JP3983061B2 - production management system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a production management system, method, and program capable of satisfying demand and producing with a small number of in-process in a multi-stage production system having a large production fluctuation due to the influence of yield fluctuation and apparatus operating rate reduction. Is.
[0002]
[Prior art]
Conventionally, thin film products such as semiconductors are manufactured with a large amount of in-process in the production process in order to cope with the time required for product completion and the decrease in production due to yield fluctuations and equipment failures. I came. However, from the viewpoints of shortening the product life cycle and complying with customer delivery dates, it has become necessary to optimize the in-process and produce the required amount in a timely manner. In addition, the production process has a very large number of processes, and the production process has a job shop layout, so the same product is different in process but passes through the same production equipment many times. Since the number is very large, there is a problem that it is difficult to manage the in-process amount in each process.
[0003]
Therefore, as in JP-A-6-69089 and JP-A-7-74226, the amount of in-process is controlled by controlling the amount of in-process to maintain the appropriate amount of in-process in each process or important process of the production process. Methods have been proposed to reduce lead time.
[0004]
Further, as in JP-A-6-69089 and JP-A-11-296208, the relationship between the production amount and the in-process amount is obtained from the past production results, and the in-process amount for the required production amount is calculated. Therefore, a method for setting an appropriate work-in-progress amount has been proposed.
[0005]
[Problems to be solved by the invention]
Even if the in-process amount is optimized in each process of the production process, it is a partial optimization, and the entire production process cannot be optimized. For example, when a work-in-process far exceeding the proper work-in-process amount occurs due to equipment failure, or when rapid production is performed at the customer's request and the work-in-process is far below the proper work-in-process amount, This is because when the in-process amount is optimized in this process, the in-process amount in the next process exceeds the appropriate amount, and the latter in-process amount is less than the appropriate amount. Therefore, the first problem of the present invention is to control the production amount so as to optimize the in-process amount of the entire production process. The second problem is to quantify an appropriate value of the in-process amount that can be produced while satisfying the demand.
[0006]
[Means for Solving the Problems]
In order to solve the above-mentioned problems, in the present invention, a means for calculating and setting a target value of the in-process amount for each process of the production process based on the production plan, and the in-process amount of the in-process amount for each process of the production process. Means for calculating a cumulative value from the final process to each process in reverse order for each of the target value and the actual value of the in-process amount, and each process according to the amount of deviation between the target value and the actual value of the accumulated in-process amount By means of controlling the production amount, the in-process amount of the entire production process can be suppressed to an appropriate value. In addition, based on the production time of each preceding process based on the production time of each preceding process based on the production time of the final process in the production process of the product, An appropriate value of the in-process amount that enables production to meet the demand can be calculated by means for calculating the in-process amount target value.
[0007]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described. In thin film products such as semiconductors, products are produced by forming a film on a substrate such as silicon and creating a circuit pattern on the substrate. Depending on the function of the product, this work may be repeated for several tens of layers. Become. FIG. 2 shows an outline of this production process using a semiconductor as an example. Thus, not only a plurality of processes exist, but also a production process in which each layer is formed in stages. In addition, since this production process has a job shop layout, the processes for performing the same processing for each layer are processed by devices belonging to the same equipment group. Therefore, each device has not only multiple products in progress but also different products in different stages (layers) even if they are the same product, so which devices can be produced to comply with the production plan. It is difficult for an operator to grasp.
[0008]
Therefore, in the present invention, by performing production based on the production management method having the function shown in FIG. 1, it is possible to perform production in compliance with the production plan in the production system consisting of the multi-stage production process as described above. is doing. Hereinafter, the production control method in the production management method of the present invention will be described with reference to FIG. The process definition acquisition unit 1 acquires and stores product production process and definition information of production time of each process. In addition, the production plan amount acquisition unit 2 acquires and stores the input amount to the production process and the plan value of the production amount in the process. Based on these pieces of information, an in-process amount target value calculation unit 3 calculates an in-process amount target value in each process of the production process. The in-process amount target value may be calculated from a statistical method based on the relationship between the production amount and the in-process amount based on past results, or set based on experience, as in the prior art. In this case, a function for changing the in-process amount target value calculation result from the in-process amount target value setting unit 4 may be set. Next, in the reverse cumulative work amount calculation unit 5, the final process of the production process is performed based on the work amount target value in each process obtained by the work amount target value calculation unit 3 or the setting unit 4. The cumulative amount of the in-process amount target value from the in-process amount target value to the respective processes in reverse order of the process is obtained. Hereinafter, this is the target value of the reverse cumulative work in progress. Similarly, the actual in-process amount of each process acquired in the production result acquisition unit 6 is calculated for each process, and this is calculated as the actual value of the reverse accumulated in-process amount. To do.
[0009]
An example of these calculation examples is shown in the table of FIG. In FIG. 10, it is assumed that the production process is simply represented by five steps. Each of these five processes has a daily production volume of 5, and the in-process amount target value of each process is 10 which is twice the production volume. For the sake of simplicity, it is assumed that these five steps represent only the steps processed by the same apparatus group as shown in FIG. 2, for example, only photolithography. Therefore, since the processing is performed in the same equipment group, the production volume of each process is not equal to 5, for example, the production volume of the process 2 can be 10 and the production volume of the process 3 can be 0. Alternatively, it is possible to increase the production amount to a maximum of 25 per day only for a specific process.
[0010]
First, in the conventional technology, it is possible to grasp the difference between the target value and the actual value in each process, but it is not possible to grasp the whole situation, for example, whether or not the amount of work in the entire production process is large. However, by managing the cumulative amount as in this method, in addition to the divergence between the target value and the actual value in each process, in what situation the in-process amount in the entire production process is relative to the target value. You can see if there is. In addition, since (in-process amount of the entire production process + 1) is proportional to the lead time of the product newly input into the production process, the product to be newly input due to the difference between the target value and the actual value of the reverse cumulative in-process amount It becomes possible to grasp the influence on the delivery date.
[0011]
In addition, when controlling the production volume so as to maintain the target in-process amount, the conventional technology has a difference in each process in order to bring the in-process amount of each process closer to the target value. Paying attention to the largest process 2, if the production volume of process 2 is set to 20, the remaining production capacity will be 5 due to the equipment capacity, and then the production volume of process 4 will be reduced in order to maintain the in-process amount of process 5. It is assumed that 5. However, when the processing in this case is completed, the in-process amount in step 4 is 0, and the in-process amount in step 3 is 25, so the target value cannot be satisfied. In this case, since the product is not produced from step 5, the newly introduced product increases the total in-process amount of the entire production process from the target value, leading to an increase in lead time. Then, in order to make the in-process amount in step 1 coincide with the target value, a new product is introduced. Therefore, in the case shown in FIG. 10, if control is performed in individual processes as in the conventional technique, it may be difficult to satisfy the conditions in all processes and it may be difficult to control the entire process. .
[0012]
Therefore, in the present invention, as shown in FIGS. 12 to 15, the production amount and the input amount of each process are controlled step by step based on the deviation amount between the target value and the actual value of the reverse cumulative work amount. Match with the target value. This makes it possible to produce without extending the lead time. In the present invention, these processes are performed in the production amount calculation unit 7. This will be described with reference to the production amount calculation processing flowchart of FIG. In the following description, the sign is described as a case where the target value is subtracted from the actual value. In the opposite case, the sign of the deviation amount is also reversed.
[0013]
First, in step S1, the divergence amount of the reverse cumulative work amount is obtained for each process from the difference between the target value and the actual value of the reverse cumulative work amount (FIG. 10: column of divergence amount of reverse cumulative work amount). In step S2, it is determined whether or not there is a process whose deviation amount is not 0. If there is no corresponding process, the production amount of each process is set as the production plan, and the process is terminated. Subsequently, when there is a process in which the deviation amount is not 0, the initial value of the production amount of each process is set to 0 in step S3. Next, in step S4, it is checked whether the maximum absolute value of the divergence amount exceeds the production capacity. If it exceeds, the deviation amount is corrected in step S5. Here, the correction coefficient 2 is calculated as “the maximum absolute value of the deviation amount × the correction coefficient 2 = the production capacity”, and the deviation coefficient calculated in step S1 is multiplied by the correction coefficient 2. In step S6, a check is made to see if there is a negative deviation. Hereinafter, the processing from step S7 to step S12 will be described in the case where there is a negative deviation amount. First, in step S7, the process n is extracted in ascending order of the divergence amount (step 3 in the example of FIG. 10), and the production amount of the process n-1 is calculated in step S8. For example, in the example of FIG. 10, since the divergence value in step 3 is negative (−15), the amount of production in step 2 of FIG. However, if the divergence amount is the same in step S7, the process having the previous process order is processed first. Here, the correction coefficient 1 is a coefficient to be multiplied in order to reduce the deviation amount when the variation in the production amount of each process becomes too large, and the correction coefficient 2 is set so that the absolute value of the deviation amount cannot exceed the production capacity and cannot be produced. In this case, the coefficient is multiplied to reduce the divergence amount. When both values are set to be less than 1, the time for convergence to the target value of the reverse cumulative work-in amount increases. In this embodiment, the correction coefficient 1 is 1. After this, in step S9, it is confirmed whether there is another one that has a negative divergence amount, and if there is, the process from step S7 to step S9 is repeated. Through the above processing, an intermediate value of the production amount in steps 3 and 4 in FIG. 12 is obtained.
[0014]
Subsequently, the remaining capacity of the production capacity is checked in step S10, and if the production capacity is exceeded in the processes up to step S10, the production amount correction process is performed in step S11. Here, the total amount of production in each process is reduced by reducing the production volume of each process by the same amount except for the production volume of the process calculated from the maximum absolute value of the deviation amount and the process of production volume 0. Correct within production capacity. In FIG. 12, the sum of the intermediate values of each process production amount is 30 and exceeds the production capacity of 25. Therefore, except for process 2, the production quantities of processes 3 and 4 are reduced by 3, and the respective production quantities are 7 and 2. And Thereafter, the final process production amount is calculated in step S12. Here, the remaining capacity of the production capacity generated by the processing up to step S11 is calculated, and the remaining capacity of the production capacity is evenly distributed to the final process and the processes other than the production amount 0. However, if equal distribution is not possible, priority is given to the final process. In addition, the input amount is the same as the production amount in the final process in order to keep the total in-process amount in the entire production process constant. Taking FIG. 12 as an example, the remaining amount of production capacity is 1 from 25− (15 + 7 + 2), so the final production amount in step 5 is 1 and the input amount is also 1. With the above processing, the determined production amount shown in FIG. 12 is obtained for each process. If there is no negative difference in step S6, first, in step S13, the process n is extracted in descending order of the deviation, and the production amount of the process n is calculated in step S14. However, if the divergence amount is the same in step S13, the process having the next process order is processed first. Then, in step S15, it is checked whether there is a positive deviation, and if there is, the processing from step S13 to step S15 is repeated. Thereafter, the same processing as that in steps S10 to S12 is performed.
[0015]
Through the above processing, the daily input amount and the production amount for each process as shown in FIG. 12 are calculated, and the in-process product shown in FIG. 10 is produced based on the input amount and the production amount shown in FIG. As a result of the production, the state of the in-process product changes to a quantity (actual value of the in-process amount) as shown in FIG. For example, in process 1, one in-process product is newly added to 5. However, since the production amount for the day is 0, the in-process amount is 6 at 5 + 1-0. Thereafter, the calculation process is similarly performed. Here, when the reverse cumulative in-process amount is calculated, it is calculated as shown in FIG. 13, and the deviation value is averaged while maintaining the in-process amount constant even when compared with the stage of FIG. It can be seen that
[0016]
Next, the case where the embodiment shown in FIG. 3 is further executed at the stage of FIG. 13 will be described below. 14 and 15 will be described as an example.
[0017]
As a result of the processing from step S1 to step S9, as shown in FIG. 14, the intermediate values of the production amounts of the respective processes are 1 in process 2, 4 in process 3, and 4 in process 4. Subsequently, since the remaining production capacity is 25− (1 + 4 + 4)> 0 in step S10, the process of step S12 is performed. Here, the remaining amount of production capacity is 16, and if the remaining amount of production capacity is evenly distributed to processes other than the production amount 0 including step 5 of the final process (4 in this example), each process as shown in FIG. The determined value of the production amount is 5 for the process 2, 8 for the process 3 and 4, and 4 for the process 5. The input amount shown in FIG. 14 and the production amount for each process are calculated by such arithmetic processing, and when the production control based on the calculation result is executed, the in-process state shown in FIG. 15 is obtained, and the in-process amount target is obtained. The value matches the actual value.
[0018]
By the above processing, the in-process amount is controlled as shown in FIG. 15, and the actual value of the reverse cumulative in-process amount can be matched with the target value. As described above, based on the input amount and the production amount calculated by the production amount calculation unit 7, the production instruction unit 8 outputs a production instruction to perform production. The above functions 1 to 8 are carried out in accordance with the production volume instruction cycle, and if necessary, the production volume can be corrected by executing the functions 5 to 8 within the daily production time. is there.
[0019]
In addition, as shown in FIG. 1, the target value 9 and the actual value 10 of the reverse cumulative work in progress calculated by the above-described method are displayed in graphs according to the process order and process, respectively. The current production progress status with respect to the production plan can be grasped from the in-process amount. Therefore, even if only the reverse cumulative work amount and the divergence amount are displayed in the production instruction unit 8, the production amount is mainly set in the process where the absolute value of the divergence amount is large so as to reduce the shaded portion in FIG. By adjusting the production, it is possible to comply with the production plan.
[0020]
Although the above-described production control method and display method based on the reverse cumulative work-in-process amount is an example of one product, it can be similarly managed when a plurality of products are mixed and produced. As an example, two products shown in FIG. 4 will be described. Assume that the processes using the same equipment have the same process name, and the production processes of products A and B can be expressed as shown in FIG. The combination of these is defined as the unified production process shown in FIG. First, production processes may be defined in the order of processes having the same process name, and the processes included between them may be freely defined. For example, in FIG. 4, the order of the process 2 and the process 3 is shown, but the order of the process 3 and the process 2 may be used. According to the unified production process defined in this way, the target value and actual value of the reverse cumulative work in progress are calculated for each of products A and B as described above. Then, the target value of the unified reverse cumulative work in progress is obtained by adding the target values of the reverse cumulative work in progress for products A and B, and the actual value is defined in the same manner. Then, the divergence amount is calculated from the difference between the target value and the actual value of each reverse cumulative work amount. An example of this numerical example is shown in FIG. The above processing can be similarly performed for two or more products. Regarding the calculation of the production amount, first, a comparison is made from the divergence amount of the unified reverse cumulative work-in-process amount, and then the divergence amount of the reverse cumulative work-in-process amount of individual products is compared with the case of the aforementioned one product. Similarly, the production volume can be determined. FIG. 5 shows the data of FIG. 11 as an example. The target value 11 and the actual value 12 of the reverse cumulative work amount of the product A, the target value 13 and the actual value 14 of the reverse cumulative work amount of the product B, and the unified reverse cumulative. The in-process amount target value 15 and the actual value 16 are displayed. Therefore, similarly to FIG. 1, even in the case of a plurality of products, the production amount can be adjusted manually by using FIG.
[0021]
Next, a method for calculating a target value of an in-process amount that enables production while meeting the demand while suppressing the number of in-process is described. First, as an example, the production process of a product is defined as shown in FIG. 6 from the data stored in the process definition acquisition unit 1 of FIG. 1, and the production process is determined from the production time data of each process of the production process. Assume that the cumulative production time from the final process to the process n is Xn days, the cumulative production time to the process 1 is X1 days, and the cumulative production time to the process 2 is also X2 days. Next, it is assumed that the final process production plan quantity of the production process acquired by the production plan quantity acquisition unit 2 of FIG. 1 can be expressed as shown in FIG. If Xn = N (day) and X1 = X2 = N + i (day) at this time, the final process production plan quantities YN and YN + i corresponding to the production dates N and N + i are obtained from the relationship shown in FIG. Is obtained. The final process production plan quantity corresponding to each process obtained in this way is equal to the production quantity required for each process on the day when the final process production plan quantity is acquired. Similarly, based on the accumulated production time from the final process to each process, the production amount Zn of each process is obtained as shown in FIG. Since the production quantity Zn of each process at that time is equal to the input quantity to each next process, the in-process quantity Wn of each process is calculated based on the queuing theory, the production activity time t of each day, the production time Tn of each process, n can be expressed by (Formula 1) as the number of steps.
Wn = (Zn-1 / t) x Tn (Equation 1)
However, if n = 1, Z0 = input amount.
[0022]
If the final process production plan quantity fluctuates significantly due to fluctuations in demand, the final process production plan quantity may be averaged within the schedule range of the production planning cycle.
[0023]
The production time of each process includes the waiting time in addition to the actual processing time, and the cumulative sum of the production time of each process is the number of days required for production to set the normal delivery time (product raw material is newly It is set to be equal to the number of days required to complete through the final process. Therefore, if production is performed with the in-process amount Wn as a target value, it is possible to produce a production plan based on the customer delivery date, that is, satisfy the demand without having an extra in-process.
[0024]
In addition, when a product outside the production plan is introduced, the actual value of the reverse cumulative work in progress will deviate from the target value, leading to an increase in lead time. To prevent this, the amount of planned product input will be reduced. Therefore, when an order is received for a product such as an unplanned customer development / prototype, the actual value of the reverse cumulative work amount 18 indicates how much the lead time of other products should be increased or the input amount must be reduced. By adding the request amount of the customer and expressing the estimated value 19 together with the target value 17, it can be visualized as shown in FIG. Also, the divergence between the target value and the estimated value of the reverse cumulative work in process 1 is the effect of the customer's request. Based on this divergence and the price information of the planned product, the loss profit is obtained, and the reverse cumulative work is obtained. By displaying together with the relationship between the quantity, it can be used for price negotiation with customers. Alternatively, the loss profit can be used as an underwriting price to disclose information to customers via the Web or the like for sales promotion activities.
[0025]
【The invention's effect】
As described above, according to the production management method of the present invention, in a multi-stage production system in which the production amount varies greatly due to the influence of the yield variation, the equipment operation rate decrease, etc., the number of devices in the entire production process is small. Production that meets demand is possible.
[Brief description of the drawings]
[Fig. 1] Functional block diagram [Fig. 2] Schematic diagram of semiconductor production process [Fig. 3] Production volume calculation process flowchart [Fig. 4] Example of handling of production process in multiple products [Fig. 5] Reverse accumulation in multiple products Example of quantity [FIG. 6] Example of cumulative production time by process from final process [FIG. 7] Example of final process production plan quantity [FIG. 8] Example of production quantity by process [FIG. 9] In other embodiments Example of reverse cumulative work in progress [FIG. 10] Table showing reverse cumulative work in progress for each process [FIG. 11] Table showing reverse cumulative work in progress for multiple products in each process [FIG. 12] Production by process Example of amount calculation result [FIG. 13] Example showing reverse cumulative work amount after first production instruction according to production amount [FIG. 14] Another example of production amount calculation result for each process [FIG. 15] An example showing the reverse cumulative work in progress after the second production instruction according to the production volume [Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Process definition acquisition part 2 Production plan amount acquisition part 3 In-process amount target value calculation part 4 In-process amount target value setting part 5 Reverse accumulation in-process amount calculation part 6 In-process acquisition part 7 Production amount calculation part 8 Production execution part 9 Target value of reverse cumulative work amount 10 Actual value of reverse cumulative work amount 11 Target value of reverse cumulative work amount of product A 12 Actual value of reverse cumulative work amount of product A 13 Reverse cumulative work amount of product B Target value of product 14 Actual value of reverse cumulative work in progress 15 Target value of unified reverse cumulative work in progress 16 Actual value of unified reverse cumulative work in progress 17 Target value of reverse cumulative work in progress 18 Reverse cumulative work in progress Actual value of 19 Estimated value of reverse cumulative work in progress

Claims (3)

  A production management system for producing a product in a multi-stage production process and managing the production process, a means for calculating and setting a target value of an in-process amount for each process of the production process based on a production plan, Means for calculating a cumulative value from the last process to each process in reverse order for each of the target value of the in-process amount and the actual value of the in-process amount for each process of the process, and the target value and actual value of the accumulated in-process amount And a means for controlling the production amount of each process in accordance with the amount of deviation from the value.   The production management system according to claim 1, further comprising display means for displaying the target value and actual value of the accumulated in-process amount in the order of processes and for each process, so that the production progress status of the production process can be grasped. A featured production management system. 2. The production management system according to claim 1, wherein means for calculating and setting the target value of the in-process amount is accumulated from the last process to each process from the production time data Tn of each process (n) of the production process. Based on the production plan quantity for each day of the final process in the production process determined by the production plan , each process has the cumulative production time based on the date on which the final process production plan quantity was acquired. the day before than production date only a few minutes the final step of indicating, as to produce the same production as the production plan of the final step, each day of production plan of the final step as a production Zn required for each step Calculate
The target value Wn of the in- process amount for each process is
Wn = (Zn-1 / t) × Tn
Where t is the daily production activity time and n is the number of processes
A production management system characterized by being calculated and set by

Images