A method, apparatus and equipment for monitoring the consumption of semiconductor consumables.

Abstract

This application provides a method, apparatus, and device for monitoring the consumption of semiconductor consumables. The method includes: acquiring consumable data related to the consumption of a target semiconductor consumable; determining a consumable analysis model corresponding to the acquired consumable data based on the acquired consumable data and the acquisition method, and a pre-established consumable analysis model corresponding to different consumable data and acquisition methods; and performing corresponding calculations based on the acquired consumable data using the determined consumable analysis model to obtain the actual consumption of the target semiconductor consumable. Using the monitoring method provided in this application, the actual consumption of different types of semiconductor consumables can be automatically and accurately determined.

Description

Technical Field

[0001] This application relates to the field of semiconductor materials technology, and in particular to a method, apparatus and equipment for monitoring the consumption of semiconductor consumables. Background Technology

[0002] IDM (Indirect Materials) refers to general consumable materials consumed by enterprises to organize, manage, and ensure the normal operation of production. In other words, it refers to various materials that contribute to the formation of goods or facilitate production during the production and processing process, but do not constitute the physical product.

[0003] In the semiconductor wafer manufacturing process, the consumption of indirect materials is essential. Commonly used indirect materials include chemicals, special gases, slurries, photoresists, and targets. In actual production, the calculation of the consumption of indirect materials is of great significance for the formulation of material requirements planning and the preparation of annual cost budgets.

[0004] However, in existing technologies, the actual consumption range of indirect materials is wide, and the combination of machines and materials results in numerous consumption points. It is difficult to manually calculate the actual consumption and occurrence points, and the differences are large. Currently, the commonly used method for detecting the actual consumption of indirect materials is to install flow meters at the pipeline to measure the consumption of materials. However, the pipeline usually connects multiple machines, making it impossible to accurately measure the actual consumption at each machine end, and the calculation results of the actual consumption have a large error. Summary of the Invention

[0005] This application provides a method, apparatus, and equipment for monitoring the consumption of semiconductor consumables, which solves the problem of inaccurate monitoring results of the actual consumption of semiconductor consumables in the prior art.

[0006] This application provides a method for monitoring the consumption of semiconductor consumables, including:

[0007] Acquire consumable data related to the consumption of the target semiconductor consumable;

[0008] Based on the acquired consumable data and the acquisition method, and the pre-established consumable analysis model corresponding to different consumable data and acquisition methods, determine the consumable analysis model corresponding to the acquired consumable data.

[0009] The actual consumption of the target semiconductor consumable is obtained by using a defined consumable analysis model based on the acquired consumable data.

[0010] The above method, by utilizing a pre-established consumable analysis model corresponding to different consumable data and consumable data collection methods, and by using the consumable analysis model to analyze the acquired semiconductor consumable data, can automatically and accurately determine the actual consumption of different types of semiconductor consumables.

[0011] This application also provides a device for monitoring the consumption of semiconductor consumables, including:

[0012] The consumables data acquisition module is used to acquire consumables data related to the consumption of the target semiconductor consumables;

[0013] The model determination module is used to determine the consumable analysis model corresponding to the acquired consumable data based on the acquired consumable data and the acquisition method, as well as the pre-established consumable analysis models corresponding to different consumable data and acquisition methods.

[0014] The consumption calculation module is used to perform corresponding calculations based on the acquired consumable data using a defined consumable analysis model to obtain the actual consumption of the aforementioned target semiconductor consumables.

[0015] This application also provides a device for monitoring the consumption of semiconductor consumables. The device includes a memory and a processor. The memory stores a computer program that can run on the processor. When the computer program is executed by the processor, the processor performs any step in the above-described method for monitoring the consumption of semiconductor consumables.

[0016] This application also provides a computer storage medium storing a computer program, which, when executed by a processor, implements any step in the above-described method for monitoring the consumption of semiconductor consumables.

[0017] Furthermore, the technical effects of any of the implementation methods of the semiconductor consumable consumption monitoring device, equipment, and computer storage medium provided in this application can be found in the technical effects of the implementation methods of the semiconductor consumable consumption monitoring method, and will not be repeated here. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic diagram of a method for monitoring the consumption of semiconductor consumables in a related art provided in this application;

[0020] Figure 2A schematic diagram illustrating an application scenario of a method for monitoring the consumption of semiconductor consumables provided in this application embodiment;

[0021] Figure 3 A flowchart illustrating a method for monitoring the consumption of semiconductor consumables provided in this application embodiment;

[0022] Figure 4 A schematic diagram of a material consumption list provided for an embodiment of this application;

[0023] Figure 5 A flowchart illustrating a method for establishing a consumable analysis model, as provided in this application embodiment;

[0024] Figure 6 A schematic diagram illustrating the process of establishing a consumable analysis model as provided in an embodiment of this application;

[0025] Figure 7 A schematic diagram of a second statistical model provided in an embodiment of this application;

[0026] Figure 8 A schematic diagram of a first statistical model provided in an embodiment of this application;

[0027] Figure 9 A schematic diagram of an area model provided in an embodiment of this application;

[0028] Figure 10 A schematic diagram of a transition summation model provided in an embodiment of this application;

[0029] Figure 11 A schematic diagram of a time-sharing integral model provided in an embodiment of this application;

[0030] Figure 12 A schematic diagram of a calculation summation model provided for an embodiment of this application;

[0031] Figure 13 A schematic diagram of a difference statistical model provided in an embodiment of this application;

[0032] Figure 14 A schematic diagram of a semiconductor consumable consumption monitoring device provided in an embodiment of this application;

[0033] Figure 15 This is a schematic diagram of a semiconductor consumable consumption monitoring device provided in an embodiment of this application. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0035] The application scenarios described in the embodiments of this invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of this invention, and do not constitute a limitation on the technical solutions provided by the embodiments of this invention. As those skilled in the art will know, with the emergence of new application scenarios, the technical solutions provided by the embodiments of this invention are also applicable to similar technical problems.

[0036] Indirect materials (IDMs) refer to general consumable materials consumed by enterprises to organize, manage, and ensure the normal operation of production. These are materials that contribute to the formation of goods or facilitate production during the manufacturing process, but do not constitute the physical product. In the semiconductor wafer manufacturing process, the consumption of indirect materials (i.e., semiconductor consumables) is essential. Commonly used indirect materials include chemicals, special gases, polishing slurries, photoresist, and target materials, etc. In actual production, the calculation of the consumption of indirect materials is of great significance.

[0037] Semiconductor consumables consumption is typically divided into product consumption and periodic consumption. Product consumption refers to the consumption directly involved in the product manufacturing process, which is set based on the consumption required for each wafer. Periodic consumption refers to the consumption other than product consumption, which is set as the average daily consumption. After testing, it was found that the average daily consumption estimated based on experience during machine testing and rework differed significantly from the actual consumption, especially during the factory construction and capacity ramp-up phases.

[0038] Figure 1 A schematic diagram illustrating a method for monitoring semiconductor consumable consumption in a related art, provided in an embodiment of this application, is shown below. Figure 1In related technologies, semiconductor consumables are typically supplied centrally by the factory or directly by the equipment. When supplied centrally by the factory, pipelines usually connect multiple equipment. In this case, the actual consumption of semiconductor consumables is typically monitored by installing flow meters at the pipelines and manually reading / estimating liquid levels daily to determine the consumption. However, because the actual consumption of consumables is wide-ranging, involving multiple equipment and combinations of consumables, this monitoring method can only measure the total consumption of consumables across multiple equipment, not down to the individual equipment level. Furthermore, manually reading the readings of the measuring instruments on the equipment and performing calculations is not only time-consuming and labor-intensive, but also inaccurate and prone to significant discrepancies. If supplied directly by the equipment, the actual consumption of semiconductor consumables is typically monitored by weighing and recording consumption monthly / weekly. This method also suffers from significant errors in calculating the actual consumption.

[0039] Semiconductor FDC (Fault Detection and Classification) systems are mainly used for monitoring the status of production equipment. During the production process, they collect data from equipment sensors to monitor and detect anomalies in real time, and issue immediate alarms.

[0040] Based on the above problems, this application proposes a method for monitoring the consumption of semiconductor consumables in conjunction with the FDC system. By establishing different statistical models for different consumables and the methods of collecting consumable data, the consumption of semiconductor consumables can be automatically and accurately calculated.

[0041] Figure 2 Please refer to the schematic diagram illustrating an application scenario of a semiconductor consumable consumption monitoring method provided in this application embodiment. Figure 2 The monitoring method provided in this application embodiment collects data from the consumable sensors 202 set on each production machine in real time through the FDC system 201, and automatically calculates the actual consumption by combining the consumable data and the collection method with the pre-set consumable analysis model in the FDC system.

[0042] Figure 3 For a flowchart illustrating a method for monitoring semiconductor consumable consumption based on the aforementioned application scenario, please refer to the embodiments of this application. Figure 3 This application proposes a method for monitoring the consumption of semiconductor consumables, including:

[0043] Step 301: Obtain consumable data related to the consumption of the target semiconductor consumable;

[0044] Step 302: Based on the acquired consumable data and the acquisition method, and the pre-established consumable analysis model corresponding to different consumable data and acquisition methods, determine the consumable analysis model corresponding to the acquired consumable data.

[0045] Step 303: Using the determined consumable analysis model, perform corresponding calculations based on the acquired consumable data to obtain the actual consumption of the target semiconductor consumable;

[0046] This application provides a method for monitoring the consumption of semiconductor consumables. By using a pre-established consumable analysis model corresponding to different consumable data and data collection methods, and by using the consumable analysis model to analyze the acquired semiconductor consumable data, the actual consumption of different types of semiconductor consumables can be automatically and accurately determined.

[0047] In some embodiments of this application, the types of target semiconductor consumables in step 301 include chemicals, gases, polishing slurries, photoresist, and targets, etc.; consumable data related to the consumption of target semiconductor consumables include the consumption data of semiconductor indirect materials in the wafer manufacturing process and non-production processes. The consumable data in the wafer manufacturing process can be the consumption amount for one wafer produced by a production machine, or the consumption amount for one production cycle of a production machine; the consumption data in non-production processes refers to the consumption data to ensure normal machine production and product quality production, such as the consumption data during machine maintenance (cleaning of the production machine, etc.) and the consumption data to ensure the environment of the machine chamber.

[0048] In some embodiments of this application, since the target semiconductor consumables are used in different types of production machines, the data acquisition method for consumables in step 302 is different. In addition to the first acquisition method that uses consumable sensors to acquire consumable data, the acquisition method also includes a second acquisition method that uses other acquisition methods besides consumable sensors to acquire data.

[0049] During implementation, after establishing corresponding consumable analysis models for different consumable data and collection methods in advance, the consumable analysis models are stored in the FDC system. After obtaining consumable data and collection methods, the FDC system determines the corresponding consumable analysis model based on the consumable data and collection methods.

[0050] In some embodiments of this application, after obtaining the actual consumption of the target semiconductor consumable in step 303, the method further includes:

[0051] Based on the actual consumption and usage behavior of the target semiconductor consumables, establish a Bill of Materials (BOM) for the target semiconductor consumables and adjust the usage of the target semiconductor consumables.

[0052] The aforementioned consumption behavior refers to the consumption behavior corresponding to the actual consumption of the target semiconductor consumables, such as consumption during a single production process, consumption when producing a single product, consumption within a specified time period, etc.

[0053] In practice, after obtaining the actual consumption of the target semiconductor consumables, the rationality of the consumption of semiconductor consumables can be determined based on the actual consumption and consumption behavior. To facilitate the judgment, a material consumption list of the target semiconductor consumables can be established, that is, the consumption related to the consumption behavior of semiconductor consumables in the production or non-production process can be recorded, and the rationality of the consumption of semiconductor consumables can be judged accordingly. Furthermore, if the consumption is determined to be unreasonable, the adjustment method of semiconductor consumables consumption can be determined based on the consumption list.

[0054] After obtaining the actual consumption of the target semiconductor consumables using the above method, a material consumption list for the target semiconductor consumables is established based on the actual consumption and consumption behavior. This facilitates the observation of the actual consumption of the target semiconductor consumables, the determination of whether the consumption is reasonable, and the determination of how to adjust it if the consumption is determined to be unreasonable.

[0055] Among them, such as Figure 4 As shown, the material consumption list for the aforementioned target semiconductor consumables includes any one or more of the following:

[0056] Material consumption of target semiconductor consumables in a single / specified number of production processes;

[0057] Material consumption of the target semiconductor consumables for a single product or a specified number of products;

[0058] Material consumption of the target semiconductor consumables within a specified time period;

[0059] Machine downtime / Material consumption of target semiconductor consumables within each set time period after downtime;

[0060] The material consumption of target semiconductor consumables before the start of each work session during non-continuous operation of the machine.

[0061] The above method specifically defines the contents of the material consumption list for the target semiconductor consumables, and its contents facilitate subsequent judgment on whether the usage of semiconductor consumables should be adjusted.

[0062] Before obtaining consumable data related to the consumption of the target semiconductor consumable in step 301 above, it is necessary to establish a consumable analysis model. The following is combined with... Figure 5 and Figure 6 The process of establishing the consumables analysis model is explained in detail.

[0063] In some embodiments of this application, consumable analysis models corresponding to different consumable data and collection methods are established, such as... Figure 5 As shown, it includes:

[0064] Step 501: For different types of semiconductor consumables, establish the correspondence between semiconductor consumables and the production machines associated with them, and determine the usage type of semiconductor consumables in the corresponding production machines.

[0065] Step 502: Based on the usage type, determine the data collection method for each type of semiconductor consumable in each corresponding production machine;

[0066] Step 503: Based on the data collection method of each type of semiconductor consumable in each corresponding production machine, establish a consumable analysis model corresponding to different consumable data and collection methods.

[0067] The above method determines the data collection method for different types of semiconductor consumables in the corresponding production machines based on their different usage types in the corresponding production machines, and establishes different consumable analysis models corresponding to the collection methods. The established models are more targeted and their calculation results are more accurate.

[0068] In step 501 above, each type of semiconductor consumable is connected to multiple production machines and will be transferred to multiple production machines during use, and its use in each machine may also be different.

[0069] In the process of establishing the consumables analysis model, such as Figure 6 As shown, firstly, based on the different types of semiconductor consumables, for each type of semiconductor consumable, determine the production machine that uses the semiconductor consumable in the production or non-production process, and record the number of the production machine corresponding to the semiconductor consumable, that is, establish the correspondence between semiconductor consumables and the production machines associated with semiconductor consumables.

[0070] Then, the usage type of the semiconductor consumable in each corresponding production machine is determined in order to determine the method of collecting the consumable data of the semiconductor consumable. The usage type refers to the purpose of the semiconductor consumable in the production machine, such as grinding wafers, cleaning production machines, etc.

[0071] It should be noted that the same type of semiconductor consumables may correspond to different usage types in the same production machine.

[0072] In step 502 above, when determining the data collection method for semiconductor consumables in each corresponding production machine based on the usage type, it is first determined whether there is a consumable sensor in the production machine that corresponds to the current usage type of the semiconductor consumable. When it is determined that there is a corresponding consumable sensor in the production machine, the data collection method is determined to be to use the consumable sensor to collect the consumption amount of the semiconductor consumable (i.e., consumable data), and then the data collected by the consumable sensor is obtained through the FDC system. When there is no corresponding consumable sensor in the production machine, the consumption amount of the semiconductor consumable (i.e., consumable data) is determined according to the operating rules of the production machine in the semiconductor process.

[0073] For example, some gases in semiconductor consumables do not have corresponding consumable sensors in the production equipment, but their usage amount is fixed each time. Similarly, polishing slurry in semiconductor consumables also does not have corresponding consumable sensors in the production equipment, but it has a fixed usage cycle and the amount used each time is also fixed. Therefore, the actual consumption amount can be determined by the corresponding usage cycle.

[0074] It should be noted that since the same type of semiconductor consumables may have different usage types in different or the same production machines, the data collection methods in the production machines may differ. For example, in production machine 1, there is a consumable sensor corresponding to usage type 1, but no consumable sensor corresponding to usage type 2. In this case, the data collection method corresponding to usage type 1 is the first data collection method, and the data collection method corresponding to usage type 2 is the second data collection method. When calculating the actual consumption, the actual consumption in production machine 1 is determined by adding the two data collection methods.

[0075] In some embodiments of this application, after establishing consumable analysis models corresponding to different consumable data and collection methods, the method further includes:

[0076] For each consumable analysis model, obtain the reference actual consumption of semiconductor consumables corresponding to the consumable analysis model, and use the consumable analysis model to calculate the test consumption of the corresponding semiconductor consumables.

[0077] The test consumption of semiconductor consumables is compared with the reference actual consumption. If the difference between the test consumption and the reference actual consumption is not less than the set threshold, the consumable analysis model corresponding to the semiconductor consumables is adjusted until the difference between the test consumption and the reference actual consumption is less than the set threshold.

[0078] During implementation, to ensure the accuracy of the established consumable analysis model, it is necessary to test the model in advance. This involves calculating the test consumption of semiconductor consumables using the established model and obtaining a reference actual consumption. The test consumption is then compared with the reference actual consumption. If the error is less than a set threshold, the consumable analysis model is considered to meet the requirements. If the error is not less than the set threshold, the model is considered to not meet the requirements and needs to be adjusted to meet them.

[0079] For example, when using semiconductor consumable NF3 for silicon etching and cleaning within a semiconductor process chamber, if the NF3 is pre-determined using a formula, the reference actual consumption for silicon etching is 30–200 ml, and the reference actual consumption for cleaning is 200–300 ml, with a preset threshold of 10%, and during the consumable analysis model test, the error between the test consumption for silicon etching and the reference actual consumption calculated by the integral model in the consumable analysis model is 0.7%, and the error between the test consumption for cleaning and the reference actual consumption is 5.4%, both less than the set threshold of 10%, then the consumable analysis model is deemed to meet the requirements.

[0080] The above method ensures the accuracy of calculating the actual consumption of semiconductor consumables by adjusting the consumable analysis model corresponding to various types of semiconductor consumables.

[0081] In some embodiments of this application, the actual consumption is determined in the following manner:

[0082] Determined based on the formulation and number of uses of semiconductor consumables;

[0083] Determined based on real-time monitoring values ​​of semiconductor consumable consumption.

[0084] The real-time monitoring values ​​of the above consumption can be determined by manually reading the readings of the consumable sensors in the production equipment in real time.

[0085] The above method establishes a reference for determining actual consumption, making the plan more specific and feasible.

[0086] In some embodiments of this application, step 503 above, which establishes consumable analysis models corresponding to different consumable data and collection methods, includes:

[0087] When the data collection method for consumables is determined to be the first collection method, the data for consumables is determined according to the collection method of the consumable sensor, and a first statistical model corresponding to the data for consumables is established.

[0088] When the data collection method for consumables is determined to be the second collection method, the data to be collected for consumables is determined, and a second statistical model corresponding to the data for consumables is established.

[0089] The first data collection method mentioned above includes, but is not limited to, real-time data collection, last consumption data collection, real-time data collection by time period for semiconductor consumable type, data collection based on pressure value change count, and cumulative data collection. The corresponding consumable data are real-time flow rate, single consumption, real-time flow rate by time period, number of pressure value changes, and cumulative consumption.

[0090] When the above-mentioned data collection method is the second data collection method, the operating rules of the production machine for determining the consumable data include, but are not limited to, the same consumption of semiconductor consumables in each fixed time / fixed time interval, different formulations of semiconductor consumables used in different batches / wafers during production, fixed consumption of semiconductor consumables in each batch / wafer production process, fixed replacement cycle of semiconductor consumables, fixed machine maintenance cycle, and fixed consumption of semiconductor consumables in each batch production process.

[0091] The above method, by classifying the consumption analysis models of semiconductor consumables with corresponding consumable sensors and those without, enables the determination of the corresponding model for different types of semiconductor consumables, thus more accurately determining their actual consumption.

[0092] In some embodiments of this application, such as Figure 7 As shown, the second statistical model described above analyzes consumable data according to any one or more of the following methods:

[0093] The first analytical method calculates consumption based on the usage time / usage interval of semiconductor consumables;

[0094] The second analytical method calculates consumption based on the number of times the production formula is executed in each production batch / wafer;

[0095] A third analytical method for calculating consumption based on the number of production batches / wafers;

[0096] The fourth analysis method calculates consumption based on the replacement cycle of semiconductor consumables;

[0097] The fifth analysis method calculates consumption based on the machine's maintenance cycle;

[0098] The sixth analysis method calculates consumption based on the number of production batches.

[0099] The above method limits the analysis method of different consumable data in the second statistical model, making the semiconductor consumable consumption monitoring method provided in this application embodiment more accurate in calculating the consumption of each consumable data.

[0100] In some embodiments of this application, the collected consumable data is determined, and a second statistical model corresponding to the consumable data is established, including:

[0101] When it is determined that the consumption of semiconductor consumables is the same in each fixed time / fixed time interval, the collected consumable data is determined to be the usage time / usage time interval of semiconductor consumables, and a second statistical model is established to analyze the consumable data according to the first analysis method;

[0102] When it is determined that the formulations used for semiconductor consumables are different in different batches / wafers during production, the collected consumable data is determined to be the number of times the production batch / wafer is executed with the production formulation, and a second statistical model is established to analyze the consumable data according to the second analysis method.

[0103] When the consumption of semiconductor consumables is fixed in each batch / wafer production process, the collected consumable data is determined to be the number of batches / wafers produced, and a second statistical model is established to analyze the consumable data according to the third analysis method.

[0104] When the consumption of semiconductor consumables is fixed in each replacement cycle, the collected consumable data is determined as the replacement cycle of semiconductor consumables, and a second statistical model is established to analyze the consumable data according to the fourth analysis method.

[0105] When the consumption of semiconductor consumables is fixed in each maintenance cycle of a production machine, the collected consumables data is determined as the maintenance cycle of the production machine, and a second statistical model is established to analyze the consumables data according to the fifth analysis method.

[0106] When the consumption of semiconductor consumables is fixed in each batch production process, the collected consumable data is determined to be the number of production batches, and a second statistical model is established to analyze the consumable data according to the sixth analysis method.

[0107] The above method determines the collected consumable data based on the operating rules of the production machine and establishes a corresponding second statistical model for different consumable data. This achieves the matching of the operating rules of the production machine, consumable data, and different analysis methods in the second statistical model, ensuring the accuracy of the calculation of actual consumption.

[0108] In some embodiments of this application, such as Figure 8 As shown, the first statistical model mentioned above includes any one or more of the following:

[0109] An area model that performs integral calculations on the collected real-time consumable data;

[0110] A transition summation model that sums up the collected single-use consumable data;

[0111] A time-sharing integral model that performs time-sharing integral calculations on the collected real-time consumable data;

[0112] A calculation and summation model that counts the number of times the fixed consumable data is collected;

[0113] A difference statistical model that performs difference calculations on the collected cumulative consumable data.

[0114] The above method establishes different types of primary statistical models for different types of sensor acquisition methods, making the models more targeted and the calculation results of the actual consumption of the models more accurate.

[0115] In some embodiments of this application, consumable data is determined based on the acquisition method of the consumable sensor, and a first statistical model corresponding to the consumable data is established, including at least one of the following steps:

[0116] When the data acquisition method of the consumable sensor is determined to be real-time acquisition, the consumable data is determined to be real-time flow based on the data acquisition method of the consumable sensor, and an area model corresponding to the real-time flow is established.

[0117] When the data acquisition method of the consumable sensor is determined to be the last consumption data acquisition, the consumable data is determined to be the single consumption data based on the data acquisition method of the consumable sensor, and a transition summation model corresponding to the single consumption data is established.

[0118] When the data acquisition method of the consumable sensor is determined to be real-time data acquisition in time segments for semiconductor consumable types, the consumable data is determined to be real-time flow in time segments based on the data acquisition method of the consumable sensor, and a time-sharing integral model corresponding to the real-time flow in time segments is established.

[0119] When the data acquisition method of the consumable sensor is determined to be based on the number of pressure value changes, the data of the consumable is determined to be the number of pressure value changes, and a calculation and summation model corresponding to the number of pressure value changes is established.

[0120] When the data acquisition method of the consumable sensor is determined to be cumulative acquisition, the consumable data is determined as the cumulative consumption based on the data acquisition method of the consumable sensor, and a statistical model of the difference corresponding to the cumulative consumption is established.

[0121] The above method specifically defines the steps for establishing the corresponding first statistical model based on the sensor acquisition method, making the scheme more concrete.

[0122] In some embodiments of this application, the types of semiconductor consumables include chemicals, gases, polishing slurries, photoresist, and targets;

[0123] The area model is associated with semiconductor consumables of the type of gas or polishing slurry;

[0124] The transition summation model is associated with semiconductor consumables that are chemical products;

[0125] The time-sharing integral model is associated with semiconductor consumables of gas type;

[0126] The calculation summation model is associated with semiconductor consumables of the photoresist type;

[0127] The difference statistical model is associated with semiconductor consumables of the target type.

[0128] The above method specifically limits the types of semiconductor consumables applicable to each type of first statistical model, making the model more targeted and the consumption calculation for each type of semiconductor consumable more accurate.

[0129] The following combination Figures 9 to 13 The first statistical model described above will be explained in detail.

[0130] Figure 9 Please refer to the schematic diagram of an area model provided in the embodiments of this application. Figure 9 The above area model applies to semiconductor consumables in fluid form, where the corresponding consumable sensor collects the real-time flow rate of the semiconductor consumables.

[0131] like Figure 9 As shown, the flow sensor collects the real-time flow rate of the semiconductor consumable NF3 in milliliters per minute (ml / min). The FDC system receives the flow monitoring value fed back by the flow sensor in real time. If no feedback is received, the real-time monitoring value is determined to be 0. Figure 9 After showing the consumable data, the consumption of NF3 during the collection period is calculated using an integral model (i.e., Figure 9 (Area of ​​the figure in the middle).

[0132] It should be noted that the consumption calculated in the above embodiments of this application is the consumption of the production machine in one production process. That is, if the production machine produces 5 batches of wafers each time, the calculation result is the consumption of semiconductor consumables for producing 5 batches of wafers.

[0133] Figure 10 Please refer to the schematic diagram of a transition summation model provided in the embodiments of this application. Figure 10 The above transition summation model is applicable to the data collection of the last consumption of semiconductor consumables by consumable sensors.

[0134] After acquiring the consumable data collected by the consumable sensor, the FDC system calculates each data inflection point, determines the data collected by the consumable sensor at the data inflection point, and sums them up to calculate the total consumption of semiconductor consumables.

[0135] like Figure 10 As shown, polysilazane (a chemical used in semiconductor consumables) is consumed in very small amounts (a few milliliters) each time it is used. The corresponding value collected by the consumable sensor is the last consumption amount. Therefore, the FDC system can calculate the consumption of polysilazane through the transition summation model.

[0136] Figure 11 Please refer to the schematic diagram of a time-sharing integral model provided in the embodiments of this application. Figure 11 The above time-sharing integral model is mainly applicable to the real-time flow rate of semiconductor consumables collected by the consumable sensor, and the same pipe collected by the consumable sensor carries different semiconductor consumables at different times. The semiconductor consumables are usually gases.

[0137] After acquiring consumable data collected by the consumable sensor, the FDC system segments the time periods in which different semiconductor consumables circulate, and then performs time-division integration based on the corresponding time periods to determine their actual consumption. The segmentation of the time periods in which different semiconductor consumables circulate can be achieved by adding filtering conditions.

[0138] like Figure 11 As shown, the consumable sensor collects data on the flow of DCS gas and N2 gas in the pipeline at different time periods. When calculating the consumption, the time-sharing integral model in the FDC system is used to divide the time periods in which DCS gas and N2 gas flow, and then the actual consumption is determined by time-sharing integration based on the corresponding time periods.

[0139] Figure 12 Please refer to the schematic diagram of a calculation summation model provided in the embodiments of this application. Figure 12 The above calculation and summation model is mainly applicable to a fixed amount of semiconductor consumables used each time, and the consumable sensor collects the real-time pressure value of the pipeline where the semiconductor consumables are located. This pressure value will change periodically with the use of the semiconductor consumables, which are usually photoresist liquids.

[0140] After acquiring consumable data collected by the consumable sensor, the FDC system determines the number of times the semiconductor consumable is used based on the number of times the pressure value changes periodically. Combined with the fixed consumption amount of the semiconductor consumable each time, the actual consumption amount is determined.

[0141] like Figure 12As shown in the figure, the horizontal axis represents the collection time, and the vertical axis represents the pressure value during collection. After collecting the consumable data shown in the figure, the number of cycles can be calculated using the preset Hit Count Below model. That is, a standard line is preset, and the number of cycles is determined by calculating the number of times the collected data changes from above the standard line to below the standard line. This number of cycles is the number of times the semiconductor consumable is used. Multiplying it by the fixed consumption amount of the semiconductor consumable each time will give the actual consumption amount.

[0142] Figure 13 Please refer to the schematic diagram of a difference statistical model provided in the embodiments of this application. Figure 13 The aforementioned difference statistical model is applicable to the cumulative consumption of semiconductor consumables collected by the consumable sensor, which is typically a target material. After acquiring the consumable data collected by the consumable sensor, the FDC system determines the actual consumption of the semiconductor consumables by calculating the difference between the collected values ​​corresponding to the start and end times of the collection.

[0143] like Figure 13 As shown in the figure, the horizontal axis represents the data acquisition time, and the vertical axis represents the cumulative data consumption during acquisition. Figure 13 After obtaining the data shown, the actual consumption of semiconductor consumables is determined by calculating the difference between the cumulative consumption at the end time and the cumulative consumption at the beginning time.

[0144] It should be noted that the method described above for calculating the actual consumption using an integral algorithm is applicable to MFC (Mass Flow Controller, flow sensor) type sensors. However, special settings are required for different production machines and semiconductor consumables. For example, although the gas consumption is set to 0 when not in use, the FDC system usually receives a very small non-zero feedback value when the consumable is not consumed during actual use. This is because MFC type sensors use current readings, which have negative errors near the zero point. Therefore, in actual settings, the minimum scale of the MFC hardware setting is 1, and readings less than 0.5 are filtered out and regarded as zero values.

[0145] This application also provides a device for monitoring the consumption of semiconductor consumables. Please refer to [link / reference]. Figure 14 The device includes:

[0146] Consumable data acquisition module 1401 is used to acquire consumable data related to the consumption of the target semiconductor consumable;

[0147] The model determination module 1402 is used to determine the consumable analysis model corresponding to the acquired consumable data based on the acquired consumable data and the acquisition method, and the pre-established consumable analysis model corresponding to different consumable data and acquisition methods.

[0148] The consumption calculation module 1403 is used to perform corresponding calculations based on the acquired consumable data using a determined consumable analysis model to obtain the actual consumption of the aforementioned target semiconductor consumable.

[0149] Optionally, the semiconductor consumable consumption monitoring device provided in this application embodiment further includes a model building module, which is used for:

[0150] For different types of semiconductor consumables, establish a correspondence between semiconductor consumables and the associated production machines, and determine the usage type of the semiconductor consumables in the corresponding production machines.

[0151] Based on the above usage types, the data collection method for each type of semiconductor consumable in each corresponding production machine is determined.

[0152] Based on the data collection method of each type of semiconductor consumable in each corresponding production machine, a consumable analysis model corresponding to different consumable data and collection methods is established.

[0153] Optionally, the above model building module is used to build consumable analysis models corresponding to different consumable data and collection methods, including:

[0154] When the data collection method for consumables is determined to be the first collection method, the data for consumables is determined according to the data collection method of the consumable sensor, and a first statistical model corresponding to the above-mentioned data for consumables is established.

[0155] When the data collection method for consumables is determined to be the second collection method, the data to be collected for consumables is determined, and a second statistical model corresponding to the above-mentioned data for consumables is established.

[0156] Optionally, after the model building module establishes the second statistical model corresponding to the above consumable data, it is also used for:

[0157] For each consumable analysis model, obtain the reference actual consumption of the semiconductor consumables corresponding to the above consumable analysis model, and use the above consumable analysis model to calculate the corresponding test consumption of the above semiconductor consumables.

[0158] The test consumption of the aforementioned semiconductor consumables is compared with the reference actual consumption. If the difference between the test consumption and the reference actual consumption is not less than a set threshold, the consumable analysis model corresponding to the aforementioned semiconductor consumables is adjusted until the difference between the test consumption and the reference actual consumption is less than the set threshold.

[0159] Optionally, the above reference actual consumption amount is determined in the following way:

[0160] Determined based on the formulation and number of uses of semiconductor consumables;

[0161] Determined based on real-time monitoring values ​​of semiconductor consumable consumption.

[0162] Optionally, the second statistical model described above analyzes the consumables data according to any one or more of the following methods:

[0163] The first analytical method calculates consumption based on the usage time / usage interval of semiconductor consumables;

[0164] The second analytical method calculates consumption based on the number of times the production formula is executed in each production batch / wafer;

[0165] A third analytical method for calculating consumption based on the number of production batches / wafers;

[0166] The fourth analysis method calculates consumption based on the replacement cycle of semiconductor consumables;

[0167] The fifth analysis method calculates consumption based on the maintenance cycle of production machines;

[0168] The sixth analysis method calculates consumption based on the number of production batches.

[0169] Optionally, the model building module described above is used to determine the collected consumable data based on the above-mentioned formula, and to build a second statistical model corresponding to the above-mentioned consumable data, including:

[0170] When it is determined that the consumption of the above-mentioned semiconductor consumables is the same in each fixed time / fixed time interval, the collected consumable data is determined to be the usage time / usage time interval of the semiconductor consumables, and a second statistical model is established to analyze the above-mentioned consumable data according to the first analysis method.

[0171] When it is determined that the formulations used for the above semiconductor consumables are different in different batches / wafers during production, the collected consumable data is determined to be the number of times the production batch / wafer is produced using the production formulation, and a second statistical model is established to analyze the above consumable data according to the second analysis method.

[0172] When the consumption of the aforementioned semiconductor consumables is fixed in each batch / wafer production process, the collected consumable data is determined to be the number of batches / wafers produced, and a second statistical model is established to analyze the aforementioned consumable data according to the third analysis method.

[0173] When the consumption of the aforementioned semiconductor consumables is fixed in each replacement cycle, the collected consumable data is determined as the replacement cycle of the semiconductor consumables, and a second statistical model is established to analyze the aforementioned consumable data according to the fourth analysis method.

[0174] When the consumption of the aforementioned semiconductor consumables is fixed in each production machine maintenance cycle, the collected consumables data is determined as the production machine maintenance cycle, and a second statistical model is established to analyze the aforementioned consumables data according to the fifth analysis method.

[0175] When the consumption of the aforementioned semiconductor consumables is fixed in each batch production process, the collected consumable data is determined to be the number of production batches, and a second statistical model is established to analyze the aforementioned consumable data according to the sixth analysis method.

[0176] Optionally, the first statistical model mentioned above includes any one or more of the following:

[0177] An area model that performs integral calculations on the collected real-time consumable data;

[0178] A transition summation model that sums up the collected single-use consumable data;

[0179] A time-sharing integral model that performs time-sharing integral calculations on the collected real-time consumable data;

[0180] A calculation and summation model that counts the number of times the fixed consumable data is collected;

[0181] A difference statistical model that performs difference calculations on the collected cumulative consumable data.

[0182] Optionally, the model building module described above is used to determine consumable data based on the acquisition method of the consumable sensor and to build a first statistical model corresponding to the consumable data, including at least one of the following steps:

[0183] When the acquisition method of the above-mentioned consumable sensor is determined to be real-time acquisition, the above-mentioned consumable data is determined to be real-time flow based on the acquisition method of the above-mentioned consumable sensor, and an area model corresponding to the above-mentioned real-time flow is established.

[0184] When the data acquisition method of the above consumable sensor is determined to be the last consumption data acquisition, the above consumable data is determined to be the single consumption data based on the data acquisition method of the above consumable sensor, and a transition summation model corresponding to the above single consumption data is established.

[0185] When the acquisition method of the above consumable sensor is determined to be real-time acquisition of semiconductor consumables in different time periods, the consumable data is determined to be real-time flow in different time periods based on the acquisition method of the above consumable sensor, and a time-sharing integral model corresponding to the real-time flow in different time periods is established.

[0186] When the data acquisition method of the above-mentioned consumable sensor is determined to be based on the counting of pressure value changes, the data of the above-mentioned consumable is determined to be the number of pressure value changes according to the data acquisition method of the above-mentioned consumable sensor, and a calculation summation model corresponding to the number of pressure value changes is established.

[0187] When the data acquisition method of the above-mentioned consumable sensor is determined to be cumulative acquisition, the above-mentioned consumable data is determined as the cumulative consumption based on the acquisition method of the above-mentioned consumable sensor, and a statistical model of the difference corresponding to the above-mentioned cumulative consumption is established.

[0188] Optionally, the types of semiconductor consumables mentioned above include chemicals, gases, polishing slurries, photoresist, and targets;

[0189] The above area model is associated with semiconductor consumables of the type of gas or polishing slurry;

[0190] The aforementioned transition summation model is associated with semiconductor consumables that are chemical products;

[0191] The above time-sharing integral model is associated with gas-type semiconductor consumables;

[0192] The above calculation and summation model is associated with semiconductor consumables of the photoresist type;

[0193] The above difference statistical model is associated with semiconductor consumables of the target type.

[0194] Optionally, after obtaining the actual consumption of the target semiconductor consumable, the consumption calculation module 1403 further includes:

[0195] Based on the actual consumption and consumption behavior of the aforementioned target semiconductor consumables, a material consumption list for the aforementioned target semiconductor consumables is established, and the consumption of the aforementioned target semiconductor consumables is adjusted.

[0196] Optionally, the material consumption list for the aforementioned target semiconductor consumables includes any one or more of the following:

[0197] Material consumption of target semiconductor consumables in a single / specified number of production processes;

[0198] Material consumption of the target semiconductor consumables for a single product or a specified number of products;

[0199] Material consumption of the target semiconductor consumables within a specified time period;

[0200] Machine downtime / Material consumption of target semiconductor consumables within each set time period after downtime;

[0201] The material consumption of target semiconductor consumables before the start of each work session during non-continuous operation of the machine.

[0202] Based on the same disclosed concept, this application also provides a monitoring device for semiconductor consumable consumption. Since this device is the same as the device in the method of this application, and the principle of the device in solving the problem is similar to that of the method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0203] Those skilled in the art will understand that various aspects of this application can be implemented as a system, method, or program product. Therefore, various aspects of this application can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, collectively referred to herein as a "circuit," "module," or "system."

[0204] In some possible implementations, the device according to this application may include at least one processor and at least one memory. The memory stores program code that, when executed by the processor, causes the processor to perform the steps in the semiconductor consumable consumption monitoring method according to various exemplary embodiments of this application described above.

[0205] The following reference Figure 15 To describe the device 1500 according to this embodiment of the present application. Figure 15 The device 1500 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0206] like Figure 15 As shown, device 1500 is presented in the form of a general-purpose device. Components of device 1500 may include, but are not limited to: at least one processor 1501, at least one memory 1502, and a bus 1503 connecting different system components (including memory 1502 and processor 1501). The memory stores program code that, when executed by the processor, causes the processor to perform the following steps:

[0207] Acquire consumable data related to the consumption of the target semiconductor consumable;

[0208] Based on the acquired consumable data and the acquisition method, and the pre-established consumable analysis model corresponding to different consumable data and acquisition methods, determine the consumable analysis model corresponding to the acquired consumable data.

[0209] The actual consumption of the target semiconductor consumables is obtained by using a defined consumables analysis model based on the acquired consumables data.

[0210] Bus 1503 represents one or more of several bus structures, including a memory bus or memory controller, peripheral bus, processor, or local bus using any of the various bus structures.

[0211] The memory 1502 may include a readable medium in the form of volatile memory, such as random access memory (RAM) 15021 and / or cache memory 15022, and may further include read-only memory (ROM) 15023.

[0212] The memory 1502 may also include a program / utility 15025 having a set (at least one) of program modules 15024, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0213] Device 1500 can also communicate with one or more external devices 1504 (e.g., keyboard, pointing device, etc.), one or more devices that enable a user to interact with device 1500, and / or any device that enables device 1500 to communicate with one or more other devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 1505. Furthermore, device 1500 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 1506. As shown, network adapter 1506 communicates with other modules used with device 1500 via bus 1503. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with device 1500, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0214] Optionally, before the processor acquires consumable data related to the consumption of the target semiconductor consumable, it further includes:

[0215] For different types of semiconductor consumables, establish a correspondence between semiconductor consumables and the associated production machines, and determine the usage type of the semiconductor consumables in the corresponding production machines.

[0216] Based on the above usage types, the data collection method for each type of semiconductor consumable in each corresponding production machine is determined.

[0217] Based on the data collection method of each type of semiconductor consumable in each corresponding production machine, a consumable analysis model corresponding to different consumable data and collection methods is established.

[0218] Optionally, the processor described above is used to establish consumable analysis models corresponding to different consumable data and acquisition methods, including:

[0219] When the data collection method for consumables is determined to be the first collection method, the data for consumables is determined according to the data collection method of the consumable sensor, and a first statistical model corresponding to the above-mentioned data for consumables is established.

[0220] When the data collection method for consumables is determined to be the second collection method, the data to be collected for consumables is determined, and a second statistical model corresponding to the above-mentioned data for consumables is established.

[0221] Optionally, the processor described above is also used for:

[0222] For each consumable analysis model, obtain the reference actual consumption of the semiconductor consumables corresponding to the above consumable analysis model, and use the above consumable analysis model to calculate the corresponding test consumption of the above semiconductor consumables.

[0223] The test consumption of the aforementioned semiconductor consumables is compared with the reference actual consumption. If the difference between the test consumption and the reference actual consumption is not less than a set threshold, the consumable analysis model corresponding to the aforementioned semiconductor consumables is adjusted until the difference between the test consumption and the reference actual consumption is less than the set threshold.

[0224] Optionally, the above reference actual consumption amount is determined in the following way:

[0225] Determined based on the formulation and number of uses of semiconductor consumables;

[0226] Determined based on real-time monitoring values ​​of semiconductor consumable consumption.

[0227] Optionally, the second statistical model described above analyzes the consumables data according to any one or more of the following methods:

[0228] The first analytical method calculates consumption based on the usage time / usage interval of semiconductor consumables;

[0229] The second analytical method calculates consumption based on the number of times the production formula is executed in each production batch / wafer;

[0230] A third analytical method for calculating consumption based on the number of production batches / wafers;

[0231] The fourth analysis method calculates consumption based on the replacement cycle of semiconductor consumables;

[0232] The fifth analysis method calculates consumption based on the maintenance cycle of production machines;

[0233] The sixth analysis method calculates consumption based on the number of production batches.

[0234] Optionally, the processor is used to determine the collected consumable data and establish a second statistical model corresponding to the consumable data, including:

[0235] When it is determined that the consumption of the above-mentioned semiconductor consumables is the same in each fixed time / fixed time interval, the collected consumable data is determined to be the usage time / usage time interval of the semiconductor consumables, and a second statistical model is established to analyze the above-mentioned consumable data according to the first analysis method.

[0236] When it is determined that the formulations used for the above semiconductor consumables are different in different batches / wafers during production, the collected consumable data is determined to be the number of times the production batch / wafer is produced using the production formulation, and a second statistical model is established to analyze the above consumable data according to the second analysis method.

[0237] When the consumption of the aforementioned semiconductor consumables is fixed in each batch / wafer production process, the collected consumable data is determined to be the number of batches / wafers produced, and a second statistical model is established to analyze the aforementioned consumable data according to the third analysis method.

[0238] When the consumption of the aforementioned semiconductor consumables is fixed in each replacement cycle, the collected consumable data is determined as the replacement cycle of the semiconductor consumables, and a second statistical model is established to analyze the aforementioned consumable data according to the fourth analysis method.

[0239] When the consumption of the aforementioned semiconductor consumables is fixed in each production machine maintenance cycle, the collected consumables data is determined as the production machine maintenance cycle, and a second statistical model is established to analyze the aforementioned consumables data according to the fifth analysis method.

[0240] When the consumption of the aforementioned semiconductor consumables is fixed in each batch production process, the collected consumable data is determined to be the number of production batches, and a second statistical model is established to analyze the aforementioned consumable data according to the sixth analysis method.

[0241] Optionally, the first statistical model mentioned above includes any one or more of the following:

[0242] An area model that performs integral calculations on the collected real-time consumable data;

[0243] A transition summation model that sums up the collected single-use consumable data;

[0244] A time-sharing integral model that performs time-sharing integral calculations on the collected real-time consumable data;

[0245] A calculation and summation model that counts the number of times the fixed consumable data is collected;

[0246] A difference statistical model that performs difference calculations on the collected cumulative consumable data.

[0247] Optionally, the processor is used to determine consumable data based on the acquisition method of the consumable sensor and to establish a first statistical model corresponding to the consumable data, including at least one of the following steps:

[0248] When the acquisition method of the above-mentioned consumable sensor is determined to be real-time acquisition, the above-mentioned consumable data is determined to be real-time flow based on the acquisition method of the above-mentioned consumable sensor, and an area model corresponding to the above-mentioned real-time flow is established.

[0249] When the data acquisition method of the above consumable sensor is determined to be the last consumption data acquisition, the above consumable data is determined to be the single consumption data based on the data acquisition method of the above consumable sensor, and a transition summation model corresponding to the above single consumption data is established.

[0250] When the acquisition method of the above consumable sensor is determined to be real-time acquisition of semiconductor consumables in different time periods, the consumable data is determined to be real-time flow in different time periods based on the acquisition method of the above consumable sensor, and a time-sharing integral model corresponding to the real-time flow in different time periods is established.

[0251] When the data acquisition method of the above-mentioned consumable sensor is determined to be based on the counting of pressure value changes, the data of the above-mentioned consumable is determined to be the number of pressure value changes according to the data acquisition method of the above-mentioned consumable sensor, and a calculation summation model corresponding to the number of pressure value changes is established.

[0252] When the data acquisition method of the above-mentioned consumable sensor is determined to be cumulative acquisition, the above-mentioned consumable data is determined as the cumulative consumption based on the acquisition method of the above-mentioned consumable sensor, and a statistical model of the difference corresponding to the above-mentioned cumulative consumption is established.

[0253] Optionally, the types of semiconductor consumables mentioned above include chemicals, gases, polishing slurries, photoresist, and targets;

[0254] The above area model is associated with semiconductor consumables of the type of gas or polishing slurry;

[0255] The aforementioned transition summation model is associated with semiconductor consumables that are chemical products;

[0256] The above time-sharing integral model is associated with gas-type semiconductor consumables;

[0257] The above calculation and summation model is associated with semiconductor consumables of the photoresist type;

[0258] The above difference statistical model is associated with semiconductor consumables of the target type.

[0259] Optionally, the processor described above is also used for:

[0260] Based on the actual consumption and consumption behavior of the aforementioned target semiconductor consumables, a material consumption list for the aforementioned target semiconductor consumables is established, and the consumption of the aforementioned target semiconductor consumables is adjusted.

[0261] Optionally, the material consumption list for the aforementioned target semiconductor consumables includes any one or more of the following:

[0262] Material consumption of target semiconductor consumables in a single / specified number of production processes;

[0263] Material consumption of the target semiconductor consumables for a single product or a specified number of products;

[0264] Material consumption of the target semiconductor consumables within a specified time period;

[0265] Machine downtime / Material consumption of target semiconductor consumables within each set time period after downtime;

[0266] The material consumption of target semiconductor consumables before the start of each work session during non-continuous operation of the machine.

[0267] In some possible implementations, various aspects of the semiconductor consumable consumption monitoring method provided in this application can also be implemented in the form of a program product, which includes program code. When the program product is run on a computer device, the program code is used to cause the computer device to perform the steps in the semiconductor consumable consumption monitoring method according to the various exemplary embodiments of this application described above.

[0268] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0269] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0270] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A monitoring method for the consumption amount of semiconductor consumables. The semiconductor consumables are indirect materials that assist in the formation or production of semiconductors during the semiconductor production and processing process but do not constitute the physical entity of semiconductor products, including at least one of chemicals, special gases, polishing liquids, photoresist liquids, and targets. It is characterized in that, Including: For different types of semiconductor consumables, establish the corresponding relationship between the semiconductor consumables and the production machines associated with the semiconductor consumables, and respectively determine the usage types of the semiconductor consumables in the corresponding production machines; Based on the usage types, respectively determine the acquisition methods of the consumable data of each type of semiconductor consumable in each corresponding production machine; the acquisition methods include the first acquisition method of acquiring consumable data by using a consumable sensor and the second acquisition method of acquiring by using other acquisition methods except the consumable sensor; According to the acquisition methods of the consumable data of each type of semiconductor consumable in each corresponding production machine, establish a consumable analysis model corresponding to different consumable data and acquisition methods; According to the acquired consumable data and acquisition method, and the pre-established consumable analysis model corresponding to different consumable data and acquisition methods, determine the consumable analysis model corresponding to the acquired consumable data; Use the determined consumable analysis model to perform corresponding calculations based on the acquired consumable data to obtain the actual consumption amount of the semiconductor consumable.

2. The method according to claim 1, wherein Establishing a consumable analysis model corresponding to different consumable data and acquisition methods includes: When it is determined that the acquisition method of the consumable data is the first acquisition method, determine the consumable data according to the acquisition method of the consumable sensor, and establish a first statistical model corresponding to the consumable data; When it is determined that the acquisition method of the consumable data is the second acquisition method, determine the acquired consumable data, and establish a second statistical model corresponding to the consumable data.

3. The method according to claim 1 or 2, characterized in that, It also includes: For each consumable analysis model, obtain the reference actual consumption amount of the semiconductor consumable corresponding to the consumable analysis model, and use the consumable analysis model to calculate the test consumption amount of the corresponding semiconductor consumable; Compare the test consumption amount of the semiconductor consumable with the reference actual consumption amount. When it is determined that the difference value between the test consumption amount and the reference actual consumption amount is not less than the set threshold, adjust the consumable analysis model corresponding to the semiconductor consumable until the difference value between the test consumption amount and the reference actual consumption amount is less than the set threshold.

4. The method according to claim 3, wherein The reference actual consumption amount is determined by the following method: Determined according to the usage formula and usage times of the semiconductor consumable; and / or, Determined according to the real-time monitoring value of the consumption amount of the semiconductor consumable.

5. The method according to claim 2, wherein The second statistical model analyzes the consumable data according to any one or any combination of the following methods: The first analysis method of calculating the consumption amount according to the usage time / usage time interval of the semiconductor consumable; The second analysis method of calculating the consumption amount according to the number of times of the production batch / wafers executing the production formula; The third analysis method of calculating the consumption amount according to the production batch / wafer quantity; The fourth analysis method of calculating the consumption amount according to the replacement cycle of the semiconductor consumable; The fifth analysis method of calculating the consumption amount according to the maintenance cycle of the production machine; The sixth analysis method of calculating the consumption amount according to the production batch number.

6. The method according to claim 5, wherein Determine the acquired consumable data, and establish a second statistical model corresponding to the consumable data, including: When it is determined that the consumption amount of the semiconductor consumable is the same within each fixed time / fixed time interval, determine that the collected consumable data is the usage time / usage time interval of the semiconductor consumable, and establish a second statistical model for analyzing the consumable data according to the first analysis method; When it is determined that the recipes used by the semiconductor consumable during production of different batches / wafers are different, determine that the collected consumable data is the number of times the production batch / wafer executes the production recipe, and establish a second statistical model for analyzing the consumable data according to the second analysis method; When it is determined that the consumption amount of the semiconductor consumable is fixed during the production process of each batch / wafer, determine that the collected consumable data is the number of production batches / wafers, and establish a second statistical model for analyzing the consumable data according to the third analysis method; When it is determined that the consumption amount of the semiconductor consumable is fixed during each replacement cycle, determine that the collected consumable data is the replacement cycle of the semiconductor consumable, and establish a second statistical model for analyzing the consumable data according to the fourth analysis method; When it is determined that the consumption amount of the semiconductor consumable is fixed during the maintenance cycle of each production machine, determine that the collected consumable data is the maintenance cycle of the production machine, and establish a second statistical model for analyzing the consumable data according to the fifth analysis method; When it is determined that the consumption amount of the semiconductor consumable is fixed during each batch production process, determine that the collected consumable data is the number of batch productions, and establish a second statistical model for analyzing the consumable data according to the sixth analysis method.

7. The method according to claim 2, wherein The first statistical model includes any one or any combination of the following: An area model obtained by performing an integration operation on the collected real-time consumable data; A turning sum model obtained by performing a summation operation on the collected single-time consumable data; A time-sharing integration model obtained by performing a time-segmented integration operation on the collected real-time consumable data; A calculation sum model obtained by performing a counting operation on the number of times of the collected fixed consumable data; A difference statistical model obtained by performing a difference operation on the collected cumulative consumable data.

8. The method according to claim 7, wherein Determine the consumable data according to the collection method of the consumable sensor, and establish a first statistical model corresponding to the consumable data, including at least one of the following steps: When it is determined that the collection method of the consumable sensor is real-time collection, determine that the consumable data is the real-time flow rate according to the collection method of the consumable sensor, and establish an area model corresponding to the real-time flow rate; When it is determined that the collection method of the consumable sensor is the last consumption amount collection, determine that the consumable data is the single consumption amount according to the collection method of the consumable sensor, and establish a turning sum model corresponding to the single consumption amount; When it is determined that the collection method of the consumable sensor is real-time collection by time segment for the semiconductor consumable type, determine that the consumable data is the real-time flow rate by time segment according to the collection method of the consumable sensor, and establish a time-sharing integration model corresponding to the real-time flow rate by time segment; When it is determined that the collection method of the consumable sensor is counting collection according to the change of the pressure value, determine that the consumable data is the number of changes of the pressure value according to the collection method of the consumable sensor, and establish a calculation sum model corresponding to the number of changes of the pressure value; When it is determined that the acquisition mode of the consumable sensor is cumulative acquisition, determine that the consumable data is the cumulative consumption amount according to the acquisition mode of the consumable sensor, and establish a difference statistical model corresponding to the cumulative consumption amount.

9. The method according to claim 7, wherein the area model is associated with semiconductor consumables of the type of gas or abrasive liquid; the turning sum model is associated with semiconductor consumables of the type of chemical; the time-sharing integration model is associated with semiconductor consumables of the type of gas; the calculation sum model is associated with semiconductor consumables of the type of photoresist; the difference statistical model is associated with semiconductor consumables of the type of target material.

10. The method according to claim 1, wherein Further comprising: According to the actual consumption amount and consumption behavior of the semiconductor consumables, establish a material consumption list of the semiconductor consumables, and adjust the consumption amount of the semiconductor consumables.

11. The method according to claim 10, wherein The material consumption list of the semiconductor consumables includes any one or any combination of the following: The material consumption amount of semiconductor consumables during a single / specified number of production processes; The material consumption amount of semiconductor consumables corresponding to a single / specified number of products; The material consumption amount of semiconductor consumables within a specified time period; The material consumption amount of semiconductor consumables at the moment of machine downtime / each set time period after downtime; The material consumption amount of semiconductor consumables before the start of each work during the non-continuous operation of the machine.

12. A semiconductor consumable consumption monitoring device based on the monitoring method according to any one of claims 1-11, characterized in that, Comprising: A consumable data acquisition module, configured to acquire consumable data related to the consumption of semiconductor consumables; A model determination module, configured to determine the consumable analysis model corresponding to the acquired consumable data according to the acquired consumable data and acquisition mode, and the pre-established consumable analysis models corresponding to different consumable data and acquisition modes; A consumption amount calculation module, configured to perform corresponding calculations based on the acquired consumable data by using the determined consumable analysis model to obtain the actual consumption amount of the semiconductor consumables.

13. A monitoring device for the consumption amount of semiconductor consumables, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the method for monitoring the consumption amount of semiconductor consumables according to any one of claims 1 to 11 are implemented.

14. A computer storage medium, on which a computer program is stored, characterized in that, When the program is executed by the processor, the steps of the method for monitoring the consumption amount of semiconductor consumables according to any one of claims 1 to 11 are implemented.

Images