Semiconductor dynamic production cycle data collection method and apparatus
By integrating multi-dimensional data and prioritizing data entry, the problems of production cycle uncertainty and missing shipment records in semiconductor manufacturing enterprises have been solved, enabling approximate calculation of workpiece production cycles and reasonable production control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI HUALI INTEGRATED CIRCUIT CORP
- Filing Date
- 2022-09-29
- Publication Date
- 2026-05-26
AI Technical Summary
In semiconductor manufacturing companies with multiple platforms and unstable mass production, existing technologies are insufficient to effectively collect production cycle data of workpieces at each process stage, resulting in production cycle uncertainty and missing shipment records, which affects production control.
By acquiring historical shipment records of workpieces on the production line, multi-dimensional data fusion is performed to calculate the production cycle of the workpiece in each process. The production cycle prediction value is fused using a priority filling principle to ensure that the approximate calculation of the workpiece production cycle is achieved under any circumstances.
It enables approximate calculation of the production cycle of a workpiece in each process under any circumstances, solves the problems of production cycle uncertainty and missing shipment records, and ensures reasonable production control.
Smart Images
Figure CN115481982B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor integrated circuit manufacturing, and in particular to a method for collecting dynamic semiconductor production cycle data. This invention also relates to a device for collecting dynamic semiconductor production cycle data. Background Technology
[0002] A flow-shop production line is typically based on a pre-defined product flow. The product flow consists of multiple sequentially arranged steps. After a workpiece enters the production line, it is processed step by step according to the predetermined product flow until all processes are completed.
[0003] Typically, large semiconductor manufacturers, due to their large scale, consistent order volume, and stable mass production, can collect and reuse production cycles for each process step once. However, for semiconductor manufacturers with multiple platforms and unstable mass production, this simple method of collecting production cycles is ineffective.
[0004] At different times, the production cycle of the same workpiece in the same process is uncertain, and simply reusing it will only lead to production control failure.
[0005] It is difficult to guarantee that the workpiece has a record of being shipped at each process within a certain time frame, which may result in a gap in the production cycle of the workpiece at a certain process.
[0006] Table 1
[0007] Step CT(t1) CT(t2) Cumulative CT(t1) 1 0.5 1.2 0.5 2 0.8 0.5 1.3 3 0.4 0.4 1.7 4 0.8 2.5 5 0.7 2.5 6 0.5 0.5 3.0
[0008] As shown in Table 1 above, t1 represents the first time range, t2 represents the second time range, and Step represents the various processes or steps in a workflow within the corresponding time range. CT represents the production cycle, CT(t1) represents the CT corresponding to time period t1, CT(t2) represents the CT corresponding to time period t2, and cumulative CT(t1) represents the cumulative CT of each process corresponding to time period t1. It can be seen that CT(t1) has no record of material handling in Step 5; CT(t2) has no record of material handling in Step 4. Due to the lack of CT(t1) data in Step 5, the cumulative CT(t1) will have a significant deviation. Similarly, if the cumulative CT is calculated based on CT(t2), the final cumulative CT will also have a significant deviation. Summary of the Invention
[0009] The technical problem to be solved by this invention is to provide a method for collecting dynamic semiconductor production cycle data, which can perform multi-dimensional data fusion and effectively ensure that the production cycle of a workpiece at each process step can be approximated under any circumstances, thereby achieving the goal of reasonable production control. To this end, this invention also provides a device for collecting dynamic semiconductor production cycle data.
[0010] To address the aforementioned technical problems, the semiconductor dynamic production cycle data collection method provided by this invention includes the following steps:
[0011] Step 1: Obtain the historical shipment records of each workpiece on the production line at each process and form the first dataset.
[0012] Step 2: Perform production cycle calculations, including:
[0013] Step 21: Calculate the average of the operation time, waiting time and stagnation time of the same product in the same process equipment group at each process stage based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form the first production cycle.
[0014] Step 22: Based on the data in the first dataset and by changing the product, process platform, or process stage, calculate the average of the operation time, waiting time, and stagnation time of the same group of process equipment, and add the average of the operation time, the waiting time, and the stagnation time to form an auxiliary production cycle.
[0015] Step 3: According to the priority filling principle, the production cycle is integrated to obtain the predicted production cycle value of the unworked workpiece in each process.
[0016] The priority filling principle is determined by the number of products, process platforms, process stages, and process equipment groups of the same type. The more products of the same type, the higher the priority level. The first production cycle has the highest priority level.
[0017] For a process of the unworked workpiece, if the first production cycle exists, the first production cycle is used as the predicted production cycle value; if the first production cycle does not exist, the auxiliary production cycle is selected as the predicted production cycle value according to the priority filling principle.
[0018] A further improvement is that, in step 22, the auxiliary production cycle includes a first auxiliary production cycle, which is obtained through the following steps:
[0019] Step 22a: Calculate the average of the operation time, waiting time and stagnation time of all products on the same process platform in the same process equipment group at each of the same process stages based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form the first auxiliary production cycle.
[0020] A further improvement is that, in step 22, the auxiliary production cycle includes a second auxiliary production cycle, which is obtained through the following steps:
[0021] Step 22b: Calculate the average of the operation time, waiting time and stagnation time of all products in the same process equipment group for each of the same process stages based on the data in the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form a second auxiliary production cycle. The priority of the first auxiliary production cycle is greater than the priority of the second auxiliary production cycle.
[0022] A further improvement is that, in step 22, the auxiliary production cycle includes a third auxiliary production cycle, which is obtained through the following steps:
[0023] Step 22c: Calculate the average of the operation time, waiting time and stagnation time of all products in the same process equipment group based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form a third auxiliary production cycle. The priority of the second auxiliary production cycle is greater than the priority of the third auxiliary production cycle.
[0024] A further improvement is that, in step one, the historical shipment records include: the process platform (tech), product name (product_name), workpiece name (lot_name), process flow (flow), process equipment group (Capability), process stage (stage_name), process sequence (step_sequence), quantity (quality), operation time (rt, run_time), waiting time (qt, queue_time), and dwell time (ht, hold_time). The first dataset can be represented as T{t1, t2…tn}, where t1 to tn represent the various data collected.
[0025] To solve the above-mentioned technical problems, the semiconductor dynamic production cycle data collection device provided by the present invention includes:
[0026] The data preparation module is used to obtain the historical shipment records of each workpiece that has been processed on the production line at each process and form the first dataset.
[0027] The production cycle calculation module is used to implement:
[0028] Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of the same product in the same process equipment group at each process stage, and add the average of the operation time, the waiting time, and the downtime to form the first production cycle;
[0029] Based on the data in the first dataset and by transforming the product, process platform, or process stage, calculate the average of the operation time, waiting time, and downtime of the same group of process equipment, and add the average of the operation time, the waiting time, and the downtime to form an auxiliary production cycle.
[0030] The production cycle integration module is used to achieve:
[0031] The production cycle prediction values for unworked workpieces in each process are obtained by integrating the production cycle according to the priority filling principle.
[0032] The priority filling principle is determined according to the number of products, process platforms, process stages and process equipment groups of the same type. The more products of the same type, the higher the priority level. The first production cycle has the highest priority level.
[0033] For a process of the unworked workpiece, if the first production cycle exists, the first production cycle is used as the predicted production cycle value; if the first production cycle does not exist, the auxiliary production cycle is selected as the predicted production cycle value according to the priority filling principle.
[0034] A further improvement is that the auxiliary production cycle includes a first auxiliary production cycle, which is obtained through the following steps:
[0035] Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of all products on the same process platform in the same process equipment group at each of the same process stages, and add the average of the operation time, the waiting time, and the downtime to form the first auxiliary production cycle.
[0036] A further improvement is that the auxiliary production cycle includes a second auxiliary production cycle, which is obtained through the following steps:
[0037] Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of all products in the same process equipment group for each of the same process stages, and add the average of the operation time, the waiting time, and the downtime to form a second auxiliary production cycle. The priority of the first auxiliary production cycle is greater than that of the second auxiliary production cycle.
[0038] A further improvement is that the auxiliary production cycle includes a third auxiliary production cycle, which is obtained through the following steps:
[0039] Based on the data in the first dataset, the average of the operation time, waiting time, and downtime for all the products in the same process equipment group is calculated, and the average of the operation time, waiting time, and downtime is added together to form a third auxiliary production cycle. The priority of the second auxiliary production cycle is greater than that of the third auxiliary production cycle.
[0040] A further improvement is that the historical shipment records include: the process platform, product name, workpiece name, process flow, process equipment group, process stage, process sequence, quantity, operation time, waiting time, and downtime.
[0041] This invention collects historical data, calculates production cycles from multiple dimensions, and fuses the multi-dimensional data. This effectively ensures that the production cycle of a workpiece at each process can be approximately calculated under any circumstances, thereby achieving the goal of reasonable production control.
[0042] This invention is particularly suitable for collecting production cycle data in semiconductor manufacturing companies with multiple platforms and unstable output.
[0043] Firstly, since both the first production cycle and the auxiliary production cycle are formed by averaging multiple data points, it can solve the problem of uncertainty in the production cycle of the same workpiece on the same process at different times in semiconductor manufacturing companies with multiple platforms and unstable output, and the resulting failure in production control.
[0044] Secondly, since the production cycle prediction value formed by the integration of production cycles can always be selected from the first production cycle and the auxiliary production cycle, it can eliminate the defect of missing production cycle of a workpiece in a certain process caused by missing shipment records in various processes of semiconductor manufacturing enterprises with multiple platforms and unstable output. Attached Figure Description
[0045] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:
[0046] Figure 1This is a flowchart of the semiconductor dynamic production cycle data collection method according to an embodiment of the present invention;
[0047] Figure 2 This is a diagram showing the relationship between various dimensions and corresponding production cycle of the semiconductor dynamic production cycle data collection method according to an embodiment of the present invention. Detailed Implementation
[0048] like Figure 1 The diagram shown is a flowchart of a semiconductor dynamic production cycle data collection method according to an embodiment of the present invention; as shown... Figure 2 The diagram shown is a relational graph of various dimensions and corresponding production cycle relationships in the semiconductor dynamic production cycle data collection method according to an embodiment of the present invention. The semiconductor dynamic production cycle data collection method according to an embodiment of the present invention includes the following steps:
[0049] Step 1: Obtain the historical shipment records of each workpiece on the production line at each process and form the first dataset. Figure 1 In this context, step one corresponds to step S102. Step S101 corresponds to the starting step.
[0050] In the method of this invention embodiment, the historical shipment record includes: the process platform, product name, workpiece name, process flow, process equipment group, process stage, process sequence, quantity, operation time, waiting time, and downtime. The first dataset can be represented as T{t1,t2…tn}, where t1 to tn represent the various data collected.
[0051] Step 2: Perform production cycle calculations, including:
[0052] Step 21: Calculate the average of the operation time, waiting time and stagnation time of the same product in the same process equipment group at each process stage based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form the first production cycle.
[0053] Step 22: Based on the data in the first dataset and by changing the product, process platform, or process stage, calculate the average of the operation time, waiting time, and stagnation time of the same group of process equipment, and add the average of the operation time, the waiting time, and the stagnation time to form an auxiliary production cycle.
[0054] In the method of this invention embodiment, the auxiliary production cycle includes a first auxiliary production cycle, which is obtained through the following steps:
[0055] Step 22a: Calculate the average of the operation time, waiting time and stagnation time of all products on the same process platform in the same process equipment group at each of the same process stages based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form the first auxiliary production cycle.
[0056] The auxiliary production cycle includes a second auxiliary production cycle, which is obtained through the following steps:
[0057] Step 22b: Calculate the average of the operation time, waiting time and stagnation time of all products in the same process equipment group for each of the same process stages based on the data in the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form a second auxiliary production cycle. The priority of the first auxiliary production cycle is greater than the priority of the second auxiliary production cycle.
[0058] The auxiliary production cycle includes a third auxiliary production cycle, which is obtained through the following steps:
[0059] Step 22c: Calculate the average of the operation time, waiting time and stagnation time of all products in the same process equipment group based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form a third auxiliary production cycle. The priority of the second auxiliary production cycle is greater than the priority of the third auxiliary production cycle.
[0060] In the method of this invention, the data for calculating various production cycles is represented by dimensions, and the four production cycles are calculated from data in four dimensions. Figure 1 In the process, step two is represented by step S103. It can be seen that the four dimensions are dimension 1, dimension 2, dimension 3 and dimension 4, CT1 represents the first production cycle, CT2 represents the second production cycle, CT3 represents the third production cycle and CT4 represents the fourth production cycle.
[0061] like Figure 2 As shown, data 101 represents data in dimension 1, namely the same product, the same process stage, and the same process equipment group, from which CT1 can be calculated.
[0062] Data 102 represents data in dimension 2, namely, data from the same platform, the same process stage, and the same process equipment group, which can be used to calculate CT2.
[0063] Data 103 represents data in dimension 3, namely the same process stage and the same process equipment group, which can be used to calculate CT3.
[0064] Data 104 represents data in dimension 4, i.e., the same process equipment group, from which CT4 can be calculated.
[0065] The expansion is explained as follows:
[0066] Dimension 1 analyzes the production cycle (CT) CT1 from the perspective of the same product, the same process stage, and the same process equipment group. The set T is grouped by (product_name, stage_name, capability), and then the following indicators are counted in each group:
[0067] Weighted average run time WAVGRT1 = ∑quality*rt / ∑quaIity;
[0068] Weighted average queue time WAVGQT1 = ∑quaIit * qt / ∑quaIity;
[0069] Weighted average hold time WAVGHT1 = ∑quality * ht / ∑quaIity;
[0070] Production cycle CT1 = WAVGRT1 + WAVGQT1 + WAVGHT1.
[0071] Where WAVGRT represents the weighted average job time, WAVGQT represents the weighted average waiting time, WAVGHT represents the weighted average downtime, quality represents the quantity, rt represents the job time, qt represents the waiting time, and ht represents the downtime.
[0072] Dimension 2 analyzes the production cycle CT2 from the perspective of the same platform, the same process stage, and the same process equipment group. The set T is grouped by (tech, stage_name, capability), and then the weighted average operation time WAVGRT2, the weighted average waiting time WAVGQT2, and the weighted average downtime WAGHT2 are calculated to calculate the production cycle CT2.
[0073] Dimension 3 analyzes the production cycle CT3 from the perspective of the same process stage and the same process equipment group. The set T is grouped by (stage_name, capability), and then the weighted average operation time WAVGRT3, weighted average waiting time WAVGQT3, and weighted average standby time WAGHT3 are calculated to calculate the production cycle CT3.
[0074] Dimension 4: Analyze the production cycle CT4 from the perspective of equipment groups with the same process. Group the set T by (capability), and then calculate the weighted average operation time WAVGRT4, weighted average waiting time WAVGQT4, and weighted average downtime WAGHT4 to calculate the production cycle CT4.
[0075] Step 3: According to the priority filling principle, the production cycle is integrated to obtain the predicted production cycle value of the unworked workpiece in each process.
[0076] The priority filling principle is determined by the number of products, process platforms, process stages, and process equipment groups of the same type. The more products of the same type, the higher the priority level. The first production cycle has the highest priority level.
[0077] For a process of the unworked workpiece, if the first production cycle exists, the first production cycle is used as the predicted production cycle value; if the first production cycle does not exist, the auxiliary production cycle is selected as the predicted production cycle value according to the priority filling principle.
[0078] Figure 1 Step three corresponds to step S104, and step S105 is the final step. It can be seen that in the method of this embodiment of the invention, CT1 to CT4 are fused and reconstructed according to a priority filling principle.
[0079] Table 2
[0080] Serial Number Production cycle Fill priority 1 CT1 1 2 CT2 2 3 CT3 3 4 CT4 4
[0081] The priority levels are shown in Table 2. CT1 has the highest priority level, and the priority levels of CT2, CT3 and CT4 decrease in that order. In Table 2, 1 represents the highest priority level, and the larger the value, the lower the priority level.
[0082] The method of this invention can be implemented using either SQL or Python. Taking SQL as an example, a procedure is created, a historical shipment record table is called, CT1, CT2, CT3, and CT4 are calculated according to multiple dimensions, and finally CT is reconstructed according to the priority filling principle.
[0083] By collecting historical data and then calculating the production cycle from multiple dimensions and fusing the multi-dimensional data, this invention can effectively ensure that the production cycle of a workpiece in each process can be approximately calculated under any circumstances, thereby achieving the goal of reasonable production control.
[0084] This invention is particularly suitable for collecting production cycle data from semiconductor manufacturing companies with multiple platforms and unstable production volumes.
[0085] Firstly, since both the first production cycle and the auxiliary production cycle are formed by averaging multiple data points, it can solve the problem of uncertainty in the production cycle of the same workpiece on the same process at different times in semiconductor manufacturing companies with multiple platforms and unstable output, and the resulting failure in production control.
[0086] Secondly, since the production cycle prediction value formed by the integration of production cycles can always be selected from the first production cycle and the auxiliary production cycle, it can eliminate the defect of missing production cycle of a workpiece in a certain process caused by missing shipment records in various processes of semiconductor manufacturing enterprises with multiple platforms and unstable output.
[0087] Because the method of this invention can ensure that the production cycle of the workpiece in each process has an approximate calculated value, it is beneficial to predict the total cumulative operation time of the workpiece. An example is given below:
[0088] Table 3
[0089] step CT1 CT2 CT3 CT4 CT Cumulative CT 1 0.5 0.6 0.4 0.4 0.5 0.5 2 0.2 0.5 0.1 0.3 0.2 0.7 3 0.5 0.6 0.5 0.5 1.2 4 0.4 0.5 0.5 0.4 1.6 5 0.3 0.4 0.3 0.4 0.3 1.9 6 0.7 0.6 0.7 2.6 7 0.1 0.2 0.1 2.7 8 0.3 0.1 0.2 0.3 3.0 9 0.3 0.3 3.3 10 0.4 0.2 0.5 0.3 0.4 3.7
[0090] As shown in Table 3, steps 1, 2, 5, and 10 all have a CT1 value, therefore the CT values for the fusion of steps 1, 2, 5, and 10 all use CT1; the CT values for the fusion of steps 3, 4, and 8 all use CT2; the CT values for the fusion of steps 6 and 7 all use CT3; and the CT values for the fusion of step 9 all use CT4. The cumulative CT is the sum of the CT values of each step. Therefore, the method of this embodiment can obtain a better approximation of the CT values of each step, overcome the defects of CT(t1) or CT(t2) in Table 1, and finally make the cumulative CT more accurate.
[0091] The semiconductor dynamic production cycle data collection device of this invention includes:
[0092] The data preparation module is used to obtain the historical shipment records of each workpiece that has been processed on the production line at each process and form the first dataset.
[0093] In this embodiment of the invention, the historical shipment record includes: the process platform, product name, workpiece name, process flow, process equipment group, process stage, process sequence, quantity, operation time, waiting time, and stoppage time.
[0094] The production cycle calculation module is used to implement:
[0095] Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of the same product in the same process equipment group at each process stage, and add the average of the operation time, the waiting time, and the downtime to form the first production cycle.
[0096] Based on the data in the first dataset and by transforming the product, process platform, or process stage, calculate the average of the operation time, waiting time, and downtime of the same group of process equipment, and add the average of the operation time, the waiting time, and the downtime to form an auxiliary production cycle.
[0097] In this embodiment of the invention, the auxiliary production cycle includes a first auxiliary production cycle, which is obtained through the following steps:
[0098] Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of all products on the same process platform in the same process equipment group at each of the same process stages, and add the average of the operation time, the waiting time, and the downtime to form the first auxiliary production cycle.
[0099] The auxiliary production cycle includes a second auxiliary production cycle, which is obtained through the following steps:
[0100] Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of all products in the same process equipment group for each of the same process stages, and add the average of the operation time, the waiting time, and the downtime to form a second auxiliary production cycle. The priority of the first auxiliary production cycle is greater than that of the second auxiliary production cycle.
[0101] The auxiliary production cycle includes a third auxiliary production cycle, which is obtained through the following steps:
[0102] Based on the data in the first dataset, the average of the operation time, waiting time, and downtime for all the products in the same process equipment group is calculated, and the average of the operation time, waiting time, and downtime is added together to form a third auxiliary production cycle. The priority of the second auxiliary production cycle is greater than that of the third auxiliary production cycle.
[0103] The production cycle integration module is used to achieve:
[0104] The production cycle prediction values for unworked workpieces in each process are obtained by integrating the production cycle according to the priority filling principle.
[0105] The priority filling principle is determined according to the number of products, process platforms, process stages and process equipment groups of the same type. The more products of the same type, the higher the priority level. The first production cycle has the highest priority level.
[0106] For a process of the unworked workpiece, if the first production cycle exists, the first production cycle is used as the predicted production cycle value; if the first production cycle does not exist, the auxiliary production cycle is selected as the predicted production cycle value according to the priority filling principle.
[0107] This invention can be implemented using either SQL or Python. Taking SQL as an example, a procedure is created, a historical shipment record table is called, CT1, CT2, CT3, and CT4 are calculated according to multiple dimensions, and finally CT is reconstructed according to the priority filling principle.
[0108] The embodiments of the present invention can be applied to the following modules, as shown in Table 4:
[0109] serial number Module Application scenarios 1 Workpiece control Predict the production time and delivery time of the workpiece in each process step. 2 Production Regulation Based on short-term forecasts, daily production targets are set and tracked. 3 Investment Based on multiple long-term forecasts, the investment plan was optimized. 4 Production capacity Based on multiple long-term forecasts, capacity early warning and analysis are conducted. 5 Delivery time forecast Based on historical order performance, predict whether the product can be completed on time.
[0110] The present invention has been described in detail above through specific embodiments, but these are not intended to limit the invention. Many modifications and improvements can be made by those skilled in the art without departing from the principles of the invention, and these should also be considered within the scope of protection of the present invention.
Claims
1. A method for collecting dynamic semiconductor production cycle data, characterized in that, Includes the following steps: Step 1: Obtain the historical shipment records of each completed workpiece on the production line at each process stage and form the first dataset; Step 2: Perform production cycle calculations, including: Step 21: Calculate the average of the operation time, waiting time and stagnation time of the same product in the same process equipment group at each process stage based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form the first production cycle; Step 22: Based on the data in the first dataset and by changing the product, process platform, or process stage, calculate the average of the operation time, waiting time, and downtime of the same group of process equipment, and add the average of the operation time, the waiting time, and the downtime to form an auxiliary production cycle; Step 3: According to the priority filling principle, the production cycle is integrated to obtain the predicted production cycle value of the unworked workpiece in each process; The priority filling principle is determined according to the number of products, process platforms, process stages and process equipment groups of the same type. The more products of the same type, the higher the priority level. The first production cycle has the highest priority level. For a process of the unworked workpiece, if the first production cycle exists, the first production cycle is used as the predicted production cycle value; if the first production cycle does not exist, the auxiliary production cycle is selected as the predicted production cycle value according to the priority filling principle.
2. The semiconductor dynamic production cycle data collection method as described in claim 1, characterized in that: In step 22, the auxiliary production cycle includes a first auxiliary production cycle, which is obtained through the following steps: Step 22a: Calculate the average of the operation time, waiting time and stagnation time of all products on the same process platform in the same process equipment group at each of the same process stages based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form the first auxiliary production cycle.
3. The semiconductor dynamic production cycle data collection method as described in claim 2, characterized in that: In step 22, the auxiliary production cycle includes a second auxiliary production cycle, which is obtained through the following steps: Step 22b: Calculate the average of the operation time, waiting time and stagnation time of all products in the same process equipment group for each of the same process stages based on the data in the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form a second auxiliary production cycle. The priority of the first auxiliary production cycle is greater than the priority of the second auxiliary production cycle.
4. The semiconductor dynamic production cycle data collection method as described in claim 3, characterized in that: In step 22, the auxiliary production cycle includes a third auxiliary production cycle, which is obtained through the following steps: Step 22c: Calculate the average of the operation time, waiting time and stagnation time of all products in the same process equipment group based on the data of the first dataset, and add the average of the operation time, the waiting time and the stagnation time to form a third auxiliary production cycle. The priority of the second auxiliary production cycle is greater than the priority of the third auxiliary production cycle.
5. The semiconductor dynamic production cycle data collection method as described in claim 1, characterized in that: In step one, the historical shipment records include: the process platform, product name, workpiece name, process flow, process equipment group, process stage, process sequence, quantity, operation time, waiting time, and downtime.
6. A semiconductor dynamic production cycle data collection device, characterized in that, include: The data preparation module is used to obtain the historical shipment records of each workpiece that has been processed on the production line and its operation in each process and form the first dataset; The production cycle calculation module is used to implement: Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of the same product in the same process equipment group at each process stage, and add the average of the operation time, the waiting time, and the downtime to form the first production cycle; Based on the data in the first dataset and by transforming the product, process platform, or process stage, calculate the average of the operation time, waiting time, and downtime of the same group of process equipment, and add the average of the operation time, the waiting time, and the downtime to form an auxiliary production cycle; The production cycle integration module is used to achieve: The production cycle prediction values for unworked workpieces in each process are obtained by integrating the production cycle according to the priority filling principle. The priority filling principle is determined according to the number of products, process platforms, process stages and process equipment groups of the same type. The more products of the same type, the higher the priority level. The first production cycle has the highest priority level. For a process of the unworked workpiece, if the first production cycle exists, the first production cycle is used as the predicted production cycle value; if the first production cycle does not exist, the auxiliary production cycle is selected as the predicted production cycle value according to the priority filling principle.
7. The semiconductor dynamic production cycle data collection device as described in claim 6, characterized in that: The auxiliary production cycle includes a first auxiliary production cycle, which is obtained through the following steps: Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of all products on the same process platform in the same process equipment group at each of the same process stages, and add the average of the operation time, the waiting time, and the downtime to form the first auxiliary production cycle.
8. The semiconductor dynamic production cycle data collection device as described in claim 7, characterized in that: The auxiliary production cycle includes a second auxiliary production cycle, which is obtained through the following steps: Based on the data in the first dataset, calculate the average of the operation time, waiting time, and downtime of all products in the same process equipment group for each of the same process stages, and add the average of the operation time, the waiting time, and the downtime to form a second auxiliary production cycle. The priority of the first auxiliary production cycle is greater than that of the second auxiliary production cycle.
9. The semiconductor dynamic production cycle data collection device as described in claim 8, characterized in that: The auxiliary production cycle includes a third auxiliary production cycle, which is obtained through the following steps: Based on the data in the first dataset, the average of the operation time, waiting time, and downtime for all the products in the same process equipment group is calculated, and the average of the operation time, waiting time, and downtime is added together to form a third auxiliary production cycle. The priority of the second auxiliary production cycle is greater than that of the third auxiliary production cycle.
10. The semiconductor dynamic production cycle data collection device as described in claim 6, characterized in that: The historical shipment records include: the process platform, product name, workpiece name, process flow, process equipment group, process stage, process sequence, quantity, operation time, waiting time, and downtime.