photomasks are different, it is counted as 1 time, and the objective function is the total number of switching times within the cycle.

[0064] The abnormal lot ABNORMAL_RT_CNT obtained based on the initial version data in step 2 of this embodiment is expressed as: The calculation rule of the running time runtime is: the lots under the machine are arranged in ascending order according to the time of entering the machine. When the time of entering the machine of a lot is greater than the time of leaving the machine of the previous lot, the runtime is the time of leaving the machine of the previous lot minus the time of entering the machine of this lot; when the time of entering the machine of a lot is less than the time of leaving the machine of the previous lot, the runtime is the time of leaving the machine of this lot minus the time of leaving the machine of the previous lot. The ABNORMAL_RT_CNT time is the number of lots with a runtime time greater than 20 minutes, and the objective function is the total number of abnormal lots defined within the cycle.

[0065] In step 2 of this embodiment, the average movement amount AVG_MASK_MOVE of the photomask obtained based on the initial version data is expressed as: the ratio of all machine movement amounts to MASK_CNT within the factory date or up to the current date, and the objective function is the ratio of the actual movement amount to MASK_CNT.

[0066] The non-continuous operation Lot NO_CONTINUE_CNT obtained based on the initial version data in step 2 of this embodiment is expressed as follows: all the lots entering the machine are arranged in order, and the difference between the exit time of the current lot and the entry time of the fourth lot in the continuous order is calculated. If the difference is greater than 15 seconds, it is counted as 1 time, and the target value is the total number of times counted.

[0067] Step 3: Develop the system based on the multiple key indicators, that is, build the system framework based on the WCF framework system and adopt a C / S three-tier structure design.

[0068] System development specifically includes:

[0069] (1) Conduct interface design and its logical implementation; The present invention further conducts interface design in step three of this embodiment, and the display of the interface includes historical date options, current date options, result display items based on the multiple key indicators, and drill-down prompt information based on the result display items. The interface design and logical implementation in step (1) mainly cover the integrated design of important information query and important algorithm indicator calculation. According to any combination of different capabilities and machines, historical date and current date options are added, and result displays under different conditions are given. According to each important algorithm indicator, its display value increases the drill-down prompt setting; in the context of different needs, users can drill down to detailed information under each algorithm indicator according to actual needs.

[0070] (2) Designing an interface project; In the present invention, the interface project in step 3 of this embodiment is a program library, and the program library contains the multiple key indicators, the drilling information corresponding to the multiple key indicators, and the declaration of the service methods of the system. The interface design in step (2) creates the key indicator interface, designs the program library to contain the interface required by the system, which involves the key indicator algorithm value and the detailed information of the drill-down under the indicator value, and returns it in the form of DataSet, and declares all the service methods involved in the system based on the data contract.

[0071] (3) Creating and implementing a service program; Further, the present invention provides that the method for creating and implementing a service program in step three of this embodiment is as follows: creating the key indicator class library project, referencing the interface project, passing in the corresponding key indicator, and logically implementing the interface project.

[0072] (4) Create a host; further, in the present invention, the host in step 3 of this embodiment references the interface project, implements server and port configuration and starts on the server. (5) Create an agent; further, in the present invention, the agent in step 3 of this embodiment is used to add configuration to implement hosting and publicize the production source data, and start the system.

[0073] Step 4, ranking the performance utilization among the lithography machines. The present invention further, in step 4 of this embodiment, the performance utilization ranking among the lithography machines is aimed at selecting the best machine with the best performance, and the criterion layer is based on 10 designed KPIs as criteria. Here, the goal is to select the best machine (best performance), and the criterion layer is based on 10 designed KPIs as criteria, assuming that 4 machines with the same capability are selected. The method for ranking the performance utilization among the lithography machines includes:

[0074] 1. Judgment matrix construction: the criterion layer is to compare 10 different factors in pairs and calculate the ratio as a ij And put it into a 10-row 10-column matrix A to be judged as an inverse matrix or a consistency matrix; Compared with the traditional weight allocation, the criterion layer here is to compare the 10 different factors in pairs (the ratio of the factors can refer to the scale table), and the ratio is calculated as a ij And put it into a matrix of 10 rows and 10 columns, this matrix is ​​A. As shown in Table 1, Table 1 Scaling Setting Table (Taking UPTIME and IDLETIME as an example)

[0075]

[0076] The mathematical model of judgment matrix construction is: if A=(a ij ) m×n Then it needs to meet the following requirements: (i) a ij >; (ii) That is, the two are mutually inverse matrices. i, j, k = 1, 2, 3…10, which is the consistency matrix.

[0077] 2. Weight calculation and consistency test: Assume that the above judgment matrix complies with the consistency principle and is completely consistent. Any column vector of the consistent matrix is ​​an eigenvector. That is, by normalizing the eigenvector corresponding to the maximum eigenvalue of the judgment matrix, the weight of each factor can be obtained as ω = (ω 1 ,...ω 10 ) T The weight coefficient corresponding to the eigenvector is based on the consistency of the judgment matrix. AHP requires reverse verification of whether the judgment matrix is ​​truly consistent.

[0078] (1) Aω=λ max ω

[0079] (2) The consistency test judgment value CR can be obtained from the above two formulas; the RI value can be obtained by looking up the table (when n=10, CR=1.49); if the CR value is less than 0.1, the consistency test is passed, otherwise it is necessary to readjust the values ​​in the judgment matrix, that is, the scale settings between factors. If the consistency test is finally passed, the weights corresponding to the 10 important KPIs are ω 1 ,...ω 10 .

[0080] 3. Candidate layer coefficient setting.

[0081] The rule for setting the candidate layer coefficients becomes the weight of the machine under a certain factor. The conventional algorithm is to construct 10 comparison matrices, and the eigenvector corresponding to the maximum eigenvalue of each matrix is ​​the weight ratio of the four machines under different factors. This article adopts a more concise and rigorous method: the KPI ratio between all machines is used as the weight ratio of the machine under a certain factor.

[0082] The weight ratio of each factor is as follows:

[0083] UPTIME:ω 1，ut :ω 2，ut :ω 3，ut :ω 4，ut

[0084] IDLETTIME:ω 1，it :ω 2，it :ω 3，it :ω 4，it

[0085]

[0086] NO_CONTINUE_CNT: ω 1，no_c_cnt :ω 2，no_c_cnt :ω 3，no_c_cnt :ω 4，no_c_cnt

[0087] The final weight of machine m is: ω 1 ×ω m，ut，... +ω 10，no_c_cnt ; After finding the final weights of each machine, the optimal target machine can be selected.

[0088] The present invention designs an evaluation index for the efficiency of photolithography machine utilization; the designed system index is tooled and systematized; the designed system can meet the display of historical data and current data under different combinations of conditions; the specific values ​​of corresponding attributes can be drilled under the algorithm results, and abnormal values ​​of detailed information can be marked. Based on important indicators, the AHP algorithm is used to find the optimal machine for efficiency utilization and its ranking. The present invention can monitor the running status of photolithography machines through the current index data and indicators displayed by the system, and provide reasonable planning control and delivery plans. Based on the established historical data, reasonable predictions can be made for key indicators; using the machine with the best efficiency as a benchmark, it can help it optimize other machines to the greatest extent, improve machine utilization, and maximize the factory's production capacity.

[0089] In summary, the present invention designs an algorithm of key indicators to better evaluate the efficiency of lithography machine utilization, and systematizes the algorithm design, predicts the machine supply situation or machine status based on historical and current data, and monitors the current running status of the machine in real time, which is conducive to finding the machine with the best performance and obtaining the efficiency utilization ranking between each machine, so that the performance evaluation of the lithography machine can be accurate. Therefore, the present invention effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0090] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A photolithography tool performance evaluation system based on a key indicator algorithm, characterized in that: At least: Production source data of lithography machines; A data processing module for processing the production source data; Initial version data formed by processing the production source data through the data processing module; A plurality of key indicators obtained based on the initial version data; the plurality of key indicators at least include: machine operation time, machine idle time, machine utilization efficiency, actual lot number, lot movement amount, number of continuous small lot operations, number of photomask switching times, abnormal lot, average photomask movement amount, and non-continuous operation lot; Based on the multiple key indicators, a system is developed to obtain a photolithography tool performance evaluation system; The system based on the plurality of key indicators comprises at least: an interface, an interface item, a service program, a host and an agent; The interface project is a program library, which contains the multiple key indicators, the drilling information corresponding to the multiple key indicators, and the declaration of the service method of the system; The service program is an implementation library of the interface program, which is used to reference the interface program and pass in the multiple key indicators to specifically implement the interface program; The designed system can meet the display of historical data and current data under different combinations of conditions, predict the machine's incoming goods or machine status based on historical and current data, and monitor the machine's current running goods status in real time, and use the AHP algorithm to find the most efficient machine and its ranking.

2. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The production source data includes at least: lot quantity, product name, product type, site information, operation step sequence number, product grade, machine entry time, machine exit time, operation start time, operation end time, historical operation time and current operation time.

3. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The machine operation time refers to the time within the factory date, the historical date of the lithography machine and the time when the lithography machine is running up to the current time.

4. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The idle time of the machine is the idle time of the lithography machine in the factory date, the historical date of the lithography machine and the current time, and the algorithm of the idle time of the machine is: i is the state parameter, σ is the state set, and I is the upper limit of the state parameter; mid is the machine identification; t and T are the time periods; stu i∈σ,end is the end time of idle state, stu i∈σ,start The idle state start time.

5. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The algorithm of the machine usage efficiency UTI is: Where ξ is a valid evaluation set, in which valid values ​​include at least: backup status, running status, and recycling status; j is a valid evaluation item, J is the last valid evaluation item; t and T are time periods; com j∈ξ,end is the effective evaluation item value at the end time, com j∈ξ,start is the effective evaluation item value at the start time; the machine utilization efficiency obtained based on the initial version data design is the utilization efficiency of the product type being the product piece within the factory date or up to the current date.

6. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The actual lot quantity refers to the lot quantity of product sheets and engineering sheets under different machines within the factory date or up to the current date.

7. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The lot movement amount is the ratio of the actual wafer movement amount generated by each machine to the actual lot number. One step of wafer movement is one movement amount. The actual wafer movement amount is the sum of each movement amount. The algorithm of the lot movement amount lotsize is: One step of movement of a wafer is called a step, c is the lot number; t and T are the time periods; C is the upper limit of the lot number, and lotqty is the number of wafers contained in a lot.

8. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The algorithm for calculating the number of consecutive small lot operations smallsize_cnt is: The number of wafers in each lot is less than or equal to 10, and the lots are continuous. If there are more than or equal to 3 consecutive lots, it is counted as 1 time; k is the lot number, cnt represents the count, and sort is sorted by the exit time. is a set of continuous lots, k++ is greater than or equal to 3, and qty is the number of lots and less than 10.

9. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The number of photomask switching is expressed as the number of photomasks required when the photolithography machine operates different products. If two consecutive photomasks are of different types, it is counted as 1 time. The objective function is the total number of switching times within a cycle.

10. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The abnormal lot is expressed as: the calculation rule of the running time is: arrange the lots under the machine in ascending order according to the time of entering the machine. When the time of entering the machine of a lot is greater than the time of leaving the machine of the previous lot, the running time is the time of leaving the machine of the previous lot minus the time of entering the machine of this lot; when the time of entering the machine of a lot is less than the time of leaving the machine of the previous lot, the running time is the time of leaving the machine of this lot minus the time of leaving the machine of the previous lot; the abnormal lot time is the number of lots whose running time is greater than 20 minutes, and the objective function is the total number of abnormal lots defined within the cycle.

11. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The average movement of the photoresist plate is expressed as: the ratio of all movement of the machine to the number of photoresist plate switching within the factory date or up to the current date.

12. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The non-continuous operation Lot is represented as follows: all the lots entering the machine are arranged in order, and the difference between the time when the current lot leaves the machine and the time when the fourth lot in the continuous order enters the machine is calculated. If the difference is greater than 15 seconds, it is counted as 1 time, and the target value is the total number of times counted.

13. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The interface includes a historical date option, a current date option, result display items based on the multiple key indicators, and drill prompt information based on the result display items.

14. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 1, characterized in that: The host is a console application created based on the Windows platform, which is used to reference the interface project and the service program, implement server and port configuration, and start on the server.

15. The photolithography tool performance evaluation system based on key indicator algorithm according to claim 14, characterized in that: The agent is used to add configuration to host and expose the production source data, and start the system.

16. The application method of the photolithography machine performance evaluation system based on the key indicator algorithm according to any one of claims 1 to 15, characterized in that: The method comprises at least the following steps: Step 1: providing production source data of a lithography machine, and processing the production source data through a data processing module to obtain initial version data; Step 2: designing and obtaining multiple key indicators based on the initial version data; Step 3: Develop the system based on the multiple key indicators, including: (1) designing the interface and implementing its logic; (2) designing the interface project; (3) creating a service program and implementing it; (4) creating a host; (5) creating an agent; Step 4: ranking the performance utilization of the lithography machines.

17. The application method of the photolithography tool performance evaluation system based on the key indicator algorithm according to claim 16, characterized in that: The method for the data processing module in step 1 to process the production source data is: extracting, converting, and loading the production source data from the source end to the destination end.

18. The application method of the photolithography tool performance evaluation system based on the key indicator algorithm according to claim 16, characterized in that: The efficiency ranking among the lithography machines described in step 4 aims to select the best machine with the best efficiency. The criterion layer uses the designed 10 KPIs as the criteria. The specific methods include:

1. Judgment matrix construction. The criterion layer compares the 10 different factors in pairs and calculates the ratio as a ij And put it into the matrix A with 10 rows and 10 columns to judge it as an inverse matrix or a consistency matrix; 2. Weight calculation and consistency check; 3. Candidate layer coefficient setting.

Images