An Internet of Things-based semiconductor production data supervision system and method

Abstract

The present invention relates to the technical field of supervision data transmission management, and specifically to a semiconductor production data supervision system and method based on the Internet of Things, including: a production data acquisition module, a time transition prediction module, a production equipment data analysis module, and a supervision data transmission management module. The semiconductor historical production information and the equipment historical inspection information for semiconductor production are collected through the production data acquisition module. A production time prediction model is established through the time transition prediction module to predict the time required to complete the semiconductor production process in the current production stage. The probability of equipment failure when using the equipment is analyzed through the production equipment data analysis module before the start of the next production stage in the current production stage, by referring to the equipment inspection data stored in the production supervision terminal to judge the operation state of the equipment. The equipment re-inspection and re-inspection data transmission management are carried out through the supervision data transmission management module, reducing the probability of equipment failure when using the equipment after referring to the inspection data.

Description

Technical Field

[0001] The present invention relates to the technical field of supervision data transmission management, and particularly to a semiconductor production data supervision system and method based on the Internet of Things. Background Technique

[0002] In the process of semiconductor production, different production equipment is used in different production stages. These production equipment need to be regularly inspected and maintained during the semiconductor production process to ensure the safe operation of the equipment, and thus ensure the safe production of semiconductors;

[0003] However, since the number of semiconductors produced each time may be different, the time for transitioning from the previous production stage to the next stage after completing the production process is also different. When the production equipment corresponding to the next production stage needs to be used, the last inspection time of the corresponding equipment may be a long time from the current time. Before using the equipment, it is necessary to view the equipment inspection data to determine whether the equipment can operate safely. For the inspection data uploaded to the supervision terminal after a long interval, referring to it to determine whether the equipment can operate safely is likely to reduce the accuracy of the judgment result. The prior art does not pre-intervene in the equipment inspection time, that is, the time for uploading inspection data, on the basis of the original equipment inspection cycle in necessary cases to maintain the accuracy of the judgment result by referring to the inspection data, and cannot reduce the equipment failure probability of using the equipment after referring to the inspection data.

[0004] Therefore, people need a semiconductor production data supervision system and method based on the Internet of Things to solve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a semiconductor production data supervision system and method based on the Internet of Things to solve the problems raised in the above background technique.

[0006] To solve the above technical problems, the present invention provides the following technical solution: A semiconductor production data supervision system based on the Internet of Things, the system includes: a production data collection module, a time transition prediction module, a production equipment data analysis module, and a supervision data transmission management module;

[0007] The output end of the production data collection module is connected to the input ends of the time transition prediction module and the production equipment data analysis module, the output end of the time transition prediction module is connected to the input end of the production equipment data analysis module, and the output end of the production equipment data analysis module is connected to the input end of the supervision data transmission management module;

[0008] The production data collection module is used to collect semiconductor historical production information and equipment historical inspection information for semiconductor production;

[0009] The time transition prediction module is used to establish a production time prediction model according to the semiconductor historical production information for different production stages, and predict the time required to complete the current number of semiconductor production processes in the current production stage;

[0010] The production equipment data analysis module is used to analyze the probability of equipment failure when using the equipment after judging the operation state of the equipment by referring to the equipment inspection data stored in the production supervision terminal before the start of the next production stage in the current production stage;

[0011] The supervision data transmission management module is used to manage equipment reinspection and reinspection data transmission according to the probability analysis result.

[0012] Furthermore, the production data acquisition module includes a production quantity acquisition unit, a completion time acquisition unit, and a production equipment information acquisition unit;

[0013] The production quantity acquisition unit is used to acquire the historical production quantity of semiconductors;

[0014] The completion time acquisition unit is used to acquire the time information spent on completing the production processes of the corresponding quantity of semiconductors in each production stage when producing the corresponding quantity of semiconductors in the past. Semiconductors include multiple production stages, including wafer preparation, lithography, thin film deposition, ion implantation, annealing, and packaging;

[0015] The production equipment information acquisition unit is used to acquire the inspection information of the equipment used for semiconductor production in each production stage, including the currently default set equipment inspection cycle, the previously set equipment inspection cycle, and the number of times the equipment fails when used for semiconductor production after obtaining the judgment result by directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally. There is corresponding equipment for semiconductor production in each production stage, such as lithography machines, thin film deposition devices, etc.

[0016] Furthermore, the time transition prediction module includes a production data retrieval unit, a time prediction model establishment unit, and a completion time prediction unit;

[0017] The input end of the production data retrieval unit is connected to the output ends of the production quantity acquisition unit and the completion time acquisition unit, the output end of the production data retrieval unit is connected to the input end of the time prediction model establishment unit, and the output end of the time prediction model establishment unit is connected to the input end of the completion time prediction unit;

[0018] The production data retrieval unit is used to retrieve the historical production quantity information of semiconductors and the time information spent on completing the production processes of the corresponding quantity of semiconductors in each production stage when producing the corresponding quantity of semiconductors in the past to the time prediction model establishment unit;

[0019] The time prediction model establishment unit is used to establish a production time prediction model for each production stage;

[0020] The completion time prediction unit is used to obtain the quantity of semiconductors that need to be produced currently, substitute the quantity of semiconductors that need to be produced currently into the production time prediction model, and predict the time required to complete the production processes of the current quantity of semiconductors in the current production stage.

[0021] Furthermore, the production equipment data analysis module includes a transmission data reference time analysis unit and an equipment failure probability analysis unit;

[0022] The input end of the transmission data reference time analysis unit is connected to the output end of the production equipment information collection unit, and the input end of the equipment failure probability analysis unit is connected to the output ends of the transmission data reference time analysis unit and the completion time prediction unit;

[0023] The transmission data reference time analysis unit is used to analyze, based on the production equipment information, the probability that the equipment used in the next production stage of the current production stage fails when being used for semiconductor production in different equipment inspection cycles in the past, due to directly referring to the equipment inspection data transmitted to the production supervision terminal to determine whether the equipment can operate normally, establish a failure probability prediction model for the corresponding equipment, and analyze the transmission time of the corresponding equipment inspection data referred to before the start of the next production stage under the currently set corresponding equipment inspection cycle;

[0024] The equipment failure probability analysis unit is used to substitute the interval duration between the transmission time and the start time of the next production stage into the failure probability prediction model, predict the probability that the corresponding equipment fails when being used for semiconductor production by directly referring to the equipment inspection data transmitted to the production supervision terminal to determine whether the corresponding equipment can operate normally in the current situation, and use the equipment after obtaining the judgment result.

[0025] Furthermore, the supervision data transmission management module includes a re-inspection judgment unit and an inspection data retransmission unit;

[0026] The input end of the re-inspection judgment unit is connected to the output end of the equipment failure probability analysis unit, and the output end of the re-inspection judgment unit is connected to the input end of the inspection data retransmission unit;

[0027] The re-inspection judgment unit is used to set a failure probability threshold and compare the predicted probability with the threshold: if the predicted probability exceeds the threshold, it is judged that the equipment used in the corresponding production stage needs to be reinspected before the start of the next production stage; if the predicted probability does not exceed the threshold, it is judged that the equipment used in the corresponding production stage does not need to be reinspected before the start of the next production stage, and it is selected to directly refer to the stored equipment inspection data in the production supervision terminal before the start of the next production stage to judge whether the corresponding equipment can operate normally;

[0028] The inspection data retransmission unit is used to, if it is judged that the equipment used in the corresponding production stage needs to be reinspected before the start of the next production stage, transmit the equipment inspection data to the production supervision terminal after reinspecting the equipment, and refer to the equipment inspection data transmitted after reinspecting before the start of the next production stage to judge whether the corresponding equipment can operate normally. After judging that the corresponding equipment can operate normally, use the corresponding equipment to complete the semiconductor production process of the corresponding stage at the start of the next production stage.

[0029] A semiconductor production data supervision method based on the Internet of Things includes the following steps:

[0030] Z1: Collect semiconductor historical production information and equipment historical inspection information for semiconductor production;

[0031] Z2: For different production stages, establish a production time prediction model based on semiconductor historical production information to predict the time required to complete the current number of semiconductor production processes in the current production stage;

[0032] Z3: Analyze the probability of equipment failure when using the equipment after judging the operating state of the equipment by referring to the stored equipment inspection data in the production supervision terminal before the start of the next production stage of the current production stage;

[0033] Z4: Perform equipment reinspection and reinspection data transmission management according to the probability analysis results.

[0034] Further, in step Z1: The set of historical production quantities for each previous semiconductor production is collected as M = {M1, M2,..., M r}, where r represents the number of historical semiconductor productions. When collecting the time information spent on completing the corresponding number of semiconductor production processes in each production stage for the corresponding number of semiconductors produced in the past, and collecting the inspection information of the equipment used for semiconductor production in each production stage, including the currently default set equipment inspection cycle, the previously set equipment inspection cycle, and the number of times the equipment failed when used in semiconductor production after obtaining the judgment result by directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally.

[0035] Further, in step Z2: when it is obtained that the semiconductor currently to be produced is in the i-th production stage, and when retrieving the corresponding number of semiconductors in the previous production set M, the time set for the i-th production stage to complete the production process of the corresponding number of semiconductors is t = {t1, t2,..., t r}, for the data points {(M1, t1), (M2, t2),..., (M r , t r )}, perform linear fitting to establish a production time prediction model for predicting the time required to complete the semiconductor production process in the current production stage: , where a and b represent the fitting coefficients of the production time prediction model, and solve a and b respectively:

[0036] ;

[0037] ;

[0038] where e = 1, 2,..., r, e represents the e-th semiconductor production, the number of semiconductors currently to be produced is obtained as N, substitute N into the production time prediction model: let X = N, and predict that the time required to complete the production process of the current number of semiconductors in the current production stage is .

[0039] Further, in step Z3: retrieve the set of equipment inspection cycles L = {L1, L2,..., L n} set in the past for semiconductor production in the (i + 1)-th production stage, n represents the number of historical settings of the inspection cycle for the corresponding equipment, retrieve the set of the number of times H = {H1, H2,..., H n} that the equipment malfunctioned when used in semiconductor production because the inspection data of the corresponding equipment transmitted to the production supervision terminal was directly referred to determine whether the equipment could operate normally after obtaining the judgment result under different inspection cycles in the past. According to calculate that when the set inspection cycle is L j , the probability P j that the equipment malfunctioned when used in semiconductor production because the inspection data of the corresponding equipment transmitted to the production supervision terminal was directly referred to determine whether the equipment could operate normally after obtaining the judgment result. Obtain the failure probability set as P = {P1, P2,..., P j ,..., P n}, for the data points {(H1, P1), (H2, P2),..., (H n , P n )}, perform linear fitting to establish a failure probability prediction model for the equipment used in semiconductor production in the (i + 1)-th production stage: , where and represent the fitting coefficients of the failure probability prediction model. The start production time of the semiconductor to be currently produced is A1, the start time of the i-th production stage of the semiconductor to be currently produced is A2, the currently default set equipment inspection cycle for the semiconductor production in the (i + 1)-th production stage is B1. The equipment inspection cycle B1 means that the equipment is inspected every interval of duration B1 after the start of the first stage of semiconductor production. The interval duration from the transmission time of the corresponding equipment inspection data referred to before the start of the (i + 1)-th production stage to the start time of the (i + 1)-th production stage under the currently set corresponding equipment inspection cycle is , where represents taking the floor of . Let . It is predicted that in the current situation, directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the corresponding equipment can operate normally. The probability that the corresponding equipment will fail when it is used in semiconductor production after obtaining the judgment result is p, .

[0040] Furthermore, in step Z4: Set the failure probability threshold as p ’ , compare p and p ’ : If p > p ’ , it is judged that the equipment used in the corresponding production stage needs to be reinspected before the start of the (i + 1)-th production stage. After reinspecting the equipment, the equipment inspection data is transmitted to the production supervision terminal, and the equipment inspection data transmitted after reinspection is referred to before the start of the (i + 1)-th production stage to judge whether the corresponding equipment can operate normally; if p ≤ p ’ , it is judged that there is no need to reinspect the equipment used in the corresponding production stage before the start of the (i + 1)-th production stage. It is selected to directly refer to the corresponding equipment inspection data stored in the production supervision terminal before the start of the (i + 1)-th production stage to judge whether the corresponding equipment can operate normally. After judging that the corresponding equipment can operate normally, the corresponding equipment is used to complete the semiconductor production process of the corresponding stage at the start of the (i + 1)-th production stage.

[0041] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0042] The present invention collects the historical production data of semiconductors and the time information required for each stage to complete the corresponding production volume through big data technology, uses the production volume and time data as training data to establish a production time prediction model, substitutes the number of semiconductors to be currently produced into the production time prediction model, and predicts the time required to complete the semiconductor production process of the current quantity in the current production stage, improving the accuracy of the predicted production completion time results in different situations;

[0043] By collecting information on equipment failures under different inspection cycles set for the equipment in the past, analyzing the equipment failure probability based on the collected information, analyzing the equipment inspection time transmitted to the supervision terminal for reference before the start of production work in the next production stage under the currently set equipment inspection cycle, predicting the reference corresponding inspection time, that is, judging the equipment operation situation result based on the inspection data corresponding to the inspection data transmission time, and using the equipment failure probability after determining that the equipment can operate normally, pre-judging whether it is necessary to re-inspect the semiconductor production equipment and re-transmit the inspection data according to the probability prediction result, and managing the equipment inspection data transmission, the equipment failure probability of using the equipment after referring to the inspection data is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0045] Figure 1 is a structural diagram of a semiconductor production data supervision system based on the Internet of Things according to the present invention;

[0046] Figure 2 is a flowchart of a semiconductor production data supervision method based on the Internet of Things according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The following describes the preferred embodiments of the present invention with reference to the drawings. It should be understood that the preferred embodiments described herein are only for explaining and illustrating the present invention, and are not used to limit the present invention.

[0048] The following combines Figure 1 - Figure 2 and specific embodiments to further illustrate the present invention.

[0049] Embodiment 1

[0050] As Figure 1 shown, this embodiment provides a semiconductor production data supervision system based on the Internet of Things. The system includes: a production data collection module, a time transition prediction module, a production equipment data analysis module, and a supervision data transmission management module;

[0051] The output end of the production data collection module is connected to the input ends of the time transition prediction module and the production equipment data analysis module. The output end of the time transition prediction module is connected to the input end of the production equipment data analysis module. The output end of the production equipment data analysis module is connected to the input end of the supervision data transmission management module;

[0052] The production data collection module is used to collect semiconductor historical production information and equipment historical inspection information for semiconductor production;

[0053] The time transition prediction module is used to establish a production time prediction model according to the semiconductor historical production information for different production stages, and predict the time required to complete the current number of semiconductor production processes in the current production stage;

[0054] The production equipment data analysis module is used to analyze the probability of equipment failure when using the equipment after judging the operating state of the equipment by referring to the equipment inspection data stored in the production supervision terminal before the start of the next production stage in the current production stage;

[0055] The supervision data transmission management module is used to manage equipment re-inspection and re-inspection data transmission according to the probability analysis results.

[0056] The production data acquisition module includes a production quantity acquisition unit, a completion time acquisition unit, and a production equipment information acquisition unit;

[0057] The production quantity acquisition unit is used to acquire the historical production quantity of semiconductors;

[0058] The completion time acquisition unit is used to acquire the time information spent on completing the production processes of the corresponding quantity of semiconductors in each production stage when producing the corresponding quantity of semiconductors in the past. Semiconductors include multiple production stages, including wafer preparation, lithography, thin film deposition, ion implantation, annealing, and packaging;

[0059] The production equipment information acquisition unit is used to acquire the inspection information of the equipment used for semiconductor production in each production stage, including the currently default set equipment inspection cycle, the previously set equipment inspection cycle, and the number of times the equipment fails when used for semiconductor production after obtaining the judgment result by directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally. Each production stage has corresponding equipment for semiconductor production, such as lithography machines, thin film deposition devices, etc.

[0060] The time transition prediction module includes a production data retrieval unit, a time prediction model establishment unit, and a completion time prediction unit;

[0061] The input end of the production data retrieval unit is connected to the output ends of the production quantity acquisition unit and the completion time acquisition unit, the output end of the production data retrieval unit is connected to the input end of the time prediction model establishment unit, and the output end of the time prediction model establishment unit is connected to the input end of the completion time prediction unit;

[0062] The production data retrieval unit is used to retrieve the historical production quantity information of semiconductors and the time information spent on completing the production processes of the corresponding quantity of semiconductors in each production stage when producing the corresponding quantity of semiconductors in the past to the time prediction model establishment unit;

[0063] The production time prediction model establishment unit is used to establish a production time prediction model for each production stage;

[0064] The completion time prediction unit is used to obtain the number of semiconductors that need to be produced currently, substitute the number of semiconductors that need to be produced currently into the production time prediction model, and predict the time required to complete the production process of the current number of semiconductors in the current production stage.

[0065] The production equipment data analysis module includes a transmission data reference time analysis unit and an equipment failure probability analysis unit;

[0066] The input end of the transmission data reference time analysis unit is connected to the output end of the production equipment information acquisition unit, and the input end of the equipment failure probability analysis unit is connected to the output ends of the transmission data reference time analysis unit and the completion time prediction unit;

[0067] The transmission data reference time analysis unit is used to analyze, based on the production equipment information, the probability that the equipment used in the semiconductor production of the next production stage of the current production stage fails when it is used in semiconductor production because it directly refers to the equipment inspection data transmitted to the production supervision terminal to determine whether the equipment can operate normally in different equipment inspection cycles, establish a failure probability prediction model for the corresponding equipment, and analyze the transmission time of the corresponding equipment inspection data referred to before the start of the next production stage in the currently set corresponding equipment inspection cycle;

[0068] The equipment failure probability analysis unit is used to substitute the time interval between the transmission time and the start time of the next production stage into the failure probability prediction model, predict the probability that the corresponding equipment will fail when it is used in semiconductor production by directly referring to the equipment inspection data transmitted to the production supervision terminal to determine whether the corresponding equipment can operate normally in the current situation, and use the equipment after obtaining the judgment result.

[0069] The supervision data transmission management module includes a reinspection judgment unit and an inspection data retransmission unit;

[0070] The input end of the reinspection judgment unit is connected to the output end of the equipment failure probability analysis unit, and the output end of the reinspection judgment unit is connected to the input end of the inspection data retransmission unit;

[0071] The re-inspection judgment unit is used to set a failure probability threshold and compare the predicted probability with the threshold: if the predicted probability exceeds the threshold, it is judged that the equipment used in the corresponding production stage needs to be reinspected before the start of the next production stage; if the predicted probability does not exceed the threshold, it is judged that the equipment used in the corresponding production stage does not need to be reinspected before the start of the next production stage, and it is selected to directly refer to the stored equipment inspection data in the production supervision terminal before the start of the next production stage to judge whether the corresponding equipment can operate normally;

[0072] The inspection data retransmission unit is used to, if it is judged that the equipment used in the corresponding production stage needs to be reinspected before the start of the next production stage, transmit the equipment inspection data to the production supervision terminal after reinspecting the equipment, and refer to the equipment inspection data transmitted after reinspecting before the start of the next production stage to judge whether the corresponding equipment can operate normally. After judging that the corresponding equipment can operate normally, use the corresponding equipment to complete the semiconductor production process of the corresponding stage at the start of the next production stage.

[0073] Embodiment 2

[0074] As Figure 2 shown, this embodiment provides an Internet of Things-based semiconductor production data supervision method, which is implemented based on the data supervision system in the embodiment, and specifically includes the following steps:

[0075] Z1: Collect semiconductor historical production information and equipment historical inspection information for semiconductor production. The collected set of historical production quantities for each previous semiconductor production is M = {M1, M2,..., M r}, where r represents the number of historical semiconductor productions. When collecting the time information spent by each production stage to complete the semiconductor production process for the corresponding quantity of semiconductors in previous productions, and collecting the inspection information of the equipment used for semiconductor production in each production stage, including the currently default set equipment inspection cycle, the previously set equipment inspection cycle, and the number of times the equipment has failed when used for semiconductor production due to directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally and obtaining the judgment result.

[0076] Z2: For different production stages, establish a production time prediction model based on semiconductor historical production information to predict the time required for the current production stage to complete the semiconductor production process for the current quantity. It is obtained that the current semiconductor to be produced is currently in the i-th production stage. When retrieving the set of times t = {t1, t2,..., t r} spent by the i-th production stage to complete the semiconductor production process for the corresponding quantity of semiconductors in the previous production set M, for the data points {(M1, t1), (M2, t2),..., (Mr , t r ) perform linear fitting to establish a production time prediction model for predicting the time required to complete the semiconductor production process in the current production stage: , where a and b represent the fitting coefficients of the production time prediction model, and solve for a and b respectively:

[0077] ;

[0078] ;

[0079] where e = 1, 2, …, r, e represents the e-th semiconductor production, the number of semiconductors to be produced currently is N, substitute N into the production time prediction model: let X = N, and predict that the time required to complete the semiconductor production process of the current quantity in the current production stage is ;

[0080] Z3: Analyze the probability of equipment failure when using the equipment after judging the operating state of the equipment by referring to the equipment inspection data stored in the production supervision terminal before the start of the next production stage in the current production stage. Retrieve the set of equipment inspection cycles L = {L1, L2, …, L n} set for semiconductor production in the (i + 1)-th production stage in the past, n represents the number of historical settings of the inspection cycle for the corresponding equipment. Retrieve the set of the number of times H = {H1, H2, …, H n} of equipment failure when the equipment is used for semiconductor production after obtaining the judgment result by directly referring to the corresponding equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally under different inspection cycles in the past. According to calculate that when the inspection cycle is set to L j , the probability P j of equipment failure when the equipment is used for semiconductor production after obtaining the judgment result by directly referring to the corresponding equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally. Obtain the set of failure probabilities as P = {P1, P2, …, P j , …, P n}. Perform linear fitting on the data points {(H1, P1), (H2, P2), …, (H n , P n )} to establish a failure probability prediction model for the equipment used for semiconductor production in the (i + 1)-th production stage: , where and denotes the fitting coefficient of the failure probability prediction model. The start production time of the semiconductor to be currently produced is A1, the start time of the i-th production stage of the semiconductor to be currently produced is A2, the device inspection cycle B1 currently set and used for the semiconductor production in the (i + 1)-th production stage. The device inspection cycle B1 means that the equipment is inspected every interval of duration B1 after the start of the first stage of semiconductor production. The obtained interval duration between the transmission time of the corresponding device inspection data referred to before the start of the (i + 1)-th production stage and the start time of the (i + 1)-th production stage under the currently set corresponding device inspection cycle is , where denotes the floor function of . Let . It is predicted that in the current situation, directly referring to the device inspection data transmitted to the production supervision terminal to judge whether the corresponding device can operate normally. The probability that the corresponding device will fail when it is used in semiconductor production after obtaining the judgment result is p. ;

[0081] Z4: According to the probability analysis result, perform equipment re-inspection and re-inspection data transmission management, and set the failure probability threshold as p ’ , compare p and p ’ : If p > p ’ , judge that it is necessary to re-inspect the equipment used in the corresponding production stage before the start of the (i + 1)-th production stage. After re-inspecting the equipment, transmit the equipment inspection data to the production supervision terminal, and refer to the equipment inspection data transmitted after re-inspection before the start of the (i + 1)-th production stage to judge whether the corresponding device can operate normally; if p ≤ p ’ , judge that it is not necessary to re-inspect the equipment used in the corresponding production stage before the start of the (i + 1)-th production stage. Choose to directly refer to the corresponding equipment inspection data stored in the production supervision terminal before the start of the (i + 1)-th production stage to judge whether the corresponding device can operate normally. After judging that the corresponding device can operate normally, use the corresponding device to complete the semiconductor production process of the corresponding stage at the start of the (i + 1)-th production stage;

[0082] For example: If the current production stage is lithography, set the failure probability threshold as p ’ = 0.4. It is predicted that in the current situation, directly referring to the device inspection data transmitted to the production supervision terminal to judge whether the corresponding device can operate normally. The probability that the corresponding device will fail when it is used in semiconductor production after obtaining the judgment result is p = 0.6, p > p ’, it is necessary to recheck the thin film deposition device before the start of the thin film deposition production stage. After rechecking the thin film deposition device, the equipment inspection data is transmitted to the production supervision terminal. Before the start of the production stage of the thin film deposition device, refer to the equipment inspection data transmitted after rechecking to determine whether the thin film deposition device can operate normally. After determining that the thin film deposition device can operate normally, use the thin film deposition device to complete the semiconductor production process of the corresponding stage at the start of the production stage of the thin film deposition device.

[0083] Finally, it should be noted that the above are only preferred examples of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A semiconductor production data supervision method based on the Internet of Things, characterized in that: Including the following steps: Z1: Collect the historical semiconductor production information and the historical equipment inspection information for semiconductor production; Z2: For different production stages, establish a production time prediction model based on the historical semiconductor production information, and predict the time required to complete the current number of semiconductor production processes in the current production stage; Z3: Analyze the probability of equipment failure when using the equipment after judging the operating state of the equipment by referring to the equipment inspection data stored in the production supervision terminal before the start of the next production stage in the current production stage; Z4: Conduct equipment re-inspection and re-inspection data transmission management according to the probability analysis results; In step Z1: The historical production quantity set collected for each previous production of the semiconductor is M = {M1, M2, …, M r}, where r represents the historical production times of the semiconductor. When collecting the corresponding quantity of semiconductors in previous productions, the time information spent on completing the production processes of the corresponding quantity of semiconductors in each production stage is collected, and the inspection information of the equipment used for semiconductor production in each production stage is collected, including the currently default set equipment inspection cycle, the previously set equipment inspection cycle, and the number of times the equipment has failed when it is used for semiconductor production after obtaining the judgment result by directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally during the corresponding inspection cycle; In step Z2: It is obtained that the semiconductor currently to be produced is in the i-th production stage. When retrieving the corresponding number of semiconductors in the previous production set M, the time set for the i-th production stage to complete the production process of the corresponding number of semiconductors is t = {t1, t2,..., t r}, for the data points {(M1, t1), (M2, t2),..., (M r , t r )}, perform linear fitting to establish a production time prediction model for predicting the time required to complete the semiconductor production process in the current production stage: , where a and b represent the fitting coefficients of the production time prediction model, and solve a and b respectively: ； ； Among them, e = 1, 2, …, r. e represents the e-th semiconductor production. The number of semiconductors that need to be produced currently is N. Substitute N into the production time prediction model: Let X = N, and the predicted time required to complete the production process of the current number of semiconductors in the current production stage is ; In step Z3: The set L = {L1, L2, …, L n} of the equipment inspection cycles that have been set in the past and are used for semiconductor production in the (i + 1)-th production stage is retrieved. n represents the number of historical settings of the inspection cycles for the corresponding equipment. The set H = {H1, H2, …, H n} of the number of times the equipment malfunctioned when used for semiconductor production after obtaining the judgment result by directly referring to the inspection data of the corresponding equipment transmitted to the production supervision terminal at different inspection cycles in the past is retrieved. According to Calculate to obtain that when the inspection cycle is set to L j , the probability P j that the equipment malfunctions when used for semiconductor production after obtaining the judgment result by directly referring to the inspection data of the corresponding equipment transmitted to the production supervision terminal. The failure probability set is P = {P1, P2, …, P j , …, P n}. Perform linear fitting on the data points {(H1, P1), (H2, P2), …, (H n , P n )} to establish a failure probability prediction model for the equipment used for semiconductor production in the (i + 1)-th production stage: , where and represent the fitting coefficients of the failure probability prediction model. The start production time of the semiconductor that needs to be produced currently is obtained as A1, the start time of the i-th production stage of the semiconductor that needs to be produced currently is A2, and the currently default set equipment inspection cycle for semiconductor production in the (i + 1)-th production stage is B1. The time interval from the transmission time of the inspection data of the corresponding equipment referred to before the start of the (i + 1)-th production stage to the start time of the (i + 1)-th production stage under the currently set inspection cycle of the corresponding equipment is obtained as , where represents taking the floor of . Let . Predict that in the current situation, the probability p that the corresponding equipment malfunctions when used for semiconductor production after obtaining the judgment result by directly referring to the inspection data of the equipment transmitted to the production supervision terminal is .

2. The semiconductor production data supervision method based on the Internet of Things according to claim 1, characterized in that: In step Z4: Set the failure probability threshold to p ’ , compare p and p ’ : If p > p ’ , it is determined that the equipment used in the corresponding production stage needs to be reinspected before the start of the (i + 1)-th production stage. After reinspecting the equipment, the equipment inspection data is transmitted to the production supervision terminal, and whether the corresponding equipment can operate normally is judged with reference to the equipment inspection data transmitted after reinspection before the start of the (i + 1)-th production stage; if p ≤ p ’ , it is determined that the equipment used in the corresponding production stage does not need to be reinspected before the start of the (i + 1)-th production stage. It is selected to directly refer to the stored corresponding equipment inspection data in the production supervision terminal to judge whether the corresponding equipment can operate normally before the start of the (i + 1)-th production stage. After judging that the corresponding equipment can operate normally, the corresponding equipment is used to complete the semiconductor production process of the corresponding stage at the start of the (i + 1)-th production stage.

3. A semiconductor production data supervision system based on the Internet of Things, which is applied to a semiconductor production data supervision method based on the Internet of Things as described in claim 1, and is characterized in that: The system includes: a production data collection module, a time transition prediction module, a production equipment data analysis module, and a supervision data transmission management module; The output end of the production data collection module is connected to the input ends of the time transition prediction module and the production equipment data analysis module, the output end of the time transition prediction module is connected to the input end of the production equipment data analysis module, and the output end of the production equipment data analysis module is connected to the input end of the supervision data transmission management module; The production data collection module is used to collect the historical semiconductor production information and the historical equipment inspection information for semiconductor production; The time transition prediction module is used to establish a production time prediction model based on the historical semiconductor production information for different production stages, and predict the time required to complete the current number of semiconductor production processes in the current production stage; The production equipment data analysis module is used to analyze the probability of equipment failure when using the equipment after judging the operating state of the equipment by referring to the equipment inspection data stored in the production supervision terminal before the start of the next production stage in the current production stage; The supervision data transmission management module is used to conduct equipment re-inspection and re-inspection data transmission management according to the probability analysis results.

4. The semiconductor production data supervision system based on the Internet of Things according to claim 3, characterized in that: The production data collection module includes a production quantity collection unit, a completion time collection unit, and a production equipment information collection unit; The production quantity collection unit is used to collect the historical production quantity of semiconductors; The completion time collection unit is used to collect the time information spent on completing the corresponding number of semiconductor production processes in each production stage when producing the corresponding number of semiconductors in the past; The production equipment information collection unit is used to collect the inspection information of the equipment used for semiconductor production in each production stage, including the currently default set equipment inspection cycle, the previously set equipment inspection cycle, and the number of times the equipment fails when used for semiconductor production after obtaining the judgment result by directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally under the corresponding inspection cycle; 5. The semiconductor production data supervision system based on the Internet of Things according to claim 4, characterized in that: The time transition prediction module includes a production data retrieval unit, a time prediction model establishment unit, and a completion time prediction unit; The input end of the production data retrieval unit is connected to the output ends of the production quantity collection unit and the completion time collection unit, the output end of the production data retrieval unit is connected to the input end of the time prediction model establishment unit, and the output end of the time prediction model establishment unit is connected to the input end of the completion time prediction unit; The production data retrieval unit is used to retrieve the historical production quantity information of semiconductors and the time information spent on the production processes of corresponding quantities of semiconductors in each production stage when producing corresponding quantities of semiconductors in the past to the time prediction model establishment unit; The time prediction model establishment unit is used to establish a production time prediction model for each production stage; The completion time prediction unit is used to obtain the quantity of semiconductors that need to be produced currently, substitute the quantity of semiconductors that need to be produced currently into the production time prediction model, and predict the time required to complete the production processes of the current quantity of semiconductors in the current production stage.

6. The semiconductor production data supervision system based on the Internet of Things according to claim 5, wherein: The production equipment data analysis module includes a transmission data reference time analysis unit and an equipment failure probability analysis unit; The input end of the transmission data reference time analysis unit is connected to the output end of the production equipment information collection unit, and the input end of the equipment failure probability analysis unit is connected to the output ends of the transmission data reference time analysis unit and the completion time prediction unit; The transmission data reference time analysis unit is used to analyze, based on the production equipment information, the probability that the equipment used in the next production stage of the current production stage fails when used in semiconductor production due to directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the equipment can operate normally in different equipment inspection cycles, establish a failure probability prediction model for the corresponding equipment, and analyze the transmission time of the corresponding equipment inspection data referred to before the start of the next production stage under the currently set corresponding equipment inspection cycle; The equipment failure probability analysis unit is used to substitute the interval duration between the transmission time and the start time of the next production stage into the failure probability prediction model, predict the probability that the corresponding equipment fails when used in semiconductor production by directly referring to the equipment inspection data transmitted to the production supervision terminal to judge whether the corresponding equipment can operate normally and then using the equipment after obtaining the judgment result; 7. The semiconductor production data supervision system based on the Internet of Things according to claim 6, characterized in that: The supervision data transmission management module includes a re-inspection judgment unit and an inspection data retransmission unit; The input end of the re-inspection judgment unit is connected to the output end of the equipment failure probability analysis unit, and the output end of the re-inspection judgment unit is connected to the input end of the inspection data retransmission unit; The re-inspection judgment unit is used to set a fault probability threshold and compare the predicted probability with the threshold: if the predicted probability exceeds the threshold, it is judged that the equipment used in the corresponding production stage needs to be reinspected before the start of the next production stage; if the predicted probability does not exceed the threshold, it is judged that the equipment used in the corresponding production stage does not need to be reinspected before the start of the next production stage, and it is selected to directly refer to the stored inspection data of the corresponding equipment in the production supervision terminal before the start of the next production stage to judge whether the corresponding equipment can operate normally; The inspection data retransmission unit is used to, if it is judged that the equipment used in the corresponding production stage needs to be reinspected before the start of the next production stage, transmit the equipment inspection data to the production supervision terminal after the equipment is reinspected, refer to the equipment inspection data transmitted after the reinspection before the start of the next production stage to judge whether the corresponding equipment can operate normally, and after judging that the corresponding equipment can operate normally, use the corresponding equipment to complete the semiconductor production process of the corresponding stage at the start of the next production stage.

Images