Product process scheduling method and device and storage medium
By calculating the index value of the intersection scaling ratio and the degree of emergency level, the priority of the semiconductor wafer manufacturer's products is determined, it solves the problem of difficult emergency in the global mixed scheduling, and improves the accuracy and production efficiency of scheduling.
Patent Information
- Application Number
- CN202310922291.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-25
- Publication Date
- 2025-01-28
AI Technical Summary
When semiconductor wafer manufacturers conduct global mixed-order scheduling, it is difficult for them to accurately compare the urgency between different processes, resulting in insufficient accuracy and timeliness of scheduling, thereby reducing production efficiency.
By obtaining the production -related information of each batch of products, calculate the expiration scaling ratio, urgent level index value and backwardness index value, determine the priority of the product, so as to perform process scheduling.
Improving the accuracy and timelyness of process scheduling, reducing the demand for artificial adjustment, and improving production efficiency.
Smart Images
Figure CN119378833A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of semiconductor technology, and in particular to a product process scheduling method, device and storage medium. Background Art
[0002] Product process scheduling is a decision-making process that is widely used in the manufacturing field. The so-called product process scheduling is to match production resources and production tasks with the purpose of optimizing single or multiple goals. Generally speaking, the optimization goals of product process scheduling mainly cover delivery time, output, and machine efficiency. Among them, delivery time is the most important optimization goal, especially when manufacturing is shifting from mass production to mass customization.
[0003] In the semiconductor manufacturing process, the wafer production line needs to process multiple batches and quantities of products at the same time, and products with different technology nodes, different production processes, and different production delivery dates all have a full production process of hundreds of steps. Due to the characteristics of semiconductor processes such as many production steps, reentry, batch processing, and random failures, the complexity and difficulty of process scheduling in semiconductor wafer manufacturing plants has increased dramatically, especially for semiconductor wafer manufacturing plants with medium production scales and high product complexity. Therefore, in order to achieve the delivery date of the product smoothly, it is usually necessary to schedule the product process based on the production resources of the entire plant, that is, global mixed scheduling.
[0004] After research, it was found that the difficulty of global mixed scheduling in semiconductor wafer manufacturing plants lies in breaking through the consistent comparison between different processes based on the urgency of their respective delivery dates.
[0005] At present, when comparing the urgency of each process based on the actual situation of the whole process, the urgency of the process is usually distinguished manually. Since the process needs to be adjusted manually at any time based on the urgency, there are defects in accuracy and timeliness. At the same time, the entire scheduling process will be terminated due to manual adjustments, which greatly reduces production efficiency. Summary of the invention
[0006] The problem to be solved by the present invention is: how to schedule the product process to improve the accuracy and timeliness of the scheduling and improve the production efficiency.
[0007] To solve the above problems, an embodiment of the present invention provides a product process scheduling method, which includes:
[0008] Obtain production-related information for each batch of products in process;
[0009] Based on the obtained production-related information, a first index value and a second index value of each batch of products in the work-in-progress are calculated; the first index value is used to represent the urgency of the product delivery demand, and the second index value is used to represent the degree of lag of the current actual processing time of the product relative to the accumulated standard time of each production step;
[0010] The priority levels of the various batches of products in process are determined in combination with the first indicator value and the second indicator value of the various batches of products in process, so as to perform process scheduling for the various batches of products in process according to the priority levels of the various batches of products in process.
[0011] Optionally, the calculating the first index value and the second index value of each batch of products in process based on the acquired production-related information includes:
[0012] Based on the obtained production-related information, a delivery date scaling ratio of the batch of products corresponding to the work-in-progress is calculated, where the delivery date scaling ratio is a scaling ratio of the production cycle requirement corresponding to the actual delivery date to the accumulated standard time of each production step;
[0013] Based on the delivery time scaling ratio of the corresponding batch of products of the work in progress, the ratio of the remaining time of the corresponding batch of products to the cumulative time from the current production location to shipment is calculated as the first indicator value;
[0014] Based on the delivery date scaling ratio of the corresponding batch of products of the work in progress, the ratio of the current actual processing time of the corresponding batch of products to the accumulated time of each production step is calculated as the second indicator value.
[0015] Optionally, the production information of each batch of products in process includes: current time, offline time, target production delivery date, inventory time and production process information.
[0016] Optionally, the production process information includes: total yellow light layer number information, each production step, standard time for completing each production step, and current production location.
[0017] Optionally, the standard time for completing each production step is read from a preset standard time database.
[0018] Optionally, the standard time database is automatically updated by a timing trigger or an event trigger.
[0019] Optionally, the target production delivery date is selected from a first production delivery date and a second production delivery date; the first production delivery date is a production delivery date calculated based on a quantity-based and point-based delivery mode, and the second production delivery date is a production delivery date calculated based on the product offline time and the promised production cycle indicator.
[0020] Optionally, the target production delivery date is the shorter production delivery date between the first production delivery date and the second production delivery date.
[0021] Optionally, combining the first index value and the second index value of each batch of products in the process to determine the priority level of each batch of products in the process includes:
[0022] Determine the production sequence of the corresponding batches of products by combining the first index value and the second index value of each batch of products in the process and sort them;
[0023] Based on the sorting results of the batches of products in process, the priority levels of the batches of products in process are determined.
[0024] Optionally, combining the first index value and the second index value of each batch of products in the process to determine the production order of the corresponding batch of products for sorting includes:
[0025] Arrange the products of each batch in the process in ascending order according to the first index value, and obtain the first sorting result of each batch in the process;
[0026] Arrange the products in descending order according to the ratio between the second index value of each batch of products in process and the number of currently completed production steps to obtain a second sorting result;
[0027] The first sorting result and the second sorting result are linearly combined to obtain the final sorting results of each batch of products in process.
[0028] Optionally, linearly combining the first sorting result and the second sorting result to obtain the final sorting result of each batch of products in process includes:
[0029] Calculating a first product of the first sorting result and a first weight;
[0030] Calculating a second product of the second sorting result and a second weight; the sum of the first weight and the second weight is 1;
[0031] The first product and the second product are summed, and the production sequence of each batch of products in the process is sorted according to the size of the summed result to obtain the final sorting result of each batch of products in the process.
[0032] Optionally, the production process is divided into a preset number of production stages, and the difference between the mean product rankings of any two production stages is less than a preset threshold; wherein the product ranking mean is the average value of the rankings of all batches of products in the same production stage.
[0033] Optionally, the first weight is 0.3, and the second weight is 0.7.
[0034] Optionally, the first weight and the second weight are both 0.5.
[0035] Optionally, the priority levels include: a main priority level and a sub-priority level.
[0036] Optionally, determining the priority level of each batch of products in the process based on the sorting results of each batch of products in the process includes:
[0037] Based on the sorting results of each batch of products in the work-in-progress and in combination with the quantity of each batch of products in the work-in-progress, determine the primary priority level of each batch of products in the work-in-progress;
[0038] For each batch of products corresponding to the same main priority level, the sub-priority level of each batch of products in process is determined based on the sorting results of each batch of products in process.
[0039] The embodiment of the present invention further provides a product process scheduling device, the device comprising:
[0040] An acquisition unit, adapted to acquire production-related information of each batch of products in process;
[0041] a calculation unit, adapted to calculate a first index value and a second index value of each batch of products in the work-in-progress based on the acquired production-related information; the first index value is used to characterize the urgency of the product delivery demand, and the second index value is used to characterize the degree of lag of the current actual processing time of the product relative to the accumulated standard time of each production step;
[0042] The priority determination unit is suitable for determining the priority level of each batch of products in process in combination with the first indicator value and the second indicator value of each batch of products in process, so as to schedule the process of each batch of products in process according to the priority level of each batch of products in process.
[0043] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. The computer program is executed by a processor to implement the steps of any of the above methods.
[0044] Compared with the prior art, the technical solution of the embodiment of the present invention has the following advantages:
[0045] The scheme of the present invention is applied, based on the obtained production-related information, the first index value and the second index value of each batch of products in the work-in-progress are calculated, wherein the first index value is used to characterize the urgency of the product delivery demand, and the second index value is used to characterize the degree of lag of the current actual processing time of the product relative to the accumulated standard time of each production step. Subsequently, based on the first index value and the second index value, the priority level of each batch of products in the work-in-progress can be reasonably determined, so that the process scheduling of each batch of products in the work-in-progress can be performed according to the priority level of each batch of products in the work-in-progress. Since there is no need to manually adjust the process based on the urgency during the entire scheduling process, the process scheduling can be made more timely, thereby effectively improving production efficiency. Moreover, when determining the priority level of each batch of products, each batch of products has a corresponding first index value and a second index value, and the ranking of each product is determined by the first index value and the second index value at the same time, so as to avoid only considering the urgency of some products when manually adjusting the process and affecting the accuracy of scheduling. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Figure 1 It is a schematic diagram of the actual production process of a semiconductor wafer manufacturing plant;
[0047] Figure 2 A flowchart of a product process scheduling method provided by an embodiment of the present invention;
[0048] Figure 3 A schematic diagram of the production process of a batch of products in process along the actual timeline and the standard timeline;
[0049] Figure 4 It is a schematic diagram for comparing the change curves of the ranking of each production stage under different situations;
[0050] Figure 5 It is a schematic diagram comparing the final sorting results of each batch of products in process according to an embodiment of the present invention with the results of manual process adjustment according to the urgency of production delivery date;
[0051] Figure 6 It is a box-and-whisker plot comparison diagram of the comprehensive index of artificially adjusting quantity and product priority mutation using the method of the embodiment of the present invention and the prior art;
[0052] Figure 7 It is a structural schematic diagram of a product process scheduling device in an embodiment of the present invention. DETAILED DESCRIPTION
[0053] In the actual production scenarios of semiconductor wafer manufacturing plants with high product complexity, products with different technology nodes, different production processes, and different production delivery times all have a full production process of hundreds of steps.
[0054] Figure 1 This is a schematic diagram of the actual production process of a semiconductor wafer manufacturing plant. Figure 1 In actual production scenarios, in a semiconductor wafer manufacturing plant, there are n products with different production delivery dates, where n is an integer greater than 1, and there are m production steps for the n products, where m is an integer greater than 1. The production steps of different products are intertwined with each other to form a complex network of n*m.
[0055] Different products and different production steps of the same product can share the same production resources (machines and people). Nodes that share the same production resources are called shared nodes. How to allocate production resources of shared nodes (different products) becomes a key problem. The rationality and precision of shared node resource allocation will directly determine whether the delivery date of each product can be achieved smoothly. Therefore, global mixed scheduling is needed to truly solve this problem.
[0056] However, in the existing global mixed scheduling process, the urgency of the process is differentiated manually, resulting in defects in accuracy and timeliness. At the same time, the entire scheduling process will be terminated due to manual adjustments, greatly reducing production efficiency.
[0057] To address this problem, the present invention provides a product process scheduling method. According to this method, during the process scheduling process, the priority level of each batch of products in the process is determined by calculating the first indicator value and the second indicator value of each batch of products in the process. Thus, the process scheduling of each batch of products in the process can be carried out according to the priority level of each batch of products in the process, without the need to manually adjust the process based on the degree of urgency. This can make the process scheduling more timely and accurate, and effectively improve production efficiency.
[0058] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.
[0059] Reference Figure 2 The embodiment of the present invention provides a product process scheduling method, which may include the following steps:
[0060] Step 21, obtaining production-related information of each batch of products in process.
[0061] In a specific implementation, the work-in-progress includes products to be processed, products in process, and products in abnormal states. The production information of each batch of the work-in-progress can be obtained from the manufacturing execution system and related databases, or can be manually configured as needed, and the specific acquisition method is not limited.
[0062] In one embodiment of the present invention, the production information of each batch of products in the work-in-progress may include: current time, current production step, offline time, target production delivery date, inventory time and production process information. Among them, the current time is the time corresponding to the current time. The offline time is the time corresponding to when the product is put on the production line and starts processing. The target production delivery date is the time when the product is delivered to the customer. The inventory time is the stagnation time caused by determining whether subsequent production is needed when the corresponding batch of products in the work-in-progress are processed in a certain production step. The production process information is used to characterize the corresponding production step information in the product manufacturing process.
[0063] In a specific implementation, the target production delivery date is an integrated delivery date, which can be specifically selected from the first production delivery date and the second production delivery date. Among them, the first production delivery date is a production delivery date calculated by the quantity and point delivery mode, and the second production delivery date is a production delivery date calculated based on the product offline time and the committed production cycle index. The first production delivery date will not lock each batch of products, that is, for each batch of products, the delivery date requirement is dynamically changing, as long as the number of products delivered to customers meets the requirements. The second production delivery date will lock each batch of products, and the time required for the locked products from offline to delivery is the second production delivery date. For each batch of products, the second production delivery date is fixed. In an embodiment of the present invention, the requirements of the two delivery dates can be combined to determine the target production cycle requirements.
[0064] In one embodiment, the target production cycle is the shorter production delivery period between the first production delivery period and the second production delivery period. For example, if the first production delivery period is 3 months and the second production delivery period is 3.5 months, the first production delivery period can be selected as the target production cycle.
[0065] In other embodiments, a first production delivery date or a second production delivery date may be provided as a target production cycle based on other requirements.
[0066] In a specific implementation, the production process information may include: total yellow light layer information, each production step, the standard time to complete each production step, and the current production position. The total number of yellow light layers can represent the number of yellow light steps required to be experienced during the product manufacturing process. The number of yellow light steps and the standard time required to complete each yellow light step can be used as the cumulative standard time of each production step of the batch of products. Each production step is the production step that the batch of products needs to go through when it is delivered. The standard time to complete each production step is the theoretical time required to complete each production step. The current production position is the production step where the work-in-progress is currently located.
[0067] In a specific implementation, the standard time for completing each production step can be read from a preset standard time database (SSC). Specifically, the standard time corresponding to different production steps can be stored in the standard time database. The standard time database can be divided into standard times for different batches of products under each production step according to technical categories. In each technical category, different matching conditions can be set for the standard times for different batches of products under production steps. Subsequently, the standard time for the corresponding batch of products under the corresponding production step can be read according to the technical category and matching conditions.
[0068] It should be noted that, in a specific implementation, the standard time database is automatically updated by a timing trigger or an event trigger. For example, the timing trigger interval can be set to one quarter, that is, the standard time in the standard time database is updated every quarter. The standard time in the standard time database can also be updated when the matching degree of the number of yellow light layers at a certain technical node or the matching degree of a specific product is less than 95%. By updating the standard time database in a timely manner, the data in the standard time database can be made more accurate, thereby improving the accuracy of process scheduling.
[0069] Step 22, based on the obtained production-related information, calculate the first index value and the second index value of each batch of products in the process.
[0070] The first indicator value is used to characterize the urgency of the product delivery demand, and the second indicator value is used to characterize the degree to which the current actual processing time of the product lags behind the accumulated standard time of each production step.
[0071] In one embodiment, when calculating the first indicator value and the second indicator value of each batch of products in the work-in-progress based on the obtained production-related information, the delivery zoom ratio (Zoom Ratio, ZR) of the corresponding batch of products in the work-in-progress can be first calculated based on the obtained production-related information. The delivery zoom ratio is the zoom ratio of the production cycle requirement corresponding to the actual delivery date and the accumulated standard time of each production step. Then, based on the delivery zoom ratio of the corresponding batch of products in the work-in-progress, the first indicator value and the second indicator value are calculated.
[0072] In a specific implementation, the delivery scaling ratio ZR = production cycle requirement corresponding to the actual delivery / cumulative standard time of each production step. Among them, the production cycle requirement corresponding to the actual delivery can be expressed as: target production delivery date (DueDay) - offline time (Wafer Start Time, WS) - inventory time Bank, that is: Due Day-WS-Bank. The cumulative standard time of each production step can be expressed as: total number of yellow light layers tpl*standard time Spec of a yellow light step, that is: tpl*Spec. Therefore, the delivery scaling ratio ZR can be expressed as:
[0073]
[0074] In a specific implementation, the ratio of the remaining time of the corresponding batch of products to the accumulated time from the current production location to shipment can be used as the first indicator value (Modified Critical Ratio, MCR).
[0075] Specifically, assuming the current time is Sys, the remaining time of the product is the difference between the target production delivery date Due Day and the current time Sys, that is: Due Day-Sys. Assuming the cumulative standard time from the current production location to shipment is Span neg , then the cumulative time from the current production location to shipment is Span neg *ZR. Therefore, the first indicator value MCR can be expressed as:
[0076]
[0077] The first indicator value MCR reflects the urgency of future delivery requirements. The lower the value of the first indicator value MCR is, the more urgent the delivery date of the batch of products is.
[0078] In a specific implementation, the ratio of the current actual processing time of the corresponding batch of products to the cumulative time of each production step can be used as the second indicator value (Gap Ratio, GR). Among them, the current actual processing time of the product can be expressed as: current time Sys-offline time WS-inventory time Bank-accumulated standard time Span from the product off the line to the current production position pos *ZR, namely: Sys-WS-Bank-Span pos *ZR, Span pos *ZR represents the cumulative time from the product coming off the line to the current production position. The cumulative time of each production step can be expressed as the product of the cumulative standard time tpl*Spec of each production step and the delivery scaling ratio ZR, that is: tpl*Spec*ZR. Therefore, the second index value GR can be expressed as:
[0079]
[0080] The second index value GR reflects the degree to which the product processing time is ahead or behind the cumulative time of each production step. The higher the value of the second index value GR is, the more serious the delay in delivery of the batch of products is.
[0081] Figure 3 The following is a production flow diagram of a batch of products in process along the actual timeline and the standard timeline. This batch of products requires a total of m production steps. When it goes off the line, production step 1 starts. The current production step is production step i. The time required from production step 1 to production step m is called the target production delivery time.
[0082] On the actual timeline, the actual time spent from production step 1 to the current production step i is the processing time of the batch of products, which can be expressed as: current time Sys-offline time WS-inventory time Ban, that is: Sys-WS-Bank. The production cycle demand corresponding to the actual delivery date can be expressed as: Due Day-WS-Bank, then the actual processing progress R of the batch of products based on the actual timeline can be expressed as:
[0083]
[0084] On the standard timeline, the standard time from production step 1 to the current production step i is the cumulative standard time span of the processed steps of this batch of products. pos The cumulative standard time of each production step = the cumulative standard time span of the processed steps pos + Cumulative standard time span of unprocessed steps neg , that is: Span pos +Span neg The processing progress R of this batch of products based on the standard timeline can be expressed as:
[0085]
[0086] Using real R and standard R to represent the first index value MCR, the expression of MCR can be obtained as follows:
[0087]
[0088] Using real R and standard R to represent the second index value GR, the expression of GR can be obtained as follows:
[0089] GR=Real R-Standard R (7)
[0090] From formula (6) and formula (7), we can see that the first indicator value MCR is the ratio of two time schedules that have not occurred, focusing on future demand and accurately reflecting the demand for product delivery. The second indicator value GR is the difference between two time schedules that have occurred, focusing on the evaluation of existing performance and fully ensuring the convergence of the product production cycle.
[0091] Step 23, combining the first index value and the second index value of each batch of products in the process, determining the priority level of each batch of products in the process, so as to perform process scheduling for each batch of products in the process according to the priority level of each batch of products in the process.
[0092] In one embodiment of the present invention, the first indicator value and the second indicator value of each batch of products in process can be combined to determine the production order of the corresponding batches of products for sorting, and then based on the sorting results of each batch of products in process, the priority level of each batch of products in process can be determined.
[0093] In a specific implementation, the first index value MCR and the second index value GR are design indexes of two dimensions, which are not in the same order of magnitude and cannot be directly linearly combined, otherwise the shuffling result will have a large deviation.
[0094] To solve this problem, in one embodiment of the present invention, each batch of products in process can be arranged in sequence according to the calculated first index value MCR and the second index value GR, to obtain corresponding sorting results, and then the sorting results can be linearly combined.
[0095] Through research, it is found that based on formula (6), the first index value MCR can be further deduced to obtain the expression between the first index value MCR and the second index value GR as follows:
[0096] MCR=1-b*GR (8)
[0097] in,
[0098] From formula (8), it can be concluded that as the production process of the product progresses, the closer the R is to 1, the value of b will increase sharply. Therefore, when the second index value GR is the same, the later the product is produced, the more rapidly the MCR will increase, and then a serious position effect will occur. The so-called position effect refers specifically to the phenomenon that as the production process of the product progresses, the ranking results of each batch of products gradually become higher or lower.
[0099] In order to eliminate the position effect, in one embodiment of the present invention, the first index value MCR of each batch of products in the process can be used to sort them in ascending order to obtain the first sorting result MCR_SEQ of each batch of products in the process. The second index value GR of each batch of products in the process can be used to sort them in descending order to obtain the second sorting result GR_SEQ according to the ratio between the second index value GR of each batch of products in the process and the number of production steps currently completed. Finally, the first sorting result MCR_SEQ and the second sorting result GR_SEQ are linearly combined to obtain the final sorting result SEQ of each batch of products in the process.
[0100] Assuming that the current production position is production step i, the second sorting result GR_SEQ is the result of sorting the GR / i ratio. GR / i represents the GR averaged to each production step. The second index value GR is the cumulative difference. In extreme cases, the position effect of long-delivery or short-delivery products will be very prominent, resulting in a serious position effect. By sorting the GR / i ratio, the influence of the second index value GR on the final sorting result can be reduced, thereby reducing the position effect.
[0101] In one embodiment, linearly combining the first sorting result and the second sorting result to obtain the final sorting result of each batch of products in the process may include: calculating the first product of the first sorting result MCR_SEQ and the first weight w1 (i.e., MCR_SEQ*w1), calculating the second product of the second sorting result GR_SEQ and the second weight w2 (i.e., GR_SEQ*w2), and then summing the first product and the second product, and sorting the production order of each batch of products in the process according to the size of the summed result to obtain the final sorting result SEQ of each batch of products in the process. The sum of the first weight w1 and the second weight w2 is 1. Among them, the final sorting result SEQ can be expressed as:
[0102] SEQ=MCR_SEQ* w1+ GR_SEQ* w2 (9)
[0103] The study found that the values of the first weight w1 and the second weight w2 have a certain degree of influence on the position effect. Reducing the first weight w1 will greatly reduce the position effect. Figure 4 Provide detailed explanation.
[0104] Reference Figure 4, the entire production process is divided into a preset number of production stages. For example, the entire production process can be divided into 8 production stages in chronological order, namely production stages S1 to S8. Each production stage includes several production steps. Multiple batches of products can be produced in each production stage, and the average value of the final sorting results of all batches of products is calculated respectively. For example, a total of 5 batches of products are produced in the first production stage S1, and the average value of the final sorting results of these 5 batches of products is used as the product sorting mean of the first production stage S1. In this way, the product sorting mean of each production stage can be obtained, and the product sorting mean of each production stage can be connected to form a change curve representing the sorting of each production stage.
[0105] Curve 41 is a ranking change curve obtained based on the ascending order of the first index value MCR and the descending order of the second index value GR, and the first weight w1 = 0.5. Curve 42 is a ranking change curve obtained based on the ascending order of the first index value MCR and the descending order of GR / i, and the first weight w1 = 0.5. Curve 42 is a ranking change curve obtained based on the ascending order of the first index value MCR and the descending order of GR / i, and the first weight w1 = 0.3.
[0106] Comparing curve 41 with curve 42, it can be seen that when the first weight w1 is the same, the final sorting result obtained based on the descending order of the second index value GR has a greater fluctuation than the final sorting result obtained based on the descending order of GR / i, and the position effect is more serious. Comparing curve 42 with curve 43, it can be seen that for the final sorting result obtained based on the descending order of GR / i, the greater the first weight w1, the greater the fluctuation of the curve, and the more serious the position effect.
[0107] In order to reduce the position effect, the first weight w1 value with smaller fluctuation of the mean ranking of products in each stage is preferred. Specifically, the first weight w1 value corresponding to the difference between the mean ranking of products in any two production stages is less than the preset threshold value can be used as the weight for the linear combination of the first indicator value MCR. Those skilled in the art can reasonably set the preset threshold value so that the mean ranking of products in each stage fluctuates less.
[0108] In an embodiment of the present invention, the first weight w1 is 0.3, and the second weight w2 is 0.7.
[0109] In some embodiments, taking into account the special processes of some new products, because their standard time database needs a period of time to be improved and cannot be 100% matched within a period of time, the first weight w1 and the second weight w2 can be selected as 0.5 to ensure that a very small number of products that cannot be calculated and judged by logic will obtain a ranking result with less fluctuation.
[0110] In a specific implementation, when determining the priority level of each batch of products in the process according to the sorting results of each batch of products in the process, the priority level can be finally assigned by comprehensively considering the needs of full differentiation and healthy distribution. For example, the priority number of the corresponding batch of products can be divided into 9 levels. The smaller the final sorting result of the corresponding batch of products, the lower the priority level, and the higher the priority of the dispatch. Products with a priority level of 1 have the highest priority, and products with a priority level of 9 have the lowest priority.
[0111] In a specific implementation, after determining the final ranking results of each batch of products, different priority levels can be assigned to the products according to a certain ratio in combination with the number of each batch of products in the process. For example, the proportion of priority level 1 in each batch of products can be 1%, and the proportion of priority level 2 can be 3%, so as to ensure the smooth and orderly production line. The number of product batches with priority levels 3 to 7 is normally distributed.
[0112] In some embodiments, in order to cope with sudden emergency situations, some priority levels may be reserved for special purposes. For example, priority levels 8 and 9 may be reserved.
[0113] In a specific implementation, after determining the priority level of each product in a corresponding batch, multiple products may have the same priority level. In this case, different priority levels may be set according to the specific production situation of the production line.
[0114] For example, the priority level may include a main priority level and a sub-priority level. Specifically, the main priority level of each batch of products in the work-in-progress may be determined based on the sorting results of each batch of products in the work-in-progress and the number of each batch of products in the work-in-progress. For each batch of products corresponding to the same main priority level, the sub-priority level of each batch of products in the work-in-progress may be determined based on the sorting results of each batch of products in the work-in-progress. Subsequently, the production order of the corresponding batch of products may be determined by combining the main priority and the sub-priority.
[0115] In a specific implementation, the priority levels of the corresponding batches of products can be divided into more levels according to the types of the corresponding batches of products. Accordingly, priority levels with different levels can be set. For example, the priority can be divided into 10 levels, and the priority level can be set to 3 levels. For example, the priority levels can be 1-1-1 and 10-2-5.
[0116] Figure 5 This is a schematic diagram comparing the final sorting results of each batch of products in the work-in-progress according to the embodiment of the present invention with the results of manual process adjustment according to the urgency of the production delivery date. Figure 5 , the horizontal axis represents the product identification of each batch of work in progress.
[0117] According to the method in the embodiment of the present invention, each batch of products has a corresponding first sorting result MCR_SEQ, a second sorting result GR_SEQ and a final sorting result SEQ. The final sorting result SEQ of each batch of products is connected into a curve. The second sorting result GR_SEQ is relatively close to the final sorting result SEQ, and the first sorting result MCR_SEQ is relatively scattered with SEQ, which is a natural result of the first weight being set to 0.3.
[0118] When manually judging the urgency of delivery on-point and in-quantity, the urgency index value of delivery on-point and in-quantity is also calculated. The larger the index value, the higher the urgency of delivery. Among them, the part where the urgency index value of delivery on-point and in-quantity exceeds the threshold is mainly concentrated in the area with a smaller final sorting result SEQ, and the part where the urgency index value of delivery on-point and in-quantity does not exceed the threshold (as shown in the dotted box) is mainly concentrated in the area with a smaller final sorting result SEQ.
[0119] According to statistics, 90% of the urgent products that are manually judged to be delivered on time have their final sorting results SEQ within the top 35% of the final sorting result SEQ curve. Therefore, the global sorting mixed priority can basically meet the needs of manually judged on-time and in-quantity delivery.
[0120] In the existing technology, the process is adjusted manually according to the urgency of the delivery date. In addition to affecting the accuracy and timeliness of process scheduling, additional patch rules need to be set to compensate for the problem of non-global mixing. The introduction of patch rules will cause a sudden change in the sorting results. The results produced in the two situations of triggering and not triggering patch rules will be very different.
[0121] Figure 6 It is a box plot comparison diagram of the comprehensive index of artificially adjusted quantity and product priority mutation using the method of the embodiment of the present invention and the prior art. Among them, the vertical axis represents the comprehensive index value of artificially adjusted quantity and product priority mutation. Box plot 61 is a box plot of the comprehensive index of artificially adjusted quantity and product priority mutation in the prior art. Box plot 62 is a box plot of the comprehensive index of artificially adjusted quantity and product priority mutation in the embodiment of the present invention.
[0122] Depend on Figure 6It can be seen that the upper and lower ends of box plot 61 have a large span, and the data distribution is not concentrated, indicating that there are more mutations in the number of artificial adjustments and product priorities. The upper and lower ends of box plot 62 have a small span, and the data distribution is concentrated, indicating that there are fewer mutations in the number of artificial adjustments and product priorities. In box plot 61, the mean of the comprehensive index of artificial adjustment of the number of products and product priority mutations is about 0.18, and in box plot 62, the mean of the comprehensive index of artificial adjustment of the number of products and product priority mutations is about 0.06. Compared with box plot 61, the improvement of the mean index in box plot 62 is as high as 67.5%, and the improvement of the median index is also 5.1%.
[0123] From the above content, it can be seen that the product process scheduling method in the embodiment of the present invention is a new method for realizing global mixed scheduling in a multi-entry complex manufacturing system represented by a semiconductor wafer manufacturing plant, and integrating it into a system solution to meet the delivery requirements of medium-sized semiconductor manufacturing plants with complex products. This method extracts ideas based on existing theoretical research, comprehensively considers the advantages and disadvantages of ratio processing and difference processing, and considers the completion of the process that has occurred. The delivery target is divided according to the process flow steps, and the key indicators of the global mixed scheduling strategy are independently designed and completed. On the basis of standard time formulation, the key indicators of the global mixed scheduling strategy are combined with the developed product priority system, and finally the guidance of shared node resource allocation is given to realize global mixed scheduling, and it is successfully applied to large-scale actual production, achieving a breakthrough in global mixed scheduling in actual production.
[0124] In order to enable those skilled in the art to better understand and implement the present invention, the device, test system, electronic device and computer-readable storage medium corresponding to the above method are described in detail below.
[0125] Reference Figure 7 The embodiment of the present invention further provides a product process scheduling device 70, which may include: an acquisition unit 71, a calculation unit 72 and a priority determination unit 73. Among them:
[0126] The acquisition unit 71 is adapted to acquire production related information of each batch of products in process;
[0127] The calculation unit 72 is adapted to calculate the first index value and the second index value of each batch of the work-in-progress product based on the acquired production-related information; the first index value is used to represent the urgency of the product delivery demand, and the second index value is used to represent the degree of lag of the current actual processing time of the product relative to the accumulated standard time of each production step;
[0128] The first determination unit 73 is suitable for determining the priority level of each batch of products in process in combination with the first indicator value and the second indicator value of each batch of products in process, so as to perform process scheduling for each batch of products in process according to the priority level of each batch of products in process.
[0129] Regarding the acquisition unit 71, the calculation unit 72 and the priority determination unit 73, they can be implemented specifically with reference to the above description of the product process scheduling method, which will not be repeated here.
[0130] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. The computer program is executed by a processor to implement the steps of any of the above methods.
[0131] In a specific implementation, the computer-readable storage medium may include: ROM, RAM, magnetic disk or optical disk, etc.
[0132] Regarding the various modules / units included in the various devices and products described in the above embodiments, they can be software modules / units, or hardware modules / units, or they can be partially software modules / units and partially hardware modules / units. For example, for various devices and products applied to or integrated in a chip, the various modules / units included therein can all be implemented in the form of hardware such as circuits, or at least some of the modules / units can be implemented in the form of software programs, which run on a processor integrated inside the chip, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in a chip module, the various modules / units included therein can all be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (such as a chip, circuit module, etc.) or different components of the chip module, or at least some of the modules / units can be implemented in the form of hardware such as circuits. The element can be implemented in the form of a software program, which runs on a processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in the terminal, the various modules / units contained therein can be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or in different components in the terminal, or, at least some modules / units can be implemented in the form of a software program, which runs on a processor integrated inside the terminal, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits.
[0133] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined by the claims.
Claims
1. A product process scheduling method, characterized in that: include: Obtain production-related information for each batch of products in process; Based on the obtained production-related information, a first index value and a second index value of each batch of products in the work-in-progress are calculated; the first index value is used to represent the urgency of the product delivery demand, and the second index value is used to represent the degree of lag of the current actual processing time of the product relative to the accumulated standard time of each production step; The priority levels of the various batches of products in process are determined in combination with the first indicator value and the second indicator value of the various batches of products in process, so as to perform process scheduling for the various batches of products in process according to the priority levels of the various batches of products in process.
2. The product process scheduling method according to claim 1, characterized in that: The calculating the first index value and the second index value of each batch of products in process based on the obtained production related information includes: Based on the obtained production-related information, a delivery date scaling ratio of the batch of products corresponding to the work-in-progress is calculated, where the delivery date scaling ratio is a scaling ratio of the production cycle requirement corresponding to the actual delivery date to the accumulated standard time of each production step; Based on the delivery date scaling ratio of the corresponding batch of products of the work in progress, the ratio of the remaining time of the corresponding batch of products to the cumulative time from the current production location to shipment is calculated as the first indicator value; Based on the delivery date scaling ratio of the corresponding batch of products of the work in progress, the ratio of the current actual processing time of the corresponding batch of products to the accumulated time of each production step is calculated as the second indicator value.
3. The product process scheduling method according to claim 2, characterized in that: The production information of each batch of products in process includes: current time, offline time, target production delivery date, inventory time and production process information.
4. The product process scheduling method according to claim 3, characterized in that: The production process information includes: total yellow light layer number information, each production step, the standard time to complete each production step, and the current production location.
5. The product process scheduling method according to claim 4, characterized in that: The standard time for completing each production step is read from a preset standard time database.
6. The product process scheduling method according to claim 5, characterized in that: The standard time database is automatically updated by a timing triggering method or an event triggering method.
7. The product process scheduling method according to claim 3, characterized in that: The target production delivery date is selected from a first production delivery date and a second production delivery date; the first production delivery date is a production delivery date calculated based on a quantity-based and point-based delivery mode, and the second production delivery date is a production delivery date calculated based on the product offline time and the promised production cycle indicator.
8. The product process scheduling method according to claim 7, characterized in that: The target production delivery date is the shorter production delivery date between the first production delivery date and the second production delivery date.
9. The product process scheduling method according to claim 1, wherein: Determining the priority level of each batch of products in the process of manufacture in combination with the first index value and the second index value of each batch of products in the process of manufacture includes: Determine the production sequence of the corresponding batches of products by combining the first index value and the second index value of each batch of products in the process and sort them; Based on the sorting results of the batches of products in process, the priority levels of the batches of products in process are determined.
10. The product process scheduling method according to claim 9, characterized in that: The step of combining the first index value and the second index value of each batch of products in the process to determine the production sequence of the corresponding batch of products for sorting includes: Arrange the products of each batch in the process in ascending order according to the first index value, and obtain the first sorting result of each batch in the process; Arrange the products in descending order according to the ratio between the second index value of each batch of products in process and the number of currently completed production steps to obtain a second sorting result; The first sorting result and the second sorting result are linearly combined to obtain the final sorting results of each batch of products in process.
11. The product process scheduling method according to claim 10, wherein: The linear combination of the first sorting result and the second sorting result to obtain the final sorting result of each batch of products in process includes: Calculating a first product of the first sorting result and a first weight; Calculating a second product of the second sorting result and a second weight; the sum of the first weight and the second weight is 1; The first product and the second product are summed, and the production sequence of each batch of products in the process is sorted according to the size of the summed result to obtain the final sorting result of each batch of products in the process.
12. The product process scheduling method according to claim 11, characterized in that: The production process is divided into a preset number of production stages, and the difference between the mean product rankings of any two production stages is less than a preset threshold; wherein the product ranking mean is the average value of the rankings of all batches of products in the same production stage.
13. The product process scheduling method according to claim 11, wherein: The first weight is 0.3, and the second weight is 0.
7.
14. The product process scheduling method according to claim 11, wherein: The first weight and the second weight are both 0.
5.
15. The product process scheduling method according to claim 9, wherein: The priority levels include: a main priority level and a sub-priority level.
16. The product process scheduling method according to claim 9, wherein: The determining the priority level of each batch of products in the process based on the sorting results of each batch of products in the process includes: Based on the sorting results of each batch of products in the work-in-progress and in combination with the quantity of each batch of products in the work-in-progress, determine the primary priority level of each batch of products in the work-in-progress; For each batch of products corresponding to the same main priority level, the sub-priority level of each batch of products in process is determined based on the sorting results of each batch of products in process.
17. A product process scheduling device, characterized in that: include: An acquisition unit, adapted to acquire production-related information of each batch of products in process; a calculation unit, adapted to calculate a first index value and a second index value of each batch of products in the work-in-progress based on the acquired production-related information; the first index value is used to characterize the urgency of the product delivery demand, and the second index value is used to characterize the degree of lag of the current actual processing time of the product relative to the accumulated standard time of each production step; The priority determination unit is suitable for determining the priority level of each batch of products in process in combination with the first indicator value and the second indicator value of each batch of products in process, so as to schedule the process of each batch of products in process according to the priority level of each batch of products in process.
18. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the steps of the method according to any one of claims 1 to 16.