Control method and control device for a semiconductor production management system
By integrating order and capacity information, developing strategic plans, and optimizing production schedules, the problem of low capacity utilization in semiconductor manufacturing plants was solved, thereby improving production efficiency and corporate revenue.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SEMICON MFG INT (BEIJING) CORP
- Filing Date
- 2024-12-09
- Publication Date
- 2026-06-09
Smart Images
Figure CN122175394A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor manufacturing, and in particular to a control method and control device for a semiconductor production management system. Background Technology
[0002] Because integrated circuits are custom-designed solutions, semiconductor manufacturing plants operate on a Make-to-Order (MTO) model. This means that a semiconductor manufacturing plant begins procurement and production only after receiving a customer order. Therefore, production planning in a semiconductor manufacturing plant is always based on customer orders.
[0003] Production plans for semiconductor manufacturing plants are typically approved by the board of directors. These plans represent the combined capacity of various process platforms within the plant. Due to varying equipment capacity limitations at different process nodes, capacity conversion between different process platforms faces barriers and cannot be achieved through a 1:1 equivalent conversion. Therefore, sufficient orders for each process platform are necessary to maximize the capacity utilization of the semiconductor manufacturing plant and ensure its full capacity is utilized.
[0004] On the other hand, due to the high price of semiconductor equipment, semiconductor manufacturing plants need to increase equipment utilization and maintain it at a high level in order to improve profitability. This places high demands on the control of semiconductor production management systems: how to maximize production capacity in a short period of time, produce high-profit products, and ultimately maximize revenue.
[0005] However, the control of current semiconductor production management systems needs further improvement to enhance production efficiency and corporate revenue. Summary of the Invention
[0006] The problem addressed by this invention is how to improve the production efficiency of semiconductor manufacturing plants and increase corporate revenue.
[0007] To address the above problems, the present invention provides a control method for a semiconductor production management system, comprising:
[0008] Obtain business information and production information of orders; obtain capacity information of semiconductor production lines; based on the business information and production information of the orders, and in conjunction with the capacity information of semiconductor production lines, obtain a strategic plan for the strategic cycle; obtain online information of semiconductor production lines; based on the strategic plan and in conjunction with the online information of semiconductor production lines, obtain a production plan for the production cycle, wherein the production cycle is shorter than the strategic cycle; and issue the production plan to the semiconductor production management system.
[0009] Optionally, the steps of obtaining the business information and production information of the order include: obtaining the product code to obtain the business information of the order, wherein the product information of the order corresponds to the product code; and obtaining the cell code to obtain the production information of the order, wherein the production information of the order corresponds to the cell code.
[0010] Optionally, the steps for obtaining a strategic plan for a strategic cycle include: determining whether to admit all orders based on the business information and production information of the orders, combined with the capacity information of the semiconductor production line; and obtaining a strategic plan for the strategic cycle based on the result of determining whether to admit all orders.
[0011] Optionally, in the step of determining whether to admit all orders and obtaining the strategic plan for the strategic period, the strategic plan for the strategic period is obtained based on all orders.
[0012] Optionally, it also includes: obtaining a production maintenance plan; when it is determined that all orders are not allowed to proceed, the step of obtaining a strategic plan for the strategic cycle further includes: before obtaining the strategic plan for the strategic cycle, obtaining a production selection plan based on the business information and production information of all orders, combined with the production maintenance plan; in the step of obtaining the strategic plan for the strategic cycle, the strategic plan for the strategic cycle is obtained based on the production selection plan and the production maintenance plan.
[0013] Optionally, the step of obtaining the selected production plan includes: obtaining the remaining orders based on the business information and production information of all orders, combined with the maintenance production plan; sorting the remaining orders; and obtaining the selected production plan based on the sorting of the remaining orders.
[0014] Optionally, the method further includes: obtaining the strategic objectives of the strategic cycle; obtaining the evaluation indicators of the orders based on the strategic objectives; and in the step of sorting the remaining orders, sorting the remaining orders based on the evaluation indicators of each order.
[0015] Optionally, in the step of obtaining the strategic objective of the strategic cycle, the strategic objective is profit maximization; in the step of obtaining the evaluation index of the order, the evaluation index is the profit-to-energy ratio.
[0016] Optionally, in the step of obtaining the strategic objective of the strategic cycle, the strategic objective is revenue maximization; in the step of obtaining the evaluation index of the order, the evaluation index is the price-time-energy ratio.
[0017] Optionally, in the step of obtaining the selected production plan, the optimal solution is obtained by linear programming based on the sorting of the remaining orders as the selected production plan.
[0018] Optionally, in the step of obtaining the production plan within the production cycle, the production plan within the production cycle is obtained by combining linear programming and particle swarm optimization based on the strategic plan.
[0019] Optionally, in the step of obtaining the production plan within the production cycle through a combination of linear programming and particle swarm optimization, an initial model is set according to the strategic plan.
[0020] Optionally, in the step of obtaining online information of the semiconductor production line, the online information of the semiconductor production line is obtained in real time; in the step of obtaining the production plan within the production cycle, the step of obtaining the production plan within the production cycle is repeatedly executed with the production cycle as the cycle; after repeatedly executing the step of obtaining the production plan within the production cycle, the step of issuing the production plan to the semiconductor production management system is repeatedly executed.
[0021] Accordingly, the present invention also provides a control device for a semiconductor production management system, comprising:
[0022] The system includes: an information integration module, suitable for obtaining business information and production information of orders; an order admission module, suitable for obtaining capacity information of semiconductor production lines; the order admission module is also suitable for obtaining a strategic plan for a strategic cycle based on the business information and production information of the orders, combined with the capacity information of the semiconductor production lines; an intelligent planning module, suitable for obtaining online information of semiconductor production lines; the intelligent planning module is also suitable for obtaining a production plan within a production cycle, where the production cycle is shorter than the strategic cycle, based on the strategic plan and the online information of the semiconductor production lines; and an output control module, suitable for issuing the production plan to the semiconductor production management system.
[0023] Optionally, the information integration module includes: a first submodule, which is suitable for obtaining business information of the order; and a second submodule, which is suitable for obtaining production information of the order.
[0024] Optionally, the order admission module includes: a first submodule, which is adapted to obtain the capacity information of the semiconductor production line; a judgment submodule, which is adapted to determine whether to admit all orders based on the business information and production information of the orders, combined with the capacity information of the semiconductor production line; and a planning submodule, which is adapted to obtain the strategic plan for the strategic cycle based on the judgment result of the judgment module.
[0025] Optionally, when the judgment submodule determines to admit all orders, the planning submodule obtains the strategic plan for the strategic period based on all orders.
[0026] Optionally, the order admission module further includes: a second submodule, which is adapted to obtain a maintenance production plan; a selection submodule, which is adapted to obtain a selection production plan based on the business information and production information of all orders, combined with the maintenance production plan, when the judgment module determines that all orders should not be admitted; and a planning submodule to obtain a strategic plan for the strategic cycle based on the selection production plan and the maintenance production plan.
[0027] Optionally, the selection submodule includes: a filter, which is adapted to obtain remaining orders based on the business information and production information of all orders, combined with the production maintenance plan; a sorter, which is adapted to sort the remaining orders; and a combiner, which is adapted to obtain the selected production plan based on the sorting of the remaining orders.
[0028] Optionally, it also includes: an evaluation module, which is adapted to obtain the strategic objectives of the strategic cycle and, based on the strategic objectives, obtain the evaluation indicators of the orders; the sorter sorts the remaining orders according to the evaluation indicators of each order.
[0029] Optionally, when the strategic objective is to maximize profits, the evaluation indicator is the price-time-energy ratio; when the strategic objective is to maximize revenue, the evaluation indicator is the price-time-energy ratio.
[0030] Optionally, the intelligent planning module obtains the production plan within the production cycle based on the strategic plan by combining linear programming and particle swarm optimization.
[0031] Optionally, it may also include: an online module, which is suitable for obtaining online information of the semiconductor production line in real time.
[0032] Compared with the prior art, the technical solution of the present invention has the following advantages:
[0033] In this invention, a strategic plan for a strategic cycle is obtained based on the business and production information of orders, combined with the capacity information of the semiconductor production line. Based on the strategic plan and the online information of the semiconductor production line, a production plan for the production cycle is obtained. The strategic plan is obtained based on the business and production information of orders. The production plan obtained from the strategic plan effectively integrates the business and production information of orders, providing an executable solution for improving the efficiency of the semiconductor production management system, optimizing annual revenue, and reducing costs. Attached Figure Description
[0034] Figure 1 This is a flowchart illustrating some embodiments of the control method for the semiconductor production management system of the present invention;
[0035] Figure 2 This is a flowchart illustrating the steps of obtaining business information and production information of an order in some embodiments of the control method of the semiconductor production management system of the present invention.
[0036] Figure 3 This is a flowchart illustrating the steps for obtaining strategic planning for a strategic cycle in some embodiments of the control method of the semiconductor production management system of the present invention.
[0037] Figure 4 This is a flowchart illustrating the steps for obtaining the production plan selection in some embodiments of the control method of the semiconductor production management system of the present invention.
[0038] Figure 5 This is a functional block diagram of some embodiments of the control device of the semiconductor production management system of the present invention. Detailed Implementation
[0039] As can be seen from the background technology, the production efficiency and corporate revenue of semiconductor manufacturing plants in the current technology need to be further improved.
[0040] Generally, in semiconductor manufacturing plants, the corporate planning department provides rolling master production schedules for the next production cycle on a cycle basis. The industrial engineering department calculates static capacity based on shipment demand and standard production cycle to adjust and confirm the master production schedule. The confirmed master production schedule is then published to the entire company and implemented by all employees.
[0041] On the other hand, the sales department determines the sales price of products based on information such as order quantity, process platform, and product type, and obtains demand orders from customers; the production planning department arranges daily production plans statically based on the requirements of the master production schedule, combined with requirements such as product delivery time, product model, product quantity, number of photomask layers, and order information, based on the standard production cycle; the production execution system schedules production based on delivery time requirements to ultimately meet customer needs.
[0042] This shows that in semiconductor manufacturing plants, the business and production information for each order is managed by different departments, resulting in a relatively static production plan that affects the plant's production efficiency and the company's revenue.
[0043] To solve the aforementioned technical problem, the present invention provides a control method for a semiconductor production management system, comprising:
[0044] Obtain business information and production information of the orders; based on the business information and production information of the orders, and in conjunction with the capacity information of the semiconductor production line, obtain a strategic plan for the strategic cycle; based on the strategic plan, and in conjunction with the online information of the semiconductor production line, obtain a production plan within the production cycle, wherein the production cycle is shorter than the strategic cycle; and issue the production plan to the semiconductor production management system.
[0045] In the technical solution of this invention, the strategic plan is obtained based on the business information and production information of the order. The production plan obtained based on the strategic plan can effectively integrate the business information and production information of the order, and can provide an executable solution for improving the efficiency of the semiconductor production management system, annual revenue and cost optimization.
[0046] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0047] refer to Figure 1 The diagram shows a flowchart of some embodiments of the control method of the semiconductor production management system of the present invention.
[0048] The control method includes:
[0049] Step S110: Obtain the business information and production information of the order; Step S109: Obtain the capacity information of the semiconductor production line; Step S120: Based on the business information and production information of the order, and combined with the capacity information of the semiconductor production line, obtain a strategic plan for the strategic cycle; Step S108: Obtain the online information of the semiconductor production line; Step S130: Based on the strategic plan and combined with the online information of the semiconductor production line, obtain a production plan within the production cycle, wherein the production cycle is shorter than the strategic cycle; Step S140: Issue the production plan to the semiconductor production management system.
[0050] The control method can effectively integrate business and production information of orders, and can provide an executable solution for improving the efficiency of semiconductor production management systems, annual revenue, and cost optimization.
[0051] The technical solution of the control method embodiment is described in detail below with reference to the accompanying drawings.
[0052] like Figure 1 As shown, step S110 is executed first to obtain order information, which includes business information and production information.
[0053] Specifically, the business information of the order is suitable to represent commercially relevant information about the order. For example, the business information of the order includes the product code, unit price, average turnover cost, order quantity, process code, and product delivery date.
[0054] The production information for the order is suitable to characterize the order and is related to the product manufacturing process. For example, the production information for the order includes the product code, cell code, quantity, dispatch level, production time required for each process step, waiting time and processing time, total production duration, and process station.
[0055] Reference Figure 2 In some embodiments of the present invention, step S110, obtaining the business information and production information of the order, includes: step S111, obtaining the product code to obtain the business information of the order, wherein the product information of the order corresponds to the product code; step S112, obtaining the cell code to obtain the production information of the order, wherein the production information of the order corresponds to the cell code.
[0056] Using the product code (ProdID) as the unique connection point, it integrates business information such as the unit price of the product, the average turnover cost per unit, the order quantity, the process flow code, and the product delivery date; using the cell code as the unique connection point, it integrates production information such as the product code, quantity, dispatch level, production time required for each process flow, waiting time and pending processing time, total production duration, and process station.
[0057] Continue to refer to Figure 1 After obtaining the business information and production information, step S120 is executed to obtain a strategic plan for the strategic cycle based on the business information and production information of the order.
[0058] Specifically, the strategic plan, namely the Master Production Schedule (MPS) in the strategic cycle, is a production plan based on the strategic objectives of the semiconductor manufacturing plant within the strategic cycle. The strategic plan is obtained based on the business information and production information of the orders, thus it can integrate the business and production information of the orders and effectively ensure the achievement of the strategic objectives.
[0059] Reference Figure 3In some embodiments of the present invention, step S120, obtaining the strategic plan for the strategic cycle, includes: step S121, determining whether to approve all orders based on the business information and production information of the orders, combined with the capacity information of the semiconductor production line; and step S122, obtaining the strategic plan for the strategic cycle based on the result of determining whether to approve all orders. By using the capacity information of the semiconductor production line to determine whether to approve the orders and monitoring the production status of the semiconductor production line, a more reasonable strategic plan can be obtained.
[0060] It should be noted that, as Figure 1 As shown, the control method of the semiconductor production management system further includes step S109, obtaining the capacity information of the semiconductor production line. Specifically, in step S109, obtaining the capacity information of the semiconductor production line, the capacity information of the semiconductor production line can be obtained directly from the semiconductor production line, or indirectly from the semiconductor production management system.
[0061] The capacity information of the semiconductor production line includes: the total capacity of the semiconductor manufacturing plant, the guided capacity combination of each process platform in the semiconductor manufacturing plant, and the single maximum capacity of each process platform.
[0062] Specifically, in step S121, the step of determining whether to approve all orders, the determination is made based on the actual capacity of the semiconductor production line and the capacity requirements of all orders. The actual capacity of the semiconductor production line can be at least one of the total capacity of the semiconductor manufacturing plant, the guided capacity combination of each process platform in the semiconductor manufacturing plant, and the single maximum capacity of each process platform; the capacity requirements of all orders are obtained based on the production information of all orders.
[0063] For example, the actual capacity of the semiconductor production line can be a combination of guided capacity for each process platform in the semiconductor manufacturing plant; the capacity of all orders is obtained based on the quantity of each order and the production site for each order.
[0064] In step S121, the step of determining whether to admit all orders, if the actual capacity of the semiconductor production line can meet the capacity requirements of all orders, that is, if the actual capacity of the semiconductor production line is not less than the capacity requirements of all orders, then all orders are admitted. If all orders are admitted, in step S122, the step of obtaining the strategic plan for the strategic period, the strategic plan for the strategic period is obtained based on all orders.
[0065] On the other hand, in step S121, the step of determining whether to admit all orders, if the actual capacity of the semiconductor production line cannot meet the capacity requirements of all orders, that is, if the actual capacity of the semiconductor production line is less than the capacity requirements of all orders, then it is determined that all orders should not be admitted.
[0066] like Figure 3 As shown, in some embodiments of the present invention, the control method further includes: step S101, obtaining a production maintenance plan; when it is determined that all orders are not allowed to enter, step S120, obtaining a strategic plan for the strategic cycle, further includes: before obtaining the strategic plan for the strategic cycle, executing step S123, obtaining a selected production plan based on the business information and production information of all orders, combined with the production maintenance plan; in step S122, obtaining the strategic plan for the strategic cycle, the strategic plan for the strategic cycle is obtained based on the selected production plan and the production maintenance plan.
[0067] The production maintenance plan includes at least one order, which is an order that must be produced during the strategic cycle.
[0068] Specifically, the production maintenance plan can be specified by the corporate planning department of the semiconductor manufacturing plant. For example, the corporate planning department of the semiconductor manufacturing plant formulates the production maintenance plan based on factors such as the company's long-term planning or strategy and the necessity of supporting key customers; that is, in step S101, the production maintenance plan is obtained from the corporate planning department.
[0069] When the actual capacity of the semiconductor production line cannot meet the production needs of all orders, all orders need to be screened to obtain a strategic plan for the strategic cycle. By screening orders, the semiconductor production line's capacity can be more fully utilized, which is beneficial to maximizing corporate profits.
[0070] Specifically, when it is determined that all orders should not be admitted, step S123 is executed to obtain a production plan based on the business information and production information of all orders, combined with the production maintenance plan.
[0071] The selected production plan includes at least one order, and the order in the selected production plan is any order other than the order in the maintenance production plan. Specifically, in step S123, the step of obtaining the selected production plan, the order in the maintenance production plan is removed from all orders based on the business information and production information of all orders to obtain the selected production plan.
[0072] Reference Figure 4In some embodiments of the present invention, step S123, obtaining the selected production plan based on the business information and production information of all orders, combined with the production maintenance plan, includes: step S1231, obtaining the remaining orders based on the business information and production information of all orders, combined with the production maintenance plan; step S1232, sorting the remaining orders; and step S1233, obtaining the selected production plan based on the sorting of the remaining orders.
[0073] Specifically, in step S1231, the step of obtaining the remaining orders, based on the business information and production information of all orders, the orders in the production maintenance plan are removed from all orders, and the remaining orders are referred to as the remaining orders.
[0074] The order ranking is appropriate to reflect the importance of the orders; the order ranking obtained by integrating business information and production information can provide an actionable solution for improving the efficiency of the semiconductor production management system, annual revenue, and cost optimization.
[0075] like Figure 4 As shown, in some embodiments, the control method further includes: step S102, obtaining the strategic objective of the strategic cycle; step S103, obtaining the evaluation index of the order based on the strategic objective; and step S1232, in the step of sorting the remaining orders, sorting the remaining orders based on the evaluation index of each order.
[0076] Based on the strategic objectives of the aforementioned strategic cycle, appropriate evaluation indicators are determined. Based on the evaluation, all orders are sorted to better assess the importance of different orders under the strategic objectives of the aforementioned strategic cycle. The production plan specified in this way can better and more efficiently achieve the strategic objectives of the aforementioned strategic cycle.
[0077] In some embodiments, the strategic period is a preset period; in some example embodiments, the strategic period is one fiscal year; for example, the strategic period can be 16 months.
[0078] It should be noted that the evaluation metrics obtained in the step of obtaining the order evaluation metrics correspond to the strategic objectives obtained in the step of obtaining the strategic objectives of the strategic cycle. For example, in some embodiments, in the step of obtaining the strategic objectives of the strategic cycle, the strategic objective is profit maximization; in the step of obtaining the order evaluation metrics, the evaluation metrics are profit-time-capacity ratio; in some embodiments, in the step of obtaining the strategic objectives of the strategic cycle, the strategic objective is revenue maximization; in the step of obtaining the order evaluation metrics, the evaluation metrics are price-time-capacity ratio.
[0079] Based on strategic objectives, we propose evaluation indicators for the ratio of efficiency to energy and the ratio of price to energy, which can better evaluate the importance of different orders under different strategic objectives, thereby providing a good quantitative basis for better production planning.
[0080] Specifically, semiconductor manufacturing plants produce a large quantity of products, and the price and profit of these products are not necessarily positively correlated: high-priced products do not necessarily have high profits, but high-priced products are beneficial to increasing the operating income of semiconductor manufacturing plants; high-profit products may have average prices, but because the cost pricing is relatively appropriate, high-profit products are beneficial to increasing the company's profits.
[0081] The performance metrics of time-to-profit ratio and price-to-time ratio are used to differentiate the different impacts of price and profit on different orders:
[0082]
[0083] Wherein, price is the unit price of the product in the business information of the order, i.e., the order quotation; profit is the profit of the order, i.e., the difference between the unit price of the product and the cost of the single average turnover; time is the total production duration of a single product in the order; the capacity coefficient is suitable for characterizing the severity of the contradiction between machine production and sales: the greater the contradiction between machine production and sales involved in the order, the greater the capacity coefficient; the smaller the contradiction between machine production and sales involved in the order, the smaller the capacity coefficient.
[0084] It should be noted that the capacity coefficient can be obtained based on the shadow price of the order. Specifically, the capacity coefficient can be based on dynamic capacity demand, and calculated by taking into account the proportion of the expected usage time of each machine in the static load, and determining the importance of each machine in the calculation period through a fuzzy algorithm, wherein the fuzzy algorithm can be the Delphi algorithm.
[0085] Based on different strategic objectives, appropriate evaluation indicators are determined, and all orders are ranked in order. The resulting strategic plan can preferentially support products or customers with high evaluation indicators, which is conducive to maximizing the company's interests under the strategic objectives.
[0086] Continue to refer to Figure 4 After sorting the remaining orders, step S1233 is executed to obtain the selected production plan based on the sorting of the remaining orders. In some embodiments, in step S1233, the step of obtaining the selected production plan involves obtaining the optimal solution as the selected production plan through linear programming based on the sorting of the remaining orders. Under the constraints of order sorting and capacity, the optimal solution combination is statically calculated through linear programming, and this combination serves as the production target for the strategic cycle, i.e., the strategic plan for the strategic cycle.
[0087] Continue to refer to Figure 3 After obtaining the selected production plan, step S122 is executed to obtain the strategic plan for the strategic cycle. Specifically, in step S122, the strategic plan for the strategic cycle is obtained based on the selected production plan and the maintenance production plan.
[0088] For example, in the step of obtaining a strategic plan for the strategic cycle based on the selected production plan and the maintained production plan, the selected production plan and the maintained production plan are combined to obtain a strategic plan for the strategic cycle.
[0089] Continue to refer to Figure 1 After obtaining the strategic plan for the strategic cycle, step S130 is executed to obtain the production plan within the production cycle based on the strategic plan and the online information of the semiconductor production line, wherein the production cycle is shorter than the strategic cycle.
[0090] It should be noted that, as Figure 1 As shown, the control method of the semiconductor production management system further includes step S108, obtaining online information of the semiconductor production line. Specifically, in step S108, obtaining online information of the semiconductor production line, the online information of the semiconductor production line can be obtained directly from the semiconductor production line, or indirectly from the semiconductor production management system.
[0091] The online information of the semiconductor production line includes information related to the status of the semiconductor production line itself, such as the status of existing order demand, production status, and the process flow of bottleneck machines, as well as the status of products on the semiconductor production line.
[0092] The production plan is obtained based on the strategic plan; and when the production capacity is insufficient to meet the capacity requirements of all orders, the production plan can maximize the company's interests under the strategic objectives while meeting the company's long-term planning or strategy and business requirements such as supporting key customers. It can effectively improve capacity utilization and is conducive to maximizing the company's interests.
[0093] In some embodiments, the production cycle is a preset cycle. In some example embodiments, the production cycle is the wafer start-up cycle of a semiconductor production line; for example, the production cycle is 24 hours.
[0094] In some embodiments of the present invention, in the step of obtaining the production plan within the production cycle, the production plan within the production cycle is obtained by combining linear programming and particle swarm optimization according to the strategic plan.
[0095] Specifically, in the steps of obtaining the production plan within the production cycle through a combination of linear programming and particle swarm optimization, constraints are set based on the production cycle and load rate increase function, and strategic objectives that reflect the strategic direction of different periods. A multi-objective production planning solution is obtained through fuzzy computation.
[0096] The constraints of linear programming can include: long term agreement (LTA) requirements, low-cost, high-return products, and process platforms, etc.
[0097] For example, the basic constraints of linear programming can be the profit-optimal strategic direction and the capacity constraint. The profit-optimal strategic direction is: Max∑ i,t (r it y it +c it x it +h it I it ); Capacity constraints are: ∑ i,t (b i y it +a i x it )≤R t , Where i represents the product name, t represents the unit time of the planning period, and r it y represents negative production costs. it A variable that is either 0 or 1, if product i is produced within the planned period t, y it If product i is not produced within the planned production cycle t, then y = 1. it Take 0, c it x represents the unit selling price. it h represents the output of product i within the planning period t. it I represents the unit inventory cost of product i within the planning period t. it b represents the inventory level of product i within the planning period t. i a represents the production capacity consumption for the preparation activities of product i. i This represents the unit capacity consumption of product i.
[0098] In addition, particle swarm optimization (PSO) is used for production cycle prediction. Specifically, based on the performance of production cycles of various process platforms under different daily load rates over the past five years as a learning standard, PSO and machine learning are used to derive the production cycle under different load rates. Furthermore, by continuously updating data with information such as the daily product mix and total quantity, the system learns and obtains valuable production cycle predictions.
[0099] For example, the particle swarm optimization algorithm involves updating ion velocity and updating particle position as follows:
[0100] Update particle velocity:
[0101] v i,j (t+1)=w*v i,j (t)+c1*r1*[pbest i,j (t)-x i,j (t)]+c2*r2*[gbest i,j (t)-x i,j (t)
[0102] Update particle position to: x i,j (t+1)=x i,j (t)+v i,j (t)
[0103] Where i represents a certain order of magnitude in the population size, j represents a certain dimension in the particle dimension, w represents the inertia factor, t represents a specific moment within the planning period, and v i,j (t+1) represents the position of the particle at time (t+1), c i The positive constant, r, represents the self-awareness learning factor. i pbest represents a random number in the interval [0,1]. i,j (t) represents the particle's optimal solution at time t, i.e., the individual optimal solution, gbest i,j (t) represents the optimal solution found so far in the entire particle swarm, i.e., the global optimal solution, x i,j (t) represents the position of the particle at time t.
[0104] In some embodiments, in the step of obtaining the production plan within the production cycle through a combination of linear programming and particle swarm optimization, an initial model is set according to the strategic plan.
[0105] Continue to refer to Figure 1 After obtaining the production plan for the production cycle, step S140 is executed to issue the production plan to the semiconductor production management system.
[0106] Specifically, the production plan is issued to the semiconductor production management system, and corresponding shipment targets and tape-out guidance assignment levels are assigned based on the customer's delivery time for the product to carry out production.
[0107] In some embodiments of the present invention, in step S108, the step of obtaining online information of the semiconductor production line, the online information of the semiconductor production line is obtained in real time; in step S130, the step of obtaining the production plan within the production cycle, the step of obtaining the production plan within the production cycle is repeatedly executed with the production cycle as the cycle; after repeatedly executing the step of obtaining the production plan within the production cycle, step S140 is repeatedly executed to issue the production plan to the semiconductor production management system.
[0108] Specifically, a production plan is obtained based on the production cycle, and the obtained production plan is rolled out to the production execution system. Based on the customer's delivery time for the product, corresponding delivery targets and tape-out guidance dispatch levels are assigned.
[0109] It should be noted that, in some embodiments of the present invention, the control method further includes: obtaining production line information such as the capacity load status, online equipment load and operating status, and delivery schedule of the semiconductor production line.
[0110] Accordingly, the present invention also provides a control device for a semiconductor production management system.
[0111] refer to Figure 5 The diagram shows functional block diagrams of some embodiments of the control device of the semiconductor production management system of the present invention.
[0112] The control device includes:
[0113] The system comprises: an information integration module 210, adapted to obtain business information and production information of orders; an order admission module 220, adapted to obtain capacity information of semiconductor production lines; the order admission module 220 is also adapted to obtain a strategic plan for a strategic cycle based on the business information and production information of the orders, combined with the capacity information of the semiconductor production lines; an intelligent planning module 230, adapted to obtain online information of semiconductor production lines; the intelligent planning module 230 is also adapted to obtain a production plan within a production cycle, where the production cycle is shorter than the strategic cycle, based on the strategic plan and the online information of the semiconductor production lines; and an output control module 240, adapted to issue the production plan to the semiconductor production management system.
[0114] The technical solution of the control device embodiment will be described in detail below with reference to the accompanying drawings.
[0115] The information integration module 210 is suitable for obtaining business information and production information of orders.
[0116] Specifically, the business information of the order is suitable to represent commercially relevant information about the order. For example, the business information of the order includes the product code, unit price, average turnover cost, order quantity, process code, and product delivery date.
[0117] The production information for the order is suitable to characterize the order and is related to the product manufacturing process. For example, the production information for the order includes the product code, cell code, quantity, dispatch level, production time required for each process step, waiting time and processing time, total production duration, and process station.
[0118] like Figure 5 As shown, in some embodiments of the present invention, the information integration module 210 includes: a first submodule 211, which is adapted to obtain business information of the order; and a second submodule 212, which is adapted to obtain production information of the order.
[0119] Specifically, the first submodule 211 obtains the product code to obtain the business information of the order, and the product information of the order corresponds to the product code; the second submodule 212 obtains the cell code to obtain the production information of the order, and the production information of the order corresponds to the cell code.
[0120] Using the product code (ProdID) as the unique connection point, it integrates business information such as the unit price of the product, the average turnover cost per unit, the order quantity, the process flow code, and the product delivery date; using the cell code as the unique connection point, it integrates production information such as the product code, quantity, dispatch level, production time required for each process flow, waiting time and pending processing time, total production duration, and process station.
[0121] The control device further includes an order admission module 220, which is adapted to obtain the capacity information of the semiconductor production line; the order admission module 220 is also adapted to obtain a strategic plan for the strategic cycle based on the business information and production information of the order, combined with the capacity information of the semiconductor production line.
[0122] Specifically, the strategic plan, namely the Master Production Schedule (MPS) in the strategic cycle, is a production plan based on the strategic objectives of the semiconductor manufacturing plant within the strategic cycle. The strategic plan is obtained based on the business information and production information of the orders, thus it can integrate the business and production information of the orders and effectively ensure the achievement of the strategic objectives.
[0123] like Figure 5 As shown, in some embodiments of the present invention, the order admission module 220 includes: a first submodule 221, which is adapted to obtain the capacity information of the semiconductor production line; a judgment submodule 222, which is adapted to determine whether to admit all orders based on the business information and production information of the orders, combined with the capacity information of the semiconductor production line; and a planning submodule 223, which is adapted to obtain a strategic plan for the strategic cycle based on the judgment result of the judgment module 222. The judgment submodule 222 determines whether to admit all orders and detects the production status of the semiconductor production line through the capacity information of the semiconductor production line, while the planning submodule 223 can obtain a more reasonable strategic plan.
[0124] The first submodule 221 is indirectly connected to the semiconductor production line through the semiconductor production management system, and indirectly obtains the production capacity information of the semiconductor production line through the semiconductor production management system. In other embodiments, the first submodule may also be directly connected to the semiconductor production line, and directly obtain the production capacity information of the semiconductor production line from the semiconductor production line.
[0125] The capacity information of the semiconductor production line includes: the total capacity of the semiconductor manufacturing plant, the guided capacity combination of each process platform in the semiconductor manufacturing plant, and the single maximum capacity of each process platform.
[0126] The judgment submodule 222 determines whether to approve all orders based on the actual capacity of the semiconductor production line and the capacity requirements of all orders. The actual capacity of the semiconductor production line can be at least one of the total capacity of the semiconductor manufacturing plant, the guided capacity combination of each process platform in the semiconductor manufacturing plant, and the single maximum capacity of each process platform; the capacity requirements of all orders are obtained based on the production information of all orders.
[0127] For example, the actual capacity of the semiconductor production line can be a combination of guided capacity for each process platform in the semiconductor manufacturing plant; the capacity of all orders is obtained based on the quantity of each order and the production site for each order.
[0128] When the actual capacity of the semiconductor production line can meet the capacity requirements of all orders, that is, when the actual capacity of the semiconductor production line is not less than the capacity requirements of all orders, the judgment submodule 222 determines to admit all orders, and the planning submodule 223 obtains the strategic plan for the strategic cycle based on all orders.
[0129] When the actual capacity of the semiconductor production line cannot meet the capacity requirements of all orders, that is, when the actual capacity of the semiconductor production line is less than the capacity requirements of all orders, the judgment submodule 222 determines that all orders should not be admitted.
[0130] like Figure 5 As shown, in some embodiments, the order admission module 220 further includes: a second submodule 224, which is adapted to obtain a production maintenance plan; a selection submodule 225, which is adapted to obtain a selection production plan based on the business information and production information of all orders, combined with the production maintenance plan, when the judgment module 222 determines that all orders are not admitted; and the planning submodule 223 obtains the strategic plan for the strategic cycle based on the selection production plan and the production maintenance plan.
[0131] The production maintenance plan includes at least one order, which is an order that must be produced during the strategic cycle.
[0132] Specifically, the production maintenance plan can be specified by the corporate planning department of the semiconductor manufacturing plant. For example, the corporate planning department of the semiconductor manufacturing plant formulates the production maintenance plan based on factors such as the company's long-term planning or strategy and the necessity of supporting key customers; that is, in step S101, the production maintenance plan is obtained from the corporate planning department.
[0133] When the actual capacity of the semiconductor production line cannot meet the production needs of all orders, all orders need to be screened to obtain a strategic plan for the strategic cycle. By screening orders, the semiconductor production line's capacity can be more fully utilized, which is beneficial to maximizing corporate profits.
[0134] Specifically, when the judgment submodule 222 determines that all orders should not be admitted, the selection submodule 225 obtains a production plan based on the business information and production information of all orders, combined with the production maintenance plan.
[0135] The selected production plan includes at least one order, and the selected production plan consists of all orders except those in the maintenance production plan. Specifically, the selection submodule 225 removes the orders in the maintenance production plan from all orders based on the business information and production information of all orders to obtain the selected production plan.
[0136] like Figure 5In some embodiments shown, the selection submodule 225 includes: a filter 225a, which is adapted to obtain remaining orders based on the business information and production information of all orders, combined with the production maintenance plan; a sorter 225b, which is adapted to sort the remaining orders; and a combiner 225c, which is adapted to obtain the selected production plan based on the sorting of the remaining orders.
[0137] Specifically, the filter 225a removes the orders in the production maintenance plan from all orders based on the business information and production information of all orders, and the remaining orders are referred to as the remaining orders.
[0138] The order ranking is appropriate to reflect the importance of the orders; the order ranking obtained by integrating business information and production information can provide an actionable solution for improving the efficiency of the semiconductor production management system, annual revenue, and cost optimization.
[0139] In some embodiments of the present invention, the control device further includes: an evaluation module 201, which is adapted to obtain the strategic objectives of the strategic cycle and, based on the strategic objectives, obtain the evaluation indicators of the orders; and the sorter 225b sorts the remaining orders according to the evaluation indicators of each order.
[0140] Based on the strategic objectives of the aforementioned strategic cycle, appropriate evaluation indicators are determined. Based on the evaluation, all orders are sorted to better assess the importance of different orders under the strategic objectives of the aforementioned strategic cycle. The production plan specified in this way can better and more efficiently achieve the strategic objectives of the aforementioned strategic cycle.
[0141] In some embodiments, the strategic period is a preset period; in some example embodiments, the strategic period is one fiscal year; for example, the strategic period can be 16 months.
[0142] It should be noted that the evaluation indicators obtained by the evaluation module 201 correspond to the strategic objectives obtained in the step of obtaining the strategic objectives of the strategic cycle. For example, in some embodiments, when the strategic objective is profit maximization, the evaluation indicator is the profit-time-capacity ratio; in some embodiments, when the strategic objective is revenue maximization, the evaluation indicator is the price-time-capacity ratio.
[0143] Based on strategic objectives, we propose evaluation indicators for the ratio of efficiency to energy and the ratio of price to energy, which can better evaluate the importance of different orders under different strategic objectives, thereby providing a good quantitative basis for better production planning.
[0144] Specifically, semiconductor manufacturing plants produce a large quantity of products, and the price and profit of these products are not necessarily positively correlated: high-priced products do not necessarily have high profits, but high-priced products are beneficial to increasing the operating income of semiconductor manufacturing plants; high-profit products may have average prices, but because the cost pricing is relatively appropriate, high-profit products are beneficial to increasing the company's profits.
[0145] The performance metrics of time-to-profit ratio and price-to-time ratio are used to differentiate the different impacts of price and profit on different orders:
[0146]
[0147] Wherein, price is the unit price of the product in the business information of the order, i.e., the order quotation; profit is the profit of the order, i.e., the difference between the unit price of the product and the cost of the single average turnover; time is the total production duration of a single product in the order; the capacity coefficient is suitable for characterizing the severity of the contradiction between machine production and sales: the greater the contradiction between machine production and sales involved in the order, the greater the capacity coefficient; the smaller the contradiction between machine production and sales involved in the order, the smaller the capacity coefficient.
[0148] It should be noted that the capacity coefficient can be obtained based on the shadow price of the order. Specifically, the capacity coefficient can be based on dynamic capacity demand, and calculated by taking into account the proportion of the expected usage time of each machine in the static load, and determining the importance of each machine in the calculation period through a fuzzy algorithm, wherein the fuzzy algorithm can be the Delphi algorithm.
[0149] Based on different strategic objectives, appropriate evaluation indicators are determined, and all orders are ranked in order. The resulting strategic plan can preferentially support products or customers with high evaluation indicators, which is conducive to maximizing the company's interests under the strategic objectives.
[0150] After the sorter 225b sorts the remaining orders, the combiner 225c obtains the selected production plan based on the sorting of the remaining orders. In some embodiments, the combiner 225c obtains the optimal solution as the selected production plan through linear programming. Under the constraints of order sorting and capacity, the optimal solution combination is statically calculated through linear programming, and this combination serves as the production target for the strategic cycle, i.e., the strategic plan for the strategic cycle.
[0151] After the combiner 225c obtains the selected production plan, the planning submodule 223 obtains the strategic plan for the strategic period. Specifically, the planning submodule 223 obtains the strategic plan for the strategic period based on the selected production plan and the maintenance production plan. For example, the planning submodule 223 combines the selected production plan and the maintenance production plan to obtain the strategic plan for the strategic period.
[0152] Continue to refer to Figure 5 In the control device, the intelligent planning module 230 is adapted to obtain online information of the semiconductor production line; the intelligent planning module 230 is also adapted to obtain a production plan within the production cycle based on the strategic plan and the online information of the semiconductor production line, wherein the production cycle is shorter than the strategic cycle.
[0153] The intelligent planning module 230 is connected to the semiconductor production line and obtains the online information from the semiconductor production line.
[0154] The online information of the semiconductor production line includes information related to the status of the semiconductor production line itself, such as the status of existing order demand, production status, and the process flow of bottleneck machines, as well as the status of products on the semiconductor production line.
[0155] The production plan is obtained based on the strategic plan; and when the production capacity is insufficient to meet the capacity requirements of all orders, the production plan can maximize the company's interests under the strategic objectives while meeting the company's long-term planning or strategy and business requirements such as supporting key customers. It can effectively improve capacity utilization and is conducive to maximizing the company's interests.
[0156] In some embodiments, the production cycle is a preset cycle. In some example embodiments, the production cycle is the wafer start-up cycle of a semiconductor production line; for example, the production cycle is 24 hours.
[0157] In some embodiments of the present invention, the intelligent planning module 230 obtains the production plan within the production cycle by combining linear programming and particle swarm optimization according to the strategic plan.
[0158] Specifically, the intelligent planning module 230 sets constraints based on the production cycle and load rate increase function, and based on the strategic objectives that reflect the strategic direction of different periods, and obtains a multi-objective production planning solution through fuzzy calculation.
[0159] The constraints of linear programming can include: long term agreement (LTA) requirements, low-cost, high-return products, and process platforms, etc.
[0160] For example, the basic constraints of linear programming can be the profit-optimal strategic direction and the capacity constraint. The profit-optimal strategic direction is: Max∑ i,t (r it y it +c it x it +h it Iit ); Capacity constraints are: ∑ i,t (b i y it +a i x it )≤R t , Where i represents the product name, t represents the unit time of the planning period, and r it y represents negative production costs. it A variable that is either 0 or 1, if product i is produced within the planned period t, y it If product i is not produced within the planned production cycle t, then y = 1. it Take 0, c it x represents the unit selling price. it h represents the output of product i within the planning period t. it I represents the unit inventory cost of product i within the planning period t. it b represents the inventory level of product i within the planning period t. i a represents the production capacity consumption for the preparation activities of product i. i This represents the unit capacity consumption of product i.
[0161] In addition, particle swarm optimization (PSO) is used for production cycle prediction. Specifically, based on the performance of production cycles of various process platforms under different daily load rates over the past five years as a learning standard, PSO and machine learning are used to derive the production cycle under different load rates. Furthermore, by continuously updating data with information such as the daily product mix and total quantity, the system learns and obtains valuable production cycle predictions.
[0162] For example, the particle swarm optimization algorithm involves updating ion velocity and updating particle position as follows:
[0163] Update particle velocity:
[0164] v i,j (t+1)=w*v i,j (t)+c1*r1*[pbest i,j (t)-x i,j (t)]+c2*r2*[gbest i,j (t)-x i,j (t)
[0165] Update particle position to: x i,j (t+1)=x i,j (t)+v i,j (t)
[0166] Where i represents a certain order of magnitude in the population size, j represents a certain dimension in the particle dimension, w represents the inertia factor, t represents a specific moment within the planning period, and v i,j (t+1) represents the position of the particle at time (t+1), c i The positive constant, r, represents the self-awareness learning factor. i pbest represents a random number in the interval [0,1]. i,j (t) represents the particle's optimal solution at time t, i.e., the individual optimal solution, gbest i,j (t) represents the optimal solution found so far in the entire particle swarm, i.e., the global optimal solution, x i,j (t) represents the position of the particle at time t.
[0167] It should be noted that in some embodiments, the intelligent planning module 230 sets an initial model based on the strategic plan.
[0168] In addition, the control device further includes an output control module 240, which is adapted to issue the production plan to the semiconductor production management system.
[0169] Specifically, the output control module 240 is connected to the semiconductor production management system, issues the production plan to the semiconductor production management system, and assigns corresponding delivery targets and tape-out guidance dispatch levels according to the customer's delivery period for the product to carry out production.
[0170] In some embodiments of the present invention, the control device further includes: an online module 202, which is adapted to obtain online information of the semiconductor production line in real time; the online module 202 is connected to the intelligent planning module 230 and provides the intelligent planning module 230 with online information of the semiconductor production line in real time; the intelligent planning module 230 repeatedly executes the step of obtaining the production plan within the production cycle; and the output control module 240 issues the production plan to the semiconductor production management system.
[0171] The intelligent planning module 230 is indirectly connected to the semiconductor production line through the online module 202, and indirectly obtains the online information of the semiconductor production line through the online module 202. In other embodiments, the intelligent planning module may also be directly connected to the semiconductor production line, and directly obtain the online information of the semiconductor production line from the semiconductor production line.
[0172] Specifically, a production plan is obtained based on the production cycle, and the obtained production plan is rolled out to the production execution system. Based on the customer's delivery time for the product, corresponding delivery targets and tape-out guidance dispatch levels are assigned.
[0173] It should be noted that, in some embodiments of the present invention, the control method further includes: obtaining production line information such as the capacity load status, online equipment load and operating status, and delivery schedule of the semiconductor production line.
[0174] The information integration module 210, the order admission module 220, the intelligent planning module 230, and the output control module 240 all have their information flowing sequentially according to their calculation speed. This eliminates the traditional, crude FIFO order sorting arrangement and transforms static production planning into dynamic production planning. The principle is to use dynamic orders and dynamic production status, combined with historical big data production cycles, to dynamically arrange production plans based on a joint optimization algorithm model.
[0175] In summary, based on order business and production information, combined with semiconductor production line capacity information, a strategic plan for the strategic cycle is obtained; based on the strategic plan, combined with the online information of the semiconductor production line, a production plan for the production cycle is obtained. The strategic plan is derived from order business and production information, and the production plan obtained from this strategic plan effectively integrates order business and production information, providing an executable solution for improving the efficiency of the semiconductor production management system, optimizing annual revenue, and reducing costs.
[0176] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.
Claims
1. A control method for a semiconductor production management system, characterized in that, include: Obtain business information and production information for the order; Obtain capacity information for semiconductor production lines; Based on the business information and production information of the order, combined with the capacity information of the semiconductor production line, a strategic plan for the strategic cycle is obtained; Obtain online information about semiconductor production lines; Based on the strategic plan and combined with online information from the semiconductor production line, a production plan is obtained within the production cycle, which is shorter than the strategic cycle. The production plan is issued to the semiconductor production management system.
2. The control method as described in claim 1, characterized in that, The steps to obtain business information and production information for an order include: Obtain the product code to obtain the business information of the order, wherein the product information of the order corresponds to the product code; Obtain the cell code to obtain the production information of the order, and the production information of the order corresponds to the cell code.
3. The control method as described in claim 1, characterized in that, The steps to obtain a strategic plan for the strategic cycle include: Based on the business information and production information of the orders, and in conjunction with the capacity information of the semiconductor production line, it is determined whether all orders should be approved. Based on the result of determining whether to grant access to all orders, a strategic plan for the aforementioned strategic cycle is obtained.
4. The control method as described in claim 3, characterized in that, In the step of determining whether to admit all orders and obtaining the strategic plan for the strategic period, the strategic plan for the strategic period is obtained based on all orders.
5. The control method as described in claim 4, characterized in that, Also includes: Obtain a production maintenance plan; When it is determined that all orders should not be admitted, the steps to obtain a strategic plan for the strategic cycle further include: before obtaining the strategic plan for the strategic cycle, obtaining a selection production plan based on the business information and production information of all orders, combined with the maintenance production plan; In the step of obtaining the strategic plan for the strategic cycle, the strategic plan for the strategic cycle is obtained based on the selected production plan and the maintained production plan.
6. The control method as described in claim 5, characterized in that, The steps to obtain a production plan include: Based on the business information and production information of all orders, and in conjunction with the production maintenance plan, the remaining orders are obtained; Sort the remaining orders; The selected production plan is obtained based on the sorting of the remaining orders.
7. The control method as described in claim 6, characterized in that, Also includes: Obtain the strategic objectives for the strategic cycle; based on the strategic objectives, obtain the evaluation indicators for the orders; In the step of sorting the remaining orders, the remaining orders are sorted according to the evaluation indicators of each order.
8. The control method as described in claim 7, characterized in that, In the step of obtaining the strategic objective of the strategic cycle, the strategic objective is profit maximization; In the step of obtaining the evaluation index for the order, the evaluation index is the Lissner ratio.
9. The control method as described in claim 7, characterized in that, In the step of obtaining the strategic objective of the strategic cycle, the strategic objective is revenue maximization; In the step of obtaining the evaluation index for the order, the evaluation index is the price-time-energy ratio.
10. The control method as described in claim 6, characterized in that, In the step of obtaining the selected production plan, the optimal solution is obtained by linear programming based on the sorting of the remaining orders as the selected production plan.
11. The control method as described in claim 1, characterized in that, In the step of obtaining the production plan within the production cycle, the production plan within the production cycle is obtained by combining linear programming and particle swarm optimization based on the strategic plan.
12. The control method as described in claim 11, characterized in that, In the step of obtaining the production plan within the production cycle through a combination of linear programming and particle swarm optimization, an initial model is set according to the strategic plan.
13. The control method as described in claim 1, characterized in that, In the process of obtaining online information of the semiconductor production line, the online information of the semiconductor production line is obtained in real time. The step of obtaining the production plan within the production cycle is to repeat the step of obtaining the production plan within the production cycle on a cycle basis. After repeatedly executing the step of obtaining the production plan within the production cycle, the step of issuing the production plan to the semiconductor production management system is repeated.
14. A control device for a semiconductor production management system, characterized in that, include: An information integration module, which is suitable for obtaining business information and production information of orders; An order admission module, which is suitable for obtaining capacity information of semiconductor production lines; The order admission module is also suitable for obtaining strategic plans for strategic cycles based on the business information and production information of the order, combined with the capacity information of the semiconductor production line. An intelligent planning module, which is suitable for obtaining online information of a semiconductor production line; The intelligent planning module is also suitable for obtaining a production plan within a production cycle based on the strategic plan and the online information of the semiconductor production line, wherein the production cycle is shorter than the strategic cycle. An output control module is provided, which is adapted to issue the production plan to the semiconductor production management system.
15. The control device as described in claim 14, characterized in that, The information integration module includes: The first submodule is suitable for obtaining business information about orders; The second submodule is suitable for obtaining production information for orders.
16. The control device as claimed in claim 14, characterized in that, The order access module includes: The first submodule is suitable for obtaining capacity information of the semiconductor production line; The judgment submodule is adapted to determine whether to approve all orders based on the business information and production information of the order, combined with the capacity information of the semiconductor production line. The planning submodule is adapted to obtain the strategic plan for the strategic cycle based on the judgment result of the judgment module.
17. The control device as claimed in claim 16, characterized in that, When the judgment submodule determines to admit all orders, the planning submodule obtains the strategic plan for the strategic period based on all orders.
18. The control device as claimed in claim 17, characterized in that, The order admission module further includes a second submodule, which is adapted to obtain a production maintenance plan; The selection submodule is suitable for obtaining a production plan based on the business information and production information of all orders, combined with the production maintenance plan, when the judgment module determines that all orders should not be admitted. The planning submodule obtains the strategic plan for the strategic cycle based on the selected production plan and the maintenance production plan.
19. The control device as claimed in claim 18, characterized in that, The selection submodule includes: A filter, which is adapted to obtain the remaining orders based on the business information and production information of all orders, in conjunction with the production maintenance plan; A sorter adapted to sort the remaining orders; A combiner adapted to obtain the selected production plan based on the sorting of the remaining orders.
20. The control device as claimed in claim 19, characterized in that, Also includes: An evaluation module, which is adapted to obtain the strategic objectives of the strategic cycle and, based on the strategic objectives, obtain the evaluation indicators of the order; The sorter sorts the remaining orders according to the evaluation metrics of each order.
21. The control device as claimed in claim 20, characterized in that, When the strategic objective is to maximize profits, the evaluation indicator is the profit-time-energy ratio; when the strategic objective is to maximize revenue, the evaluation indicator is the price-time-energy ratio.
22. The control device as claimed in claim 14, characterized in that, The intelligent planning module obtains the production plan within the production cycle based on the strategic plan by combining linear programming and particle swarm optimization.
23. The control device as described in claim 14, characterized in that, Also includes: An online module, which is suitable for obtaining real-time online information of the semiconductor production line.