JP6657562B2 - Evaluation program, evaluation device and evaluation method - Google Patents

Description

  The present invention relates to an evaluation program and the like.

  In a production line having a plurality of processes, the required processing time varies depending on each process, but the process that takes the longest time among the plurality of processes becomes a bottleneck. The bottleneck process often determines the throughput of the production line. Here, the throughput refers to the number of finished products output from the production line per unit time.

  In the production line, the objects to be processed are sequentially input from the entrance of the production line, a plurality of processes are sequentially processed by the respective devices, and the final process is performed from the exit of the production line in the order in which the products are input. Output the object to be processed. Only one processing object can be placed on one device at a time. A substrate is an example of an object to be processed.

  Here, some production lines do not have a tank serving as a buffer or a mechanism for retracting a substrate between processes for performing processing. In such a production line, for example, since the front substrate is being processed in the process C, the processing of the rear substrate may not proceed to the process C, and a waiting time may occur in the process B of the previous process.

  Although there is a tank serving as a buffer and a mechanism for retracting the substrate, there is a small amount of the tank. In such a production line, for example, there is only a small amount of a mechanism for retracting the tank or the substrate between the process C and the process B of the preceding process. B is completed with the substrate which is waiting for the start of the process C, and the process of the subsequent substrate does not proceed to the process C as in the case of the above-mentioned tank which serves as a buffer or has no mechanism for retracting the substrate. In the process B, a waiting time may occur.

  Further, the actual processing time of each step is not constant, but may be accidentally increased due to a trouble or the like and fluctuated. If the processing time of a certain process increases, the start time of the subsequent process will be delayed by that much, and in a production line without or with a small amount of the above-mentioned buffer, the progress of the previous process will be affected. May give. For example, consider a case where the buffer between the previous process B and the process C is full, and thus a waiting time occurs on the previous process B. In this case, when the processing of the substrate in the process C is completed and the process proceeds to the next process, the substrate in the buffer between the previous process B and the process C proceeds to the process C, and the substrate waiting in the process B is replaced with the substrate in the previous process B And enters the buffer between step C. At this time, the case where the processing time of the substrate in the process C did not elongate was compared with the waiting time of the substrate in the process B in the case where the elongation occurred. The waiting time becomes longer. If the waiting time of the substrate in the process B becomes longer, the start of the process in the process B of the next substrate is delayed, that is, the elongation generated in the process C affects the progress of the process in the preceding process B. As described above, the elongation caused in a certain process may affect not only the progress of the process but also the progress of processes before and after the certain process.

  Elongation affects the progress of the process, which may affect the throughput of the production line.On the other hand, depending on the process in which the elongation occurred, the magnitude of the elongation, the timing of processing the previous substrate, etc. It may not affect throughput. For example, consider a case in which the buffer between the previous process B and the process C is full and a waiting time occurs on the previous process B. In this case, even if the elongation occurs in the previous step B and the processing time becomes longer, if the elongation is shorter than the waiting time, it does not affect the progress of any step and does not affect the throughput.

  As described above, the elongation does not affect the throughput of the production line depending on the process in which the elongation occurs, the magnitude of the elongation, the timing of processing the front substrate, and the like. For this reason, it is difficult to simply evaluate the influence of the elongation of the process on the throughput by the amount of elongation of the processing time of a certain process. Therefore, in order to improve the throughput, it is important to quantitatively evaluate the effect of the increase in the processing time of the process.

  2. Description of the Related Art In a production line having a plurality of processes, a technology related to a production plan creating apparatus capable of creating an appropriate production plan even when the process time of each process frequently changes is disclosed. 1). In this technique, the production plan creation device detects the physical state of the work in process on the production line, and uses the detected physical state of the work in process to process the work in process in each process. Calculate the processing time. Then, using the calculated process time, the production plan creation device calculates the cost index from the cost caused by the extension of the lead time, the cost generated by shortening the process time, etc., and minimizes the calculated cost index. Update the process processing time. Then, the production plan creation device creates a production plan using the updated process processing time.

  In addition, a method is disclosed in which a load distribution of each process is obtained when the processing time is accidentally increased (see Patent Document 2). In this technique, the production schedule apparatus determines production schedule information including distribution items of start date and time and lead time distribution of each process for all orders based on the order information and the product information. Then, the production schedule device obtains the load distribution of each process of all orders based on the order information, the product information, and the production schedule information.

JP 2012-133633 A JP 09-16671 A

  However, in the methods of Patent Documents 1 and 2, in a production line having a plurality of steps, there is a step where accidental elongation occurs, and there is no buffer or only a small amount. There is a problem that it is difficult to quantitatively and accurately evaluate the influence on the throughput when affecting not only the process but also the progress of the process of the preceding process.

  Patent Document 1 minimizes a cost index by optimally controlling the physical state of a controllable work-in-progress, utilizing the fact that the processing time of each process is deterministically determined based on the physical state of the work-in-progress. This method does not quantitatively determine the effect on the throughput when the processing time of the process accidentally increases.

  Further, in Patent Document 2, the load distribution of each process is obtained when the growth occurs accidentally. However, the load distribution of each process is limited in a production line in which the growth of a certain process affects only the subsequent processes. It does not quantitatively determine the effect on throughput in a production line where the growth of a certain process also affects the progress of the process of the preceding process.

  SUMMARY OF THE INVENTION In one aspect, the present invention has been made to solve the above-mentioned problems, and has a plurality of steps, a step where accidental elongation occurs, and a step where there is no buffer or only a small amount. It is an object of the present invention to quantitatively and accurately evaluate the influence on the throughput in a production line in which the growth of the process may affect not only the process after the process but also the progress of the process of the preceding process.

  The evaluation program disclosed in the present application is provided in a case where a computer does not have a buffer between steps of the plurality of steps in a production line having a plurality of steps, or the capacity of the buffer is smaller than a prescribed amount even if a buffer is present. Using a model of the processing time of each process, sampling the processing time of each process for each production object when a predetermined number of production objects are produced, Using the processing time, calculate the production time that is the time required for the production line to produce, repeat the process of calculating the production time a plurality of times, using the production time calculated each time, the throughput of the production line Is executed.

  According to one aspect of the evaluation program disclosed in the present application, it is possible to quantitatively and accurately evaluate the influence on the throughput in a production line having a plurality of processes.

FIG. 1 is a diagram illustrating an example of a production process of a production line. FIG. 2A is a diagram illustrating an example of a case where the elongation of a single process does not affect the throughput. FIG. 2B is a diagram illustrating another example of the case where the growth of the single process does not affect the throughput. FIG. 2C is a diagram illustrating another example of the case where the growth of the single process does not affect the throughput. FIG. 3 is a functional block diagram illustrating the configuration of the evaluation device according to the embodiment. FIG. 4 is a diagram illustrating an example of a data structure of the configuration information according to the embodiment. FIG. 5 is a diagram illustrating sampling generation according to the embodiment. FIG. 6 is a diagram illustrating an example of a data structure of sampling information (processing time) according to the embodiment. FIG. 7 is a diagram showing the relationship between each variable of the recurrence formula in formula (1). FIG. 8 is a diagram illustrating an example of a result of the production time distribution calculation according to the embodiment. FIG. 9 is a diagram illustrating an example of a result of the influence distribution calculation according to the embodiment. FIG. 10 is a diagram illustrating an example of a result output according to the embodiment. FIG. 11 is a diagram illustrating another example of the result output according to the embodiment. FIG. 12 is a diagram illustrating a flowchart of the entire evaluation process according to the embodiment. FIG. 13 is a diagram illustrating a flowchart of a production time calculation process when the processing times of all the processes are sampled processing times. FIG. 14 is a diagram illustrating a flowchart of a production time calculation process when the processing times of all the processes are replaced with ideal processing times. FIG. 15 is a diagram showing a flowchart of a production time calculation process in a case where one process is an ideal process time and the other processes are sampled process times. FIG. 16 is a flowchart illustrating a process of calculating a production time when one process is sampled and the other process is an ideal process time. FIG. 17 is a diagram illustrating an example of a computer that executes an evaluation program.

  Hereinafter, embodiments of an evaluation program, an evaluation device, and an evaluation method disclosed in the present application will be described in detail with reference to the drawings. The present invention is not limited by the embodiments.

[Example of production process of production line]
FIG. 1 is a diagram illustrating an example of a production process of a production line. As shown in FIG. 1, the production line has M (M is a natural number) unit processes. In step 1, for example, a substrate to be produced is received, and a series of steps are processed. In step M, a finished substrate is output as a product.

  It is assumed that the production line according to the embodiment has the following four restrictions. First, there is a first restriction that the order of processing in a series of steps cannot be changed for each substrate. Further, no tank serving as a buffer or equipment for retreating the substrate is provided between single processes, or the buffer is small. In the case where each single process has a transport facility such as a conveyor, the conveyor does not operate so as to accumulate the transported objects, and the conveyor operating at a constant speed does not function as a buffer. Therefore, for example, when the previous substrate is being processed in step 3, the processing of the subsequent substrate may not proceed to step 3, and a waiting time may be generated in step 2 of the previous process. On the other hand, when the conveyor operates to temporarily accumulate the conveyed goods, for example, the conveyor operates to hold the substrate on the conveyor until the processing of the previous substrate in a later process is completed. A small conveyor acts as a small buffer. In a production line having such a conveyor, for example, when a previous substrate is being processed in step 3 and there is also a substrate on the conveyor between step 3 and step 2 of the previous step, the substrate in step 2 Even if the above process is completed, the process may not proceed, and a waiting time may occur in the process 2 of the previous process. In other words, both in the case where there is no buffer and in the case where there is a small amount of buffer, the production line has a second restriction that a substrate input later cannot overtake processing of a substrate input earlier. . Note that “a small amount of buffer” means that the capacity of the buffer is smaller than a specified amount, and the specified amount is a sufficient amount such that the stagnation of a certain process does not propagate to the previous process. Means capacity.

  In addition, there is a third restriction that a plurality of processes cannot be performed simultaneously on one substrate. Also, only one substrate can be placed on one process at a time. That is, the production line has a fourth restriction that one process cannot be performed on a plurality of substrates at the same time.

  Further, the actual processing time of each step may be increased due to trouble or the like. The elongation may or may not affect the throughput of the production line depending on the process in which the elongation occurred, the magnitude of the elongation, the timing of processing the previous substrate, and the like. The term "throughput" as used herein refers to the number of finished substrates output from a production line per unit time.

  2A to 2C are diagrams illustrating an example of a case where the growth of a single process does not affect the throughput in a production line without a buffer. It should be noted that processes 1 to 5 represent processes of steps 1 to 5, respectively. Further, it is assumed that the process 3 is a bottleneck process which is the process having the longest processing time.

  As shown in the upper part of FIG. 2A, there are first to fourth restrictions. For example, even if the processing on the substrate i-1 is completed, the processing on the previous substrate i-2 is still a bottleneck processing. Since the process 3 is performed, the process does not proceed to the process 3 and a waiting time occurs (a1). Further, in the substrate i, in the processing 1, the processing time elongates (a2). That is, in the substrate i, the elongation occurs in the processing before the processing immediately before the processing 3 which is the bottleneck processing. Then, in the substrate i, it is possible to recover the time extended in the process 1 during the bottleneck process of the previous substrate i-1. Therefore, the extended time does not affect the throughput.

  As shown in the lower part of FIG. 2A, there are first to fourth restrictions. For example, even if the substrate i-1 has completed the process 2, the previous substrate i-2 is still a bottleneck process. Since the process 3 is performed, the process does not proceed to the process 3 and a waiting time occurs (a1). In the case of the substrate i, the processing time is increased in the processing 2 (a3). That is, in the substrate i, elongation has occurred in the process immediately before the process 3 which is the bottleneck process. Then, in the substrate i, the time extended in the process 2 during the bottleneck process of the previous substrate i-1 cannot be recovered. Therefore, the extended time affects the throughput.

  That is, even if the extension time is the same, the influence on the throughput differs depending on whether the extension occurs in the process immediately before the bottleneck process (process 2) or in the process before the bottleneck process (process 1). .

  As shown in the upper part of FIG. 2B, in the substrate i-2, in the process 5, the processing time has been extended (a4). That is, in the substrate i-2, elongation occurs in the processing immediately after the processing 3 which is the bottleneck processing. Then, the substrate i-2 does not affect the subsequent bottleneck processing (processing 3) of the substrate i-1. Therefore, the extended time does not affect the throughput.

  As shown in the lower part of FIG. 2B, in the substrate i-2, in the processing 4, the processing time has been extended (a5). That is, in the substrate i-2, the elongation occurs in the processing immediately after the processing 3 which is the bottleneck processing. Then, in the substrate i-2, the processing 4 is extended even after the subsequent bottleneck processing of the substrate i-1 ends, so that a wait after the bottleneck processing (processing 3) of the substrate i-1 occurs. Therefore, the extended time affects the throughput.

  That is, even if the extension time is the same, the influence on the throughput differs depending on whether the extension occurs in the process immediately after the bottleneck process (process 4) or in the process immediately after the bottleneck process (process 5). .

  As shown in the upper part of FIG. 2C, in the substrate i, in the processing 2, the processing time has been extended (a6). In addition, as shown in the middle part of FIG. 2C, in the substrate i, the processing time has increased in the processing time (a7). However, since only one process causes the extension, the extension time does not affect the throughput.

  As shown in the lower part of FIG. 2C, in the substrate i, the processing time increases in the processing 1 and the processing 2 (a8, a9). Then, since there are a plurality of processes that cause elongation on the same substrate, the elongation time affects the throughput.

  That is, the effect on the throughput differs between the case where the number of the processes that causes the extension of one substrate is one and the case where the number of the processes is plural.

  As shown in FIG. 2A to FIG. 2C, the influence on the throughput cannot be accurately evaluated only by evaluating the elongation of the processing time in a single process. Therefore, in the embodiment, the evaluation device 1 described later quantitatively evaluates the influence on the throughput of the production line before actually operating the production line while considering the restrictions between processes. That is, the evaluation device 1 inputs the probability distribution model of the processing time of each process under a situation where the processing time of each process is ideally performed, and there is a case where the processing time may increase from the time (ideal time). Then, using the input probability distribution model, the evaluation device 1 derives an index for quantitatively evaluating how much an increase in the processing time of each process deteriorates the throughput of the production line. In the embodiment, the elongation is a positive value.

[Configuration of evaluation device]
FIG. 3 is a functional block diagram illustrating the configuration of the evaluation device according to the embodiment. As shown in FIG. 3, the evaluation device 1 has a control unit 10 and a storage unit 20. The evaluation device 1 samples the processing time of each process for each substrate when a plurality of substrates are produced, using a probability distribution model of the processing time of a plurality of processes forming a production line. The evaluation device 1 calculates the production time, which is the time required for the production line to perform production, using the processing time of each process for each sampled substrate. The evaluation device 1 repeats the process of calculating the production time a plurality of times while changing the sampling process time. Then, the evaluation device 1 evaluates the throughput of the production line using the production time calculated each time. The production time when a plurality of substrates are produced means that when the number of substrates is N, the first substrate is output from the production line and then the N-th substrate is output from the production line. Time.

  The control unit 10 has an internal memory for storing programs and control data defining various processing procedures, and executes various processing by using these. The control unit 10 corresponds to, for example, an electronic circuit of an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). Alternatively, the control unit 10 corresponds to an electronic circuit such as a CPU (Central Processing Unit) and an MPU (Micro Processing Unit). The control unit 10 includes an input data interpretation unit 11, a sampling generation unit 12, a processing time adjustment unit 13, a production line simulation unit 14, a production time distribution calculation unit 15, an influence distribution calculation unit 16, and a result output unit 17.

  The storage unit 20 corresponds to, for example, a storage device such as a nonvolatile semiconductor memory element such as a flash memory or a FRAM (registered trademark) (Ferroelectric Random Access Memory). The storage unit 20 may correspond to a storage device using a volatile semiconductor memory element such as a random access memory (RAM). In this case, the storage unit 20 may appropriately load data from a storage device such as a nonvolatile semiconductor memory device. The storage unit 20 stores configuration information 21, sampling information (processing time) 22, production time distribution information 23, and influence degree distribution information 24.

  The configuration information 21 is information indicating the configuration of each process forming the production line. In the configuration information 21, the order of each process and the ideal processing time are set. If there is a small amount of buffer between processes, the amount of buffer between processes is also set in the configuration information 21. The configuration information 21 is generated by, for example, the input data interpretation unit 11. The sampling information (processing time) 22 is information indicating the processing time sampled for each of a plurality of substrates and each process produced by the production line. In the sampling information (processing time) 22, a processing time to be sampled for a plurality of substrates using a probability distribution model of the processing time of each process is set. The sampling information (processing time) 22 is generated by the sampling generation unit 12. The production time distribution information 23 is distribution information of production time. Note that the production time distribution information 23 is generated by the production time distribution calculator 15. The influence distribution information 24 is distribution information of the influence on the throughput of each process. The influence distribution information 24 is generated by the influence distribution calculator 16. An example of the data structure of the configuration information 21, the sampling information (processing time) 22, and the influence distribution information 24 will be described later.

  The input data interpreter 11 interprets input data. Here, the input data refers to production line configuration information and a probability distribution model of the processing time of each process.

  The configuration information of the production line also includes, for each process forming the production line, the order, the process name and the ideal processing time, and if there is a small amount of buffer between processes, the amount of buffer between processes is also included. The input data interpretation unit 11 stores the configuration information of the production line as the input data in the configuration information 21. Here, the data structure of the configuration information 21 will be described with reference to FIG.

  FIG. 4 is a diagram illustrating an example of a data structure of the configuration information according to the embodiment. As shown in FIG. 4, in the configuration information 21, a process order 21a, a process name 21b, an ideal processing time 21c, and an inter-process buffer amount 21d are set in association with each other. The process order 21a is the order of each process in the production line. The process name 21b is the name of each process. The ideal processing time 21c is an ideal processing time of each process. The unit of time set as the ideal processing time 21c is, for example, seconds. The inter-process buffer amount 21d is the amount of buffer between successive processes. The amount set in the inter-process buffer amount 21d is, for example, the maximum number of substrates that can be stored in the buffer in the case of a substrate. Although the configuration information 21 has been described as being stored by the input data interpreting unit 11, the present invention is not limited to this. For example, the configuration information 21 may be set in advance by a production line manager.

  As an example, when the process order 21a is "1", the configuration information 21 sets "load 1" as the process name 21b and "2" as the ideal processing time 21c.

  Returning to FIG. 3, for example, the following is a probability distribution model of the processing time of each process as input data. As an example, there is a frequency distribution of observation values of processing time when the same type of substrate has been produced in the past. Another example is a model obtained by modeling an observation value of a processing time when a substrate of the same type was produced in the past with an existing probability distribution. Existing probability distributions include an exponential distribution, a Poisson distribution, and a normal distribution. Another example is a probability distribution predicted from a model representing the relationship between the probability distribution of the processing time and the feature amount of the substrate. Here, the feature amount of the substrate refers to the number of components used for the substrate, the component size, the size of the substrate, and the like.

  The sampling generation unit 12 receives the probability distribution model of the processing time of each process as input, and samples the processing time for a plurality of substrates for each process. For example, when the probability distribution model is a frequency distribution of observation values of processing time, the sampling generation unit 12 generates random numbers according to the frequency distribution, and samples the generated random numbers as processing time. When the probability distribution model is an existing probability distribution, the sampling generation unit 12 generates a random number according to the modeled probability distribution, and samples the generated random number as a processing time. If the probability distribution model is a probability distribution predicted from a model representing the relationship between the probability distribution of the processing time and the feature amount of the substrate, the sampling generation unit 12 generates a random number according to the predicted probability distribution, and generates the generated random number. Is sampled as the processing time.

  In addition, the sampling generation unit 12 samples a plurality of sets of processing time, with sampling of a plurality of substrates in each step as one set. Hereinafter, the unit of one set is referred to as one lot. Here, the sampling generation by the sampling generation unit 12 will be described with reference to FIG.

  FIG. 5 is a diagram illustrating sampling generation according to the embodiment. FIG. 5 illustrates a case where the frequency distribution of the observed values of the processing time is used as a probability distribution model. It is assumed that the number of steps is M (M is a natural number greater than 2). It is assumed that the number of substrates is N (N is a natural number greater than 2). Assume that the number of lots for which sampling is generated is 1,000.

  As shown in FIG. 5, the sampling generation unit 12 inputs a probability distribution model for each process for each lot, and samples the processing time for the substrates 1 to N.

  For example, the sampling generation unit 12 inputs the probability distribution model of the process 1 for the first lot and samples the processing time for the substrates 1 to N. The processing time for the substrate 1 is sampled as “2”. The processing time for the substrate 2 is sampled as “3”. The processing time for the substrate N is sampled as “4.5”.

  The sampling generation unit 12 inputs the probability distribution model of the process 2 for the first lot and samples the processing time for the substrates 1 to N. The processing time for the substrate 1 is sampled as “30”. The processing time for the substrate 2 is sampled as “32”. The processing time for the substrate N is sampled as “30”.

  For the first lot, the sampling generation unit 12 inputs the probability distribution model of the process M as input, and samples the processing time for the substrates 1 to N. The processing time for the substrate 1 is sampled as "1". The processing time for the substrate 2 is sampled as “1.5”. The processing time for the substrate N is sampled as “2”.

  Similarly, for the 1000th lot, the sampling generation unit 12 inputs the probability distribution model of the process 1 as input, and samples the processing time for the substrates 1 to N. The processing time for the substrate 1 is sampled as “3”. The processing time for the substrate 2 is sampled as “2”. The processing time for the substrate N is sampled as “2”.

  The sampling generation unit 12 inputs the probability distribution model of the process 2 for the first lot and samples the processing time for the substrates 1 to N. The processing time for the substrate 1 is sampled as “30”. The processing time for the substrate 2 is sampled as “30”. The processing time for the substrate N is sampled as “34”.

  For the first lot, the sampling generation unit 12 inputs the probability distribution model of the process M as input, and samples the processing time for the substrates 1 to N. The processing time for the substrate 1 is sampled as “1”. The processing time for the substrate 2 is sampled as “1”. The processing time for the substrate N is sampled as “1”.

  Returning to FIG. 3, the sampling generation unit 12 stores the processing time for sampling a plurality of lots in the sampling information (processing time) 22. Here, the data structure of the sampling information (processing time) 22 will be described with reference to FIG.

FIG. 6 is a diagram illustrating an example of a data structure of sampling information (processing time) according to the embodiment. As shown in FIG. 6, the sampling information (processing time) 22 includes a lot ID (identification) 22a, a substrate ID 22b, a process 1 processing time 22c ₁ , a process 2 processing time 22c ₂ ,..., A process M processing time 22c. _M is set in association with each other. The lot ID 22a is an identifier for identifying a lot, and is represented by, for example, a continuous number. The board ID 22b is an identifier for identifying the board, and is represented by, for example, a continuous number. Step 1 processing time 22c ₁ ~ Step M processing time 22c _M is the processing time sampled from the probability distribution model of each process. Step 1 processing time 22c ₁ ~ Step M processing unit of time period is set to 22c _M is, for example, in seconds.

As an example, the sampling information (process time) 22 is "30 when the lot ID22a is" 1 ", the substrate ID22b is" 1 "," 2 "as a step 1 process time 22c _1, as a step 2 process time 22c ₂ "is set to" 1 "as the process M processing time 22c _M. Lot ID22a is "1000", if the substrate ID22b is "N", "2" as a step 1 process time 22c _1, "34" as the step of processing time 22c _2, "1" as the process M processing time 22c _M Is set.

  The processing time adjustment unit 13 adjusts various processing times of each process for each substrate.

  For example, the processing time adjustment unit 13 adjusts the processing time for each of the plurality of substrates so that the processing time is the ideal processing time. The processing time adjustment unit 13 adjusts the processing time of a plurality of lots. As an example, the processing time adjustment unit 13 uses the configuration information 21 to set the ideal processing time 21c for each process as the processing time.

  Further, the processing time adjustment unit 13 adjusts the processing time for a plurality of substrates so that one process is an ideal processing time and the other processes are sampled processing times. The processing time adjustment unit 13 adjusts the processing time of a plurality of lots.

  Further, the processing time adjustment unit 13 adjusts the processing time so that one processing is sampled for a plurality of substrates and the other processing is an ideal processing time. The processing time adjustment unit 13 adjusts the processing time of a plurality of lots.

  The production line simulation unit 14 simulates a time-series position at which each process of each substrate is processed at each processing time using various processing times of each process for each substrate. The simulation is performed so that the time-series position of processing each process of each substrate satisfies constraints between processes. Then, the production line simulation unit 14 calculates the time (production time) required for a plurality of substrates to come out of the production line. The production line simulation unit 14 performs a simulation for each of a plurality of lots.

  For example, the production line simulation unit 14 simulates a time-series position at which each process of a plurality of substrates is processed using the sampled processing time (calculation processing R). Further, the production line simulation unit 14 simulates a time-series position at which each process of a plurality of substrates is processed using the ideal processing time (calculation process 0). Further, the production line simulation unit 14 simulates a time-series position in which each process of a plurality of substrates is processed using a process time in which one process is an ideal process time and the other processes are sampled process times. (Calculation process A). Further, the production line simulation unit 14 simulates a time-series position in which each process of a plurality of substrates is processed, by using a processing time obtained by sampling one process and an ideal processing time for other processes. (Calculation process B).

Here, the simulation using the sampled processing time is performed as follows. Note that the steps are described as 1 to M, and the substrate is described as 1 to N. For example, the production line simulation unit 14 sets the sampled processing time of the process i on the substrate j to τ ′ _{i, j} (i = 1 to M, j = 1 to N). If there is no buffer between the processes, the production line simulation unit 14 sets the time S _{i, j} until the start of the process i of the substrate j and the time S _i, j of the The time E _{i, j} until the process i exits is calculated in order from the substrate 1 by the following recurrence formula, that is, formula (1).
S _{i, j} = max (E _{i-1, j} , E _{i, j-1} ), E _{i, j} = max (S _{i, j} + τ ′ _{i, j} , E _{i + 1, j-1} )・ Equation (1)
When the buffer between the steps is small, the buffer amount between the steps i and i + 1 is set as B _i , and is calculated by the following recurrence formula, that is, formula (1 ′).
S _{i, j} = max (E _{i-1, j} , E _{i, j-1} ), E _{i, j} = max (S _{i, j} + τ ′ _{i, j} , E _{i + 1, j-1-Bi} ) ... Equation (1 ')

  Here, the relationship between the variants of the recurrence formula in the formula (1) will be described with reference to FIG. FIG. 7 is a diagram showing the relationship between each variable of the recurrence formula in formula (1). It should be noted that processes 1 to 5 represent processes of steps 1 to 5, respectively. The processing time delimited by each processing is an ideal processing time. “Extend” indicates that the processing has been extended, and “Wait” indicates that the processing has been completed but cannot proceed to the next step. In FIG. 7, it is assumed that M representing the last step is 5. It is assumed that the processing time divided by each processing is an ideal processing time.

As shown in FIG. 7, for example, until the start S _{2, j} of the processing of Step 2 in the substrate j exits time E _{1, j} until leaving step 1 in the substrate j, the step 2 in the substrate j-1 Is the maximum of the time E2 _{, j-1} . Then, the time E _{2, j} up to exiting step 2 on the substrate j is the time obtained by adding the sampled processing time τ ′ _{2, j} to the start S _{2, j} of the processing of step 2 on the substrate j, and the substrate j− The maximum time is the time E3 _{, j-1} until the process 3 in step 1 is completed. Here, the processing time .tau. _{'2, j} sampled in step 2 in the substrate j, since extended than ideal processing time, E _{2, j} is, S ₂ from _{E 3, j-1, j} + τ' 2, j , S _{2, j} + τ ' _{2, j} is applied.

Then, the production line simulation unit 14 calculates the production time using the output times at which the first substrate and the Nth substrate are output from the production line, respectively. As an example, assuming that the time at which the first substrate is output from the Mth process is EM _{, 1} and the time at which the Nth substrate is output from the Mth process is EM _{, N} , The line simulation unit 14 calculates the production time T for one lot by the following equation (2).
T = E _{M, N} −E _{M, 1} ...

Returning to FIG. 3, the production time (production time by the calculation process 0) when the processing time of all the processes is set to the ideal processing time for all the substrates is expressed by the following equation (1) or (1 ′). It is calculated as follows. That is, the production line simulation unit 14 replaces the sampled processing time τ ′ _{i, j} (i = 1 to M, j = 1 to N) of the equation (1) or (1 ′) with the ideal processing time. Then, the production line simulation unit 14 calculates the production time T by using equation (2). Note that the production time in the calculation process 0 is the same time for any lot, so the production line simulation unit 14 only needs to calculate once.

For all the substrates, the production time (production time by calculation processing A) when the processing of one process is an ideal processing time and the processing of the other processes is a sampling processing time is expressed by the equation (1) or the equation (1 ′). ) Is calculated as follows. That is, the production line simulation unit 14 calculates the sampled processing time τ ′ _{i, j} (i = 1 to M, j = 1 to N) of the equation (1) or the equation (1 ′) for one process. Replace with time. Then, the production line simulation unit 14 samples the processing time τ ′ _{i, j} (i = 1 to M, j = 1 to N) of the equation (1) or the equation (1 ′) for steps other than one step. Is used as it is. Then, the production line simulation unit 14 calculates the production time T by using equation (2).

For all the substrates, the processing time obtained by sampling the processing of one step and the production time (production time obtained by the calculation processing B) when the processing of the other steps are set to the ideal processing time are expressed by Equation (1) or Equation (1 ′). ) Is calculated as follows. That is, the production line simulation unit 14 uses the sampled processing time τ ′ _{i, j} (i = 1 to M, j = 1 to N) of Expression (1) or Expression (1 ′) for one process. . Then, the production line simulation unit 14 calculates the sampled processing time τ ′ _{i, j} (i = 1 to M, j = 1 to N) of the equation (1) or the equation (1 ′) for one process. Replace with time. Then, the production line simulation unit 14 calculates the production time T by using equation (2).

  The production time distribution calculator 15 calculates the distribution of production time. For example, the production time distribution calculation unit 15 calculates the frequency with respect to the production time using the sampled production times for a plurality of lots calculated by the production line simulation unit 14. The production time distribution calculator 15 may calculate not only the frequency for the production time but also the cumulative probability for the production time. Then, the production time distribution calculation unit 15 stores the frequency and the cumulative probability with respect to the calculated production time in the production time distribution information 23. Here, the data structure of the production time distribution information 23 will be described with reference to FIG.

  FIG. 8 is a diagram illustrating an example of a result of the production time distribution calculation according to the embodiment. FIG. 8 shows an example of the result of the production time distribution calculation in the production time distribution information 23. As shown in FIG. 8, in the production time distribution information 23, a production time 23a, a frequency 23b, and a cumulative probability 23c are set in association with each other.

  As an example, when the production time 23a is "15", "0.001" is stored as the frequency 23b and "0.001" is stored as the cumulative probability 23c. When the production time 23a is "16", "0.014" is stored as the frequency 23b and "0.015" is stored as the cumulative probability 23c. When the production time 23a is "25", "0.001" is stored as the frequency 23b and "0.999" is stored as the cumulative probability 23c.

Returning to FIG. 3, the production time distribution calculation unit 15 calculates an estimated value of the throughput. For example, the production time distribution calculation unit 15 calculates an average value of the production time using the sampled production times for a plurality of lots calculated by the production line simulation unit 14. Then, the production time distribution calculation unit 15 calculates an estimated value of the throughput by using the calculated average value of the production time and the number of substrates. The production time distribution calculation unit 15 calculates the estimated value V of the throughput, for example, by the following equation (3). Incidentally, N is the number of substrate, T _R is the production time, _T'R is assumed to be the average value of the production time.
V = N / _T'R ··· formula (3)

  Thus, the production time distribution calculation unit 15 can accurately estimate the production plan based on the distribution of the calculated production time and the estimated value of the calculated throughput.

  The influence distribution calculation unit 16 calculates the influence on the throughput in each process. The impact on the throughput includes an upper limit of the impact and a lower limit of the impact.

For example, various production times for the u-th lot are as follows. The production time in the case where the processing time of all the steps is the processing time sampled is defined as T _R (u). _Let T _Ai (u) be the production time when the ideal processing time is set for the process i and the processing time is sampled for the other processes. The processing time when only the process i is sampled and the production time when the other processes are set as the ideal processing time are _defined as T _Bi (u). For one lot, the production time in the case of the ideal processing time the entire process and T _0.

Using various production time, the influence distribution calculating unit 16, the u-th lot, the step i (i = 1 to M) of the degree of influence of the upper limit _H i (u), the following equation (4) calculate.

Using various production time, the influence distribution calculating unit 16, the u-th lot, the step i (i = 1 to M) of the impact of the lower limit _L i (u), the following equation (5) calculate.

  Then, the influence distribution calculation unit 16 sets the result of dividing the frequency distribution of Hi (u) in each lot by the number of lots as the upper limit distribution of the influence on the throughput. The influence distribution calculation unit 16 sets the result of dividing the frequency distribution of Li (u) in each lot by the number of lots as the distribution of the lower limit of the influence on the throughput. The influence distribution calculation unit 16 stores the frequency of the upper limit influence and the cumulative probability for each process i in the influence distribution information 24. Here, the data structure of the influence distribution information 24 will be described with reference to FIG.

  FIG. 9 is a diagram illustrating an example of a result of the influence distribution calculation according to the embodiment. FIG. 9 shows an example of the result of the influence distribution calculation in the influence distribution information 24. As shown in FIG. 9, in the influence distribution information 24, the influence 24a, the frequency 24b, and the cumulative probability 24c are set in association with the upper limit for each process. The influence level 24a, the frequency 24b, and the cumulative probability 24c are set in association with the lower limit of each process.

  As an example, for the upper limit of step 1, when the degree of influence 24a is "0.1", "0.01" is stored as the frequency 24b and "0.01" is stored as the cumulative probability 24c. When the degree of influence 24a is "0.2", "0.14" is stored as the frequency 24b and "0.15" is stored as the cumulative probability 24c. When the degree of influence 24a is "1.5", "0.01" is stored as the frequency 24b and "0.99" is stored as the cumulative probability 24c. Regarding the lower limit of step 1, when the degree of influence 24a is "0.1", "0.01" is stored as the frequency 24b and "0.01" is stored as the cumulative probability 24c. When the degree of influence 24a is "0.2", "0.14" is stored as the frequency 24b and "0.15" is stored as the cumulative probability 24c. When the degree of influence 24a is "1.5", "0.01" is stored as the frequency 24b and "0.99" is stored as the cumulative probability 24c.

Returning to FIG. 3, the upper limit H _i, the lower limit L _i of impact of step i in each lot, the formula (4) for the following reasons, is represented as the formula (5). First, the upper limit _{H i,} the denominator of the lower limit _{L i} are both a _{T 0.} This T ₀ is a production time when all the processes are set to the ideal processing time, and is always a value of 0 or more.

The value of (T _R -T _Ai ) of the molecule is a value indicating how short the production time becomes if the process i does not elongate. Elongation step i other steps because accept the sampling values, it can be said that one of the values to measure (T R _{-T Ai)} is the minute value production time is extended due to the influence of elongation of step i. This value is always 0 or more.

(T _Bi −T ₀ ) of the numerator is a value indicating how long the production time becomes long if only the process i is a sampling value. Steps other than the process i use the ideal processing time and use the sampling value only for the process i. Therefore, (T _Bi −T ₀ ) is one value that measures the value of the increase in the production time due to the extension of the process i. I can say. This value is always 0 or more.

_Then, the magnitude of (T R _{-T Ai)} and _(T Bi -T _0), because it is not the same at all times, the influence distribution calculating unit 16 calculates the case of the upper limit max, is calculated in the case of the lower limit min . Then, the influence distribution calculating unit 16, with respect to the value of the production time T ₀ in the case of an ideal processing time the entire process to define the amount of value production time is extended due to the influence of elongation steps i and impact. Note that the value of the increase in the production time due to the influence of the elongation in the process i of the molecular portion may be defined as the influence degree. Thereby, the influence distribution calculation unit 16 can calculate the influence in a short time and at a high speed.

  The result output unit 17 outputs the distribution of the production time calculated by the production time distribution calculation unit 15. For example, the result output unit 17 outputs the frequency distribution of the production time using the production time distribution information 23.

  The result output unit 17 outputs the degree of influence on the throughput of each process calculated by the degree of influence distribution calculation unit 16. For example, the result output unit 17 uses the influence distribution information 24 to output a frequency distribution of an upper limit and a lower limit of the influence on the throughput in each process.

FIG. 10 is a diagram illustrating an example of a result output according to the embodiment. FIG. 10 shows a frequency distribution of production time as a result output. As shown in FIG. 10, the horizontal axis indicates the production time, and the vertical axis indicates the frequency. The result output unit 17 outputs the frequency distribution of the production time in a line graph. Also, the result output unit 17 outputs the processing time of the entire process the combined production time T ₀ in the case of the ideal processing time. In this way, the result output unit 17 outputs the frequency distribution of the production time, so that the probability distribution for each process at the time of sampling can be reviewed and the production plan can be estimated accurately.

  FIG. 11 is a diagram illustrating another example of the result output according to the embodiment. FIG. 11 shows the upper limit and lower limit frequency distributions of the degree of influence on the throughput in each process as the result output. As shown in FIG. 11, in each step, the horizontal axis indicates the degree of influence on the throughput, and the vertical axis indicates the frequency. In each step, the upper frequency distribution is shown as an upper frequency distribution, and the lower frequency distribution is shown as a lower frequency distribution. Here, the influence of the step 3 is larger in both the upper limit and the lower limit than in the other steps. As described above, the result output unit 17 outputs the frequency distribution of the upper limit and the lower limit of the degree of influence on the throughput in each process, thereby changing the probability distribution of each process when the throughput is sampled, and deteriorating the deterioration. It is possible to specify a process that is a main factor. As a result, the result output unit 17 can change the probability distribution for each process at the time of sampling, and can actually review the process that is the main factor of deterioration.

[Overall flowchart of evaluation processing]
FIG. 12 is a diagram illustrating a flowchart of the entire evaluation process according to the embodiment.

  As shown in FIG. 12, the input data interpretation unit 11 receives input data (Step S11). For example, the input data interpretation unit 11 receives, as input data, production line configuration information and a probability distribution model of the processing time of each process.

  Then, the sampling generation unit 12 samples the processing times of all the processes for all the substrates from the probability distribution model received as the input data (Step S12). Then, the sampling generation unit 12 determines whether or not Z (Z is a natural number greater than 1) lots has been sampled (step S13). If it is determined that sampling has not been performed for the Z lots (Step S13; No), the sampling generation unit 12 proceeds to Step S12 to perform sampling for the Z lots.

On the other hand, when it is determined that sampling has been performed for the Z lots (Step S13; Yes), the production line simulation unit 14 uses the processing time sampled for all the substrates in units of the lot u (u = 1 to Z). A simulation process of the production time is executed (Step S14). That is, the production line simulation unit 14 executes the calculation process R for each lot. The processing flow of the calculation processing R will be described later. It is assumed that the production time of the lot u calculated by the calculation process R is represented by T _R (u).

Then, the production line simulation unit 14 executes a simulation process of the production time when the processing time of all the processes is replaced with the ideal processing time for all the substrates (step S15). That is, the production line simulation unit 14 executes the calculation process 0. The processing flow of the calculation processing 0 will be described later. Has been production time calculated by the calculation process 0, and is represented by T _0.

Then, the production line simulation unit 14 calculates, for each of the lots u, the production time when one process i (i = 1 to M) is an ideal processing time and the other processes are processing times sampled for all the substrates. A simulation process is performed (Step S16). That is, the production line simulation unit 14 executes the calculation process A for each lot. The processing flow of the calculation processing A will be described later. It is assumed that the production time of the lot u calculated by the calculation process A is represented by T _Ai (u).

Then, the production line simulation unit 14 performs, for each lot u, a simulation process of the production time when one process i is sampled and the other processes are set to the ideal process time for all the substrates (step). S17). That is, the production line simulation unit 14 executes the calculation process B for each lot. The processing flow of the calculation processing B will be described later. It is assumed that the production time of the lot u calculated by the calculation process B is represented by T _Bi (u).

  Subsequently, the production time distribution calculation unit 15 calculates the distribution of the production time using the production times sampled for a plurality of lots (step S18). In addition, the production time distribution calculator 15 calculates an estimated value of the throughput (step S19). For example, the production time distribution calculation unit 15 calculates an average value of the production time using the production times sampled for a plurality of lots. Then, the production time distribution calculation unit 15 calculates an estimated value of the throughput by using the calculated average value of the production time and the number of substrates.

Then, the influence distribution calculation unit 16, by the lot u _{units, T 0, T R (u} ), T Ai (u), by using a _T Bi (u), to the throughput of each process i influence the The upper and lower limits (Hi (u), Li (u)) are calculated (step S20). Hi (u) is the upper limit of the degree of influence of the process i of the lot u, and is calculated by Expression (4). Li (u) is the lower limit of the degree of influence of the process i of the lot u, and is calculated by equation (5).

  Then, the result output unit 17 outputs the distribution of the production time calculated by the production time distribution calculation unit 15 (Step S21). The result output unit 17 outputs the influence on the throughput of each process calculated by the influence distribution calculation unit 16 (Step S22).

[Flowchart of calculation processing R]
FIG. 13 is a diagram illustrating a flowchart of a calculation process of the production time when the processing time of all the processes is set as the sampled processing time in the production line having no buffer between the processes. In FIG. 13, the variable j indicates the value of the board ID, and is assumed to be 1 to N (N: natural number). The variable i indicates a process order, and is assumed to be 1 to M (M: natural number). Hereinafter, for convenience of explanation, the variable j is described as a substrate j, and the variable i is described as a process i. The variable τ _{i, j} is the sampled processing time in the process i and the substrate j.

First, the production line simulation unit 14 sets 1 to the substrate j (step S31). The production line simulation unit 14 sets 1 to the process i (step S32). The production line simulation unit 14, the processing time .tau. _'I step i in the substrate _j, the _j, set the processing time tau _{i, j} sampled in step i in the substrate j (step S33). The processing time τ _{i, j} sampled in the process i on the substrate j is obtained from the sampling information (processing time) 22.

  Then, the production line simulation unit 14 adds 1 to the process i in order to perform the next process (step S34). The production line simulation unit 14 determines whether or not the process i is larger than the maximum number M in the process order (Step S35). If the process i is not larger than the maximum number M in the process order (Step S35; No), the production line simulation unit 14 proceeds to Step S33 to set the processing time of the next process.

  On the other hand, when the process i is larger than the maximum number M in the process order (Step S35; Yes), the production line simulation unit 14 adds 1 to the substrate j to make it the next substrate (Step S36). The production line simulation unit 14 determines whether the substrate j is larger than the maximum number N of the substrates (Step S37). If the substrate j is not larger than the maximum number N of the substrates (Step S37; No), the production line simulation unit 14 proceeds to Step S32 in order to set the processing time of each process in the next substrate.

On the other hand, when the substrate j is larger than the maximum number N of the substrates (Step S37; Yes), the production line simulation unit 14 determines the start time S _i, j of the process i on the substrate j and the output time of the process i on the substrate j. Calculate E _{i, j} . The production line simulation unit 14 sets the start time S _1,1 of the process 1 on the substrate ₁ to 0. The production line simulation unit 14 sets a value obtained by adding the start time S _1,1 and the processing time τ ′ _1,1 to the output time E _1,1 of the process 1 on the substrate 1 (step S38). Here, the processing time τ ′ _1,1 is the processing time of the substrate 1 sampled in the step 1.

Then, the production line simulation unit 14 sets 2 to the process i (step S39). The production line simulation unit 14 sets the output time E _i−1,1 of the _immediately preceding process on the same substrate 1 as the start time S _i, 1 of the process i on the substrate 1. The production line simulation unit 14 sets a value obtained by adding the processing time τ ′ _{i, 1} to the start time S _{i, 1} to the output time E _i, 1 of the process i on the substrate 1 (step S40).

  Then, the production line simulation unit 14 adds 1 to the process i in order to perform the next process (step S41). The production line simulation unit 14 determines whether or not the process i is larger than the maximum number M in the process order (Step S42). If the process i is not larger than the maximum number M in the process order (Step S42; No), the production line simulation unit 14 proceeds to Step S40 to calculate the start time and output time of the next process on the substrate 1. I do.

On the other hand, when the process i is larger than the maximum number M in the process order (Step S42; Yes), the production line simulation unit 14 sets 2 to the substrate j (Step S43). The production line simulation unit 14 sets the output time E1 _{, j-1} of the same process 1 on the immediately preceding substrate j-1 as the start time S1 _{, j} of the process 1 on the substrate j. The production line simulation unit 14 sets a value obtained by adding the start time S _{1, j} and the processing time τ ′ _{1, j} to the output time E _{1, j−1} of the process 1 on the substrate j (step S44).

Then, the production line simulation unit 14 sets 2 to the process i (step S45). Then, the production line simulation unit 14 sets the output time E _{i−1, j of the} previous process i−1 on the same substrate j as the output time E _{i−1, j} of the previous substrate on the start time S _i, j of the process i on the substrate j. , The larger of the output times _{Ei, j-1} of the same process i is set. The production line simulation unit 14 calculates the value obtained by adding the processing time τ ′ _{i, j} to the start time S _i, j of the process i on the substrate j, and the output time E _{i + 1, j−} of the next process on the previous substrate. to set the larger _one (step S46).

  Then, the production line simulation unit 14 adds 1 to the process i in order to perform the next process (step S47). The production line simulation unit 14 determines whether or not the process i matches the maximum number M in the process order (Step S48). If the process i does not match the maximum number M in the process order (step S48; No), the production line simulation unit 14 proceeds to step S46 to calculate the start time and output time of the next process on the substrate j. I do.

On the other hand, when the process i matches the maximum number M in the process order (Step S48; Yes), the production line simulation unit 14 sets the start time S _{i, j} of the process i on the substrate j to 1 in the same substrate j. one previous step i-1 of the output time E _{i-1, j} and output time E _i of the same step i in the previous _board, sets a larger _j-1. The production line simulation unit 14 sets a value obtained by adding the processing time τ ′ _{i, j} to the start time S _i, j of the process i on the substrate j (step S49).

  Then, the production line simulation unit 14 adds 1 to the substrate j to make the next substrate (step S50). The production line simulation unit 14 determines whether or not the substrate j is larger than the maximum number N of the substrates (Step S51). If the substrate j is not larger than the maximum number N of substrates (step S51; No), the production line simulation unit 14 proceeds to step S44 to calculate the start time and output time of each process on the next substrate. .

On the other hand, when the substrate j is larger than the maximum number N of the substrates (Step S51; Yes), the production line simulation unit 14 calculates the output time of the process M on the substrate 1 from the output time EM _{, N} of the process M on the substrate _N. EM _{, 1} is subtracted. Then, the production line simulation unit 14 calculates the subtracted value as the production time T _R (u) (Step S52). That is, the production line simulation unit 14 calculates the production time for the lot u when the processing time of all the processes is sampled processing time. Then, the production line simulation unit 14 ends the calculation process R for the lot u. Steps S38 to S51 of the calculation processing R are referred to as “S _{i, j} , E _{i, j} calculation processing”.

[Flowchart of calculation process 0]
FIG. 14 is a diagram showing a flowchart of a calculation process of the production time when the processing time of all the processes is replaced with the ideal processing time in the production line having no buffer between the processes. In FIG. 14, the variable j indicates the value of the board ID, and is assumed to be 1 to N (N: natural number). The variable i indicates a process order, and is assumed to be 1 to M (M: natural number). Hereinafter, for convenience of explanation, the variable j is described as a substrate j, and the variable i is described as a process i.

First, the production line simulation unit 14 sets 1 to the substrate j (step S61). The production line simulation unit 14 sets 1 to the process i (step S62). The production line simulation unit 14, the processing time .tau. _'I step i in the substrate _j, the _j, set the ideal processing time _{x i} of step i (step S63). The ideal processing time xi of the process _i is obtained from the configuration information 21.

  Then, the production line simulation unit 14 adds 1 to the process i in order to perform the next process (step S64). The production line simulation unit 14 determines whether or not the process i is larger than the maximum number M in the process order (Step S65). If the process i is not larger than the maximum number M in the process order (Step S65; No), the production line simulation unit 14 proceeds to Step S63 to set the processing time of the next process.

  On the other hand, when the process i is larger than the maximum number M in the process order (Step S65; Yes), the production line simulation unit 14 adds 1 to the substrate j to make it the next substrate (Step S66). The production line simulation unit 14 determines whether the substrate j is larger than the maximum number N of the substrates (step S67). If the substrate j is not larger than the maximum number N of the substrates (Step S67; No), the production line simulation unit 14 proceeds to Step S62 to set the processing time of each process in the next substrate.

On the other hand, when the substrate j is larger than the maximum number N of the substrates (Step S67; Yes), the production line simulation unit 14 determines the start time S _i, j of the process i on the substrate j and the output time of the process i on the substrate j. Calculate E _{i, j} . The production line simulation unit 14 sets the start time S _1,1 of the process 1 on the substrate ₁ to 0. The production line simulation unit 14 sets a value obtained by adding the start time S _1,1 and the processing time τ ′ _1,1 to the output time E _1,1 of the process 1 on the substrate 1 (step S68). Here, the processing time τ ′ _1,1 is an ideal processing time of the process 1 on the substrate 1.

Then, the production line simulation unit 14 sets 2 to the process i (step S69). The production line simulation unit 14 sets the output time E _i−1,1 of the _immediately preceding process on the same substrate 1 as the start time S _i, 1 of the process i on the substrate 1. The production line simulation unit 14 sets a value obtained by adding the processing time τ ′ _{i, 1} to the start time S _{i, 1} to the output time E _i, 1 of the process i on the substrate 1 (step S70).

  Then, the production line simulation unit 14 adds 1 to the process i in order to perform the next process (step S71). The production line simulation unit 14 determines whether or not the process i is larger than the maximum number M in the process order (Step S72). If the process i is not greater than the maximum number M in the process order (Step S72; No), the production line simulation unit 14 proceeds to Step S70 to calculate the start time and output time of the next process on the substrate 1. I do.

On the other hand, when the process i is larger than the maximum number M in the process order (Step S72; Yes), the production line simulation unit 14 sets 2 to the substrate j (Step S73). The production line simulation unit 14 sets the output time E1 _{, j-1} of the same process 1 on the immediately preceding substrate j-1 as the start time S1 _{, j} of the process 1 on the substrate j. The production line simulation unit 14 sets a value obtained by adding the processing time τ ′ _{1, j} to the start time S _{1, j} to the output time E _{1, j−1} of the process 1 on the substrate j (step S74).

Then, the production line simulation unit 14 sets 2 to the process i (step S75). Then, the production line simulation unit 14 sets the output time E _{i−1, j of the} previous process i−1 on the same substrate j as the output time E _{i−1, j} of the previous substrate on the start time S _i, j of the process i on the substrate j. , The larger of the output times _{Ei, j-1} of the same process i is set. The production line simulation unit 14 calculates the value obtained by adding the processing time τ ′ _{i, j} to the start time S _i, j of the process i on the substrate j, and the output time E _{i + 1, j−} of the next process on the previous substrate. to set the larger _one (step S76).

  Then, the production line simulation unit 14 adds 1 to the process i in order to perform the next process (step S77). The production line simulation unit 14 determines whether or not the process i matches the maximum number M in the process order (Step S78). If the process i does not match the maximum number M in the process order (step S78; No), the production line simulation unit 14 proceeds to step S76 to calculate the start time and output time of the next process on the substrate j. I do.

On the other hand, when the process i matches the maximum number M in the process order (Step S78; Yes), the production line simulation unit 14 sets the start time S _{i, j} of the process i on the substrate j to 1 on the same substrate j. one previous step i-1 of the output time E _{i-1, j} and output time E _i of the same step i in the previous _board, sets a larger _j-1. The production line simulation unit 14 sets a value obtained by adding the processing time τ ′ _{i, j} to the start time S _i, j of the process i on the substrate j (step S79).

  Then, the production line simulation unit 14 adds 1 to the substrate j to make the next substrate (step S80). The production line simulation unit 14 determines whether or not the board j is larger than the maximum number N of boards (step S81). If the substrate j is not larger than the maximum number N of substrates (step S81; No), the production line simulation unit 14 proceeds to step S74 to calculate the start time and output time of each process on the next substrate. .

On the other hand, when the substrate j is larger than the maximum number N of the substrates (Step S81; Yes), the production line simulation unit 14 calculates the output time of the process M on the substrate 1 from the output time EM _{, N} of the process M on the substrate _N. E _{M, 1} and calculates the subtraction value as production time _{T 0} (step S82). That is, the production line simulation unit 14 calculates the production time when the processing time of all the processes is set to the ideal processing time. Then, the production line simulation unit 14 ends the calculation process 0. Steps S68 to S81 of the calculation process 0 are referred to as “S _{i, j} , E _{i, j} calculation processes”.

[Flowchart of Calculation Processing A]
FIG. 15 is a diagram showing a flowchart of a process of calculating a production time when one process is an ideal processing time and the other processes are sampled processing times in a production line having no buffer between the processes. Note that in FIG. 15, the variable j indicates the value of the board ID, as in FIG. 13, and is 1 to N (N: natural number). The variables i and k indicate the order of steps, as in FIG. 13, and are 1 to M (M: natural number). Hereinafter, for convenience of explanation, a variable j is described as a substrate j, a variable i is described as a process i, and a variable k is described as a process k. The variable τ _{i, j} is the sampled processing time in the process i and the substrate j.

In the subsequent processing, the production line simulation unit 14 calculates the production time T _Ak for the lot u when the process k is an ideal processing time and the other processes are processing times sampled.

  First, the production line simulation unit 14 sets 1 to the process k (step S91). Then, the production line simulation unit 14 sets 1 to the substrate j (step S92). The production line simulation unit 14 sets 1 to the process i (step S93).

Then, the production line simulation unit 14 determines whether or not the process i matches the process k (step S94). If step i matches the process k (step S94; Yes), the production line simulation unit 14, a treatment time .tau. _'I of step _i, the _j, the ideal treatment time _{x i} step i in the substrate j (Step S95). The ideal processing time xi of the process _i is obtained from the configuration information 21. On the other hand, when the process i does not coincide with the process k (Step S94; No), the production line simulation unit 14 samples the process i on the substrate j during the processing time τ ′ _i, j of the process i on the substrate j. The processing time τ _{i, j} is set (step S96). The processing time τ _{i, j} sampled in the process i on the substrate j is obtained from the sampling information (processing time) 22.

  Then, the production line simulation unit 14 adds 1 to the process i in order to perform the next process (step S97). The production line simulation unit 14 determines whether or not the process i is larger than the maximum number M in the process order (Step S98). If the process i is not larger than the maximum number M in the process order (Step S98; No), the production line simulation unit 14 proceeds to Step S94 to set the processing time of the next process.

  On the other hand, when the process i is larger than the maximum number M in the process order (Step S98; Yes), the production line simulation unit 14 adds 1 to the substrate j to make the next substrate (Step S99). The production line simulation unit 14 determines whether the substrate j is larger than the maximum number N of the substrates (Step S100). If the substrate j is not larger than the maximum number N of the substrates (Step S100; No), the production line simulation unit 14 proceeds to Step S93 to set the processing time of each process in the next substrate.

On the other hand, when the substrate j is larger than the maximum number N of the substrates (step S100; Yes), the production line simulation unit 14 executes the “S _{i, j} , E _{i, j} calculation process” of the calculation process R ( Step S101). That is, the production line simulation unit 14 calculates the start time S _i, j of the process i on the substrate j and the output time E _{i, j} of the process i on the substrate j.

Then, the production line simulation unit 14 subtracts the output time EM _{, 1} of the process M on the substrate 1 from the output time EM _{, N} of the process M on the substrate _N. Then, the production line simulation unit 14 calculates the subtracted value as the production time T _Ak (u) (step S102). That is, the production line simulation unit 14 calculates the production time for the lot u when the process k is an ideal processing time and the other processes are sampled processing times.

Then, the production line simulation unit 14 adds 1 to the process k in order to change the process to be the ideal processing time (step S103). The production line simulation unit 14 determines whether the process k is larger than the maximum number M of processes (Step S104). If the process k is not larger than the maximum number M of processes (Step S104; No), the production line simulation unit 14 proceeds to Step S92 to calculate the production time T _Ak (u) for the next process k. .

  On the other hand, when the process k is larger than the maximum number M of processes (Step S104; Yes), the production line simulation unit 14 ends the calculation process A for the lot u.

[Flowchart of Calculation Processing B]
FIG. 16 is a diagram illustrating a flowchart of a process of calculating a production time when one process is sampled in a production line having no buffer between processes and the other process is an ideal process time. In FIG. 16, the variable j indicates the value of the board ID, as in FIG. 13, and is assumed to be 1 to N (N: natural number). The variables i and k indicate the order of steps, as in FIG. 13, and are 1 to M (M: natural number). Hereinafter, for convenience of explanation, a variable j is described as a substrate j, a variable i is described as a process i, and a variable k is described as a process k. The variable τ _{i, j} is the sampled processing time in the process i and the substrate j.

In the subsequent processing, the production line simulation unit 14 calculates the processing time of sampling the process k for the lot u and the production time _TBk when the other processes are set to the ideal processing time.

  First, the production line simulation unit 14 sets 1 to the process k (step S111). Then, the production line simulation unit 14 sets 1 to the substrate j (step S112). The production line simulation unit 14 sets 1 to the process i (step S113).

Then, the production line simulation unit 14 determines whether or not the process i matches the process k (step S114). If step i does not match the process k (step S114; No), the production line simulation unit 14, a treatment time .tau. _'I of step _i, the _j, the ideal treatment time _{x i} step i in the substrate j (Step S115). The ideal processing time xi of the process _i is obtained from the configuration information 21. On the other hand, when the process i matches the process k (Step S114; Yes), the production line simulation unit 14 samples the process i on the substrate j at the processing time τ ′ _i, j of the process i on the substrate j. The processing time τ _{i, j} is set (step S116). The processing time τ _{i, j} sampled in the process i on the substrate j is obtained from the sampling information (processing time) 22.

  Then, the production line simulation unit 14 adds 1 to the process i so as to perform the next process (step S117). The production line simulation unit 14 determines whether or not the process i is larger than the maximum number M in the process order (Step S118). If the process i is not larger than the maximum number M in the process order (Step S118; No), the production line simulation unit 14 proceeds to Step S114 to set the processing time of the next process.

  On the other hand, when the process i is larger than the maximum number M in the process order (Step S118; Yes), the production line simulation unit 14 adds 1 to the substrate j so as to be the next substrate (Step S119). The production line simulation unit 14 determines whether the substrate j is larger than the maximum number N of the substrates (Step S120). If the substrate j is not larger than the maximum number N of the substrates (Step S120; No), the production line simulation unit 14 proceeds to Step S113 to set the processing time of each process in the next substrate.

On the other hand, when the substrate j is larger than the maximum number N of the substrates (step S120; Yes), the production line simulation unit 14 executes the “S _{i, j} , E _{i, j} calculation process” of the calculation process R ( Step S121). That is, the production line simulation unit 14 calculates the start time S _i, j of the process i on the substrate j and the output time E _{i, j} of the process i on the substrate j.

Then, the production line simulation unit 14 subtracts the output time EM _{, 1} of the process M on the substrate 1 from the output time EM _{, N} of the process M on the substrate _N. Then, the production line simulation unit 14 subtracts the output time EM _{, 1} of the process M on the substrate 1 from the output time EM _{, N} of the process M on the substrate _N. Then, the production line simulation unit 14 calculates the subtracted value as the production time T _Bk (u) (step S122). In other words, the production line simulation unit 14 calculates the processing time for the lot u when the process k is sampled, and the production time when the other processes are the ideal processing time.

Then, the production line simulation unit 14 adds 1 to the process k in order to change the process to be the ideal processing time (step S123). The production line simulation unit 14 determines whether the process k is larger than the maximum number M of processes (Step S124). If the process k is not larger than the maximum number M of processes (Step S124; No), the production line simulation unit 14 _proceeds to Step S112 to calculate the production time T _Bk (u) for the next process k. .

  On the other hand, when the process k is larger than the maximum number M of processes (Step S124; Yes), the production line simulation unit 14 ends the calculation process B for the lot u.

[Effects of Embodiment]
According to the above embodiment, when there is no buffer between a plurality of processes in a production line having a plurality of processes, the evaluation apparatus 1 can produce a predetermined number of substrates by using a processing time model of each process. The processing time of each process for each substrate when the processing is performed is sampled. The evaluation device 1 calculates the production time, which is the time required for the production line to perform production, using the processing time of each process for each sampled substrate. The evaluation device 1 repeats the process of calculating the production time a plurality of times while changing the sampling process time. The evaluation device 1 evaluates the throughput of the production line using the production time calculated each time. According to such a configuration, the evaluation device 1 can quantitatively and accurately evaluate the throughput of the production line, that is, the productivity.

  Further, according to the above embodiment, the evaluation device 1 calculates the distribution of the production time using the production time calculated each time. According to this configuration, the evaluation device 1 can improve the production time according to the purpose, for example, by calculating the distribution of the production time. For example, for the purpose of shortening the production time on average, the evaluation device 1 can move the high frequency part in the direction of the production time smaller by eliminating the cause of the high frequency part. it can. In addition, in the case where the evaluation device 1 aims to avoid a part where the production time is greatly extended even if the part has a low frequency, the frequency of the part is moved in a lower direction by eliminating the cause of the part. Can be done.

  Further, according to the above embodiment, the evaluation device 1 calculates the first production time indicating the time during which a predetermined number of substrates are produced from the processing time of each process for each sampled substrate. The evaluation device 1 calculates a second production time indicating a time during which a predetermined number of substrates are produced when the sampling values of the processing times of the plurality of processes are replaced with ideal values. The evaluation device 1 indicates the time during which a predetermined number of substrates are produced when the sampling value of the processing time of the process to be evaluated is replaced with an ideal value and the processing time of the process other than the process to be evaluated is set as the sampling value. A third production time is calculated. The evaluation device 1 indicates the time during which a predetermined number of substrates are produced when the processing time of the process to be evaluated is set as a sampling value and the sampling value of the processing time of the process other than the process to be evaluated is replaced with an ideal value. A fourth production time is calculated. The evaluation device 1 calculates the influence on the throughput of the process to be evaluated using the first production time, the second production time, the third production time, and the fourth production time. According to such a configuration, the evaluation apparatus 1 calculates the influence of the evaluation target process on the throughput by using the dependency relationship between the evaluation target process and the process other than the evaluation target process, thereby obtaining the evaluation target process. Can accurately evaluate the effect on throughput. In addition, the evaluation device 1 can identify the process that causes the deterioration of the throughput by calculating the influence of each process on the throughput.

  Further, according to the above embodiment, the evaluation apparatus 1 calculates the distribution of the upper limit and the lower limit of the degree of influence on the throughput of the process to be evaluated. According to such a configuration, the evaluation device 1 can quantitatively evaluate the degree of influence on the throughput of the process to be evaluated.

  Further, according to the above embodiment, the evaluation device 1 outputs the distribution calculated by the evaluation process. According to such a configuration, the evaluation device 1 can visually or audibly evaluate the influence on the throughput.

[Others]
In addition, the evaluation device 1 is provided in a known information processing device such as a personal computer or a work station by adding the above-mentioned sampling generation unit 12, processing time adjustment unit 13, production line simulation unit 14, production time distribution calculation unit 15, influence distribution This can be realized by mounting each function such as the calculation unit 16.

  Also, each component of the illustrated apparatus does not necessarily need to be physically configured as illustrated. That is, the specific mode of the distribution / integration of the apparatus is not limited to the illustrated one, and all or a part of the apparatus is functionally or physically distributed / integrated in an arbitrary unit according to various loads and usage conditions. Can be configured. For example, the sampling generation unit 12 and the processing time adjustment unit 13 may be integrated as one unit. On the other hand, the production time distribution calculation unit 15 may be divided into a first calculation unit that calculates the distribution with respect to the production time and a second calculation unit that calculates the estimated value of the throughput. In addition, the storage unit 20 may be stored in an external device of the evaluation device 1, or the external device storing the storage unit 20 may be connected to the evaluation device 1 via a network.

  Further, the various processes described in the above embodiments can be realized by executing a prepared program on a computer such as a personal computer or a workstation. Therefore, an example of a computer that executes an evaluation program that realizes the same function as the evaluation device 1 illustrated in FIG. 3 will be described below. FIG. 17 is a diagram illustrating an example of a computer that executes an evaluation program.

  As illustrated in FIG. 17, the computer 200 includes a CPU 203 that executes various types of arithmetic processing, an input device 215 that receives input of data from a user, and a display control unit 207 that controls a display device 209. Further, the computer 200 includes a drive device 213 that reads a program or the like from a storage medium, and a communication control unit 217 that exchanges data with another computer via a network. Further, the computer 200 has a memory 201 for temporarily storing various information and an HDD 205. The memory 201, CPU 203, HDD 205, display control unit 207, drive device 213, input device 215, and communication control unit 217 are connected by a bus 219.

  The drive device 213 is a device for the removable disk 211, for example. The HDD 205 stores an evaluation program 205a and evaluation processing related information 205b.

  The CPU 203 reads the evaluation program 205a, expands it in the memory 201, and executes it as a process. Such a process corresponds to each functional unit of the evaluation device 1. The evaluation process-related information 205b stores information such as configuration information 21, sampling information (processing time) 22, production time distribution information 23, and influence degree distribution information 24.

  Note that the evaluation program 205a does not necessarily need to be stored in the HDD 205 from the beginning. For example, the program is stored in a “portable physical medium” such as a flexible disk (FD), a CD-ROM, a DVD disk, a magneto-optical disk, or an IC card inserted into the computer 200. Then, the computer 200 may read out the evaluation program 205a from these and execute it.

  Regarding the embodiment including the above-described examples, the following supplementary notes are further disclosed.

(Appendix 1)
In a production line having a plurality of steps, if there is no buffer between the steps of the plurality of steps, or if the buffer has a buffer capacity smaller than a specified amount even if there is a buffer, a processing time model of each step is used. The processing time of each process for each production object when a predetermined number of production objects is produced is sampled, and the production line is produced by using the processing time of each process for each sampled production object. Calculate the production time, which is the time required for
The process of calculating the production time is repeated a plurality of times by changing the processing time of sampling,
An evaluation program for executing a process of evaluating the throughput of the production line using the production time calculated each time.

(Supplementary Note 2) The process of evaluating
The evaluation program according to claim 1, wherein the distribution of the production time is calculated using the production time calculated each time.

(Supplementary Note 3) The process of calculating the production time includes:
Calculating a first production time indicating a time during which a predetermined number of production objects are produced from the processing time of each process for each sampled production object;
Calculating a second production time indicating a time during which a predetermined number of production objects are produced when the sampling values of the processing times of the plurality of processes are replaced with ideal values;
A third value indicating the time during which a predetermined number of production objects are produced when the sampling value of the processing time of the process to be evaluated is replaced with an ideal value and the processing time of the process other than the process to be evaluated is set as the sampling value. Calculate the production time of
When the processing time of the process to be evaluated is a sampling value and the sampling value of the processing time of the process other than the process to be evaluated is replaced with an ideal value, a time indicating a time when a predetermined number of production objects are produced is Calculate the production time of 4,
The process of evaluating uses the first production time, the second production time, the third production time, and the fourth production time to calculate the degree of influence on the throughput of the process to be evaluated. 2. The evaluation program according to claim 1, wherein

(Supplementary Note 4) The processing to be evaluated includes:
4. The evaluation program according to claim 3, wherein a distribution of an upper limit and a lower limit of the degree of influence of the process to be evaluated on the throughput is calculated.

(Supplementary note 5) The evaluation program according to Supplementary note 2 or 4, wherein the distribution calculated by the evaluating process is output.

(Supplementary Note 6) In a production line having a plurality of steps, when there is no buffer between the steps of the plurality of steps, or when the capacity of the buffer is smaller than a specified amount even if there is a buffer, the processing time of each step By using the model, the processing time of each step for each production object when a predetermined number of production objects is produced is sampled, and using the processing time of each step for each sampled production object, A first calculator for calculating a production time, which is a time required for the production line to produce,
A second calculating unit that repeats the processing by the first calculating unit a plurality of times while changing a sampling processing time;
An evaluation unit for evaluating the throughput of the production line using the production time calculated each time.

(Appendix 7)
In a production line having a plurality of steps, if there is no buffer between the steps of the plurality of steps, or if the buffer has a buffer capacity smaller than a specified amount even if there is a buffer, a processing time model of each step is used. The processing time of each process for each production object when a predetermined number of production objects is produced is sampled, and the production line is produced by using the processing time of each process for each sampled production object. Calculate the production time, which is the time required for
The process of calculating the production time is repeated a plurality of times by changing the processing time of sampling,
An evaluation method characterized by executing each processing for evaluating the throughput of the production line using the production time calculated each time.

Reference Signs List 1 evaluation device 10 control unit 11 input data interpretation unit 12 sampling generation unit 13 processing time adjustment unit 14 production line simulation unit 15 production time distribution calculation unit 16 influence distribution calculation unit 17 result output unit 20 storage unit 21 configuration information 22 sampling information (processing time)
23 Production time distribution information 24 Impact distribution information

Claims (5)

On the computer,
In a production line having a plurality of steps, if there is no buffer between the steps of the plurality of steps, or if the buffer has a buffer capacity smaller than a specified amount even if there is a buffer, a processing time model of each step is used. The processing time of each process for each production object when a predetermined number of production objects is produced is sampled, and a predetermined number of production objects are produced from the processing time of each process for each sampled production object. Calculating the first production time indicating the time to be performed, and using the ideal processing time of each process for each sampled production target, replacing the sampled values of the processing times of the plurality of processes with ideal values. Calculating a second production time indicating a time during which a predetermined number of production objects are produced, processing time of each process for each of the sampled production objects, and Using the ideal processing time, the sampling value of the processing time of the process to be evaluated is replaced with the ideal value, and the processing time of the process other than the process of the evaluation target is used as the sampling value. A third production time indicating the time to be performed is calculated, and the processing time of the process to be evaluated is used as a sampling value using the processing time of each process and the ideal processing time of each process for each sampled production target. Calculating a fourth production time indicating a time during which a predetermined number of production objects are produced when the sampling value of the processing time of a process other than the process to be evaluated is replaced with an ideal value;
The process of calculating the first production time, the second production time, the third production time, and the fourth production time is repeated a plurality of times by changing the sampling time of each process ,
Using the first production time, the second production time, the third production time, and the fourth production time calculated each time, the throughput of the process to be evaluated in the production line is reduced. An evaluation program for executing a process of evaluating the degree of influence . The evaluating process includes:
The distribution of the production time is calculated using the first production time, the second production time, the third production time, and the fourth production time calculated each time. The evaluation program according to claim 1. The evaluating process includes:
The evaluation program according to claim 1 , wherein a distribution of an upper limit and a lower limit of the degree of influence of the process to be evaluated on the throughput is calculated. In a production line having a plurality of steps, if there is no buffer between the steps of the plurality of steps, or if the buffer has a buffer capacity smaller than a specified amount even if there is a buffer, a processing time model of each step is used. The processing time of each process for each production object when a predetermined number of production objects is produced is sampled, and a predetermined number of production objects are produced from the processing time of each process for each sampled production object. Calculating the first production time indicating the time to be performed, and using the ideal processing time of each process for each sampled production target, replacing the sampled values of the processing times of the plurality of processes with ideal values. Calculating a second production time indicating a time during which a predetermined number of production objects are produced, processing time of each process for each of the sampled production objects, and Using the ideal processing time, the sampling value of the processing time of the process to be evaluated is replaced with the ideal value, and the processing time of the process other than the process of the evaluation target is used as the sampling value. A third production time indicating the time to be performed is calculated, and the processing time of the process to be evaluated is used as a sampling value using the processing time of each process and the ideal processing time of each process for each sampled production target. A first calculating unit that calculates a fourth production time indicating a time during which a predetermined number of production objects are produced when a sampling value of a processing time of a process other than the process to be evaluated is replaced with an ideal value. When,
A second calculating unit that repeats the processing by the first calculating unit a plurality of times by changing a processing time for sampling in each step ;
Using the first production time, the second production time, the third production time, and the fourth production time calculated each time, the throughput of the process to be evaluated in the production line is reduced. An evaluation device, comprising: an evaluation unit that evaluates the degree of influence . Computer
In a production line having a plurality of steps, if there is no buffer between the steps of the plurality of steps, or if the buffer has a buffer capacity smaller than a specified amount even if there is a buffer, a processing time model of each step is used. The processing time of each process for each production object when a predetermined number of production objects is produced is sampled, and a predetermined number of production objects are produced from the processing time of each process for each sampled production object. Calculating the first production time indicating the time to be performed, and using the ideal processing time of each process for each sampled production target, replacing the sampled values of the processing times of the plurality of processes with ideal values. Calculating a second production time indicating a time during which a predetermined number of production objects are produced, processing time of each process for each of the sampled production objects, and Using the ideal processing time, the sampling value of the processing time of the process to be evaluated is replaced with the ideal value, and the processing time of the process other than the process of the evaluation target is used as the sampling value. A third production time indicating the time to be performed is calculated, and the processing time of the process to be evaluated is used as a sampling value using the processing time of each process and the ideal processing time of each process for each sampled production target. Calculating a fourth production time indicating a time during which a predetermined number of production objects are produced when the sampling value of the processing time of a process other than the process to be evaluated is replaced with an ideal value;
The process of calculating the first production time, the second production time, the third production time, and the fourth production time is repeated a plurality of times by changing the sampling time of each process ,
Using the first production time, the second production time, the third production time, and the fourth production time calculated each time, the throughput of the process to be evaluated in the production line is reduced. Evaluating the degree of influence An evaluation method characterized by executing each processing.

Images