JPH10283014A - Load progress prediction method and system therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To predict a product completion number and the time by predicting the progress of processing products based on stored process flow, in-process grasping information, reference production period setting, management unit time setting and throughput setting. SOLUTION: This load progress prediction system 1 is provided with a process flow storage means 3 for storing the process flow for indicating through which processes from feeding to completion the completion is performed for respective item kinds, a reference production period setting means 4 for storing a reference production period, a management unit setting means 5 for storing management unit time and a throughput setting means 6 for storing throughput. Then, the process flow, the reference production period, the management unit time and the throughput are data-inputted by a user by an input means 2. Also, an in-process grasping means 7 grasps the in-process recognition information in real time and stores it. A load progress prediction means 8 manages the process flow, the reference production period, the management unit time, the throughput and the in-process grasping information and predicts the progress of the products.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and a system for predicting the progress of a process management load in product manufacturing, and more particularly to a method and system for predicting the progress of a product in each process of a semiconductor device manufacturing line comprising a plurality of continuous processes. It is about effective technology.

[0002]

2. Description of the Related Art Customer satisfaction in the manufacturing field is as follows.
Grasping customer needs, (2) production according to needs,
(3) It depends on whether the cycle of receiving orders according to production volume works well.

[0003] In (1), the marketing department, (2), the manufacturing department, and (3), the sales department, are important issues when receiving orders according to the production volume.
It is the accuracy of the number of completed products, which products can be produced when and how many.

According to the study by the present inventors, as a method of estimating the number of completed products based on actual work in progress, for example, “Produ,” published by Nikkan Kogyo Shimbun in 1974,
ction and Inventory Manag
element in theComputer Ag
e ", Oliver W. As shown in Wright (author), Ryuichi Yoshitani (translation), pages 26 to 43, a reference production period is set, the remaining processes are determined from the process in which the product is being processed, and the reference production periods are added. There is.

[0005]

However, the present inventor has found that the above-described method of estimating the number of completed products has the following problems.

That is, although the calculation time is short because the calculation method of the prediction is simple, a fixed reference production period is used on the assumption that the capability of the process and the load in the process are balanced. There is a problem that if there is a variation in the process of (1), a calculation is performed in which a large number of products that cannot be actually processed are processed at once.

It is an object of the present invention to provide a load progress prediction method and system capable of predicting the number of completed products and the completion time in a short time and with high accuracy.

The above and other objects and novel features of the present invention will become apparent from the description of the present specification and the accompanying drawings.

[0009]

SUMMARY OF THE INVENTION Among the inventions disclosed in the present application, the outline of a representative one will be briefly described.
It is as follows.

That is, the load progress predicting method of the present invention comprises:
Storing the process flow indicating the order in which the products are processed in the first storage means, and storing the work-in-process grasping information for grasping the work-in-process for each type of each process in the second storage means; Storing in a third storage means a reference production period setting which is data of a reference production period for each process, and storing a management unit time setting which is data defining a management unit time in a fourth storage means. And storing the processing capacity setting, which is data of the quantity that can be processed for each management unit time for each process and product type, in the fifth storage means, the stored process flow, in-process grasp information, and reference production period. The load progress predicting means predicts the progress of the product through the process to be processed based on the setting, the management unit time setting and the processing capacity setting, and predicts the number of completed products and the timing thereof. It is intended.

In the load progress prediction method according to the present invention, the step of predicting the number of completed products and the timing of the completion includes storing a stored process flow, in-process grasp information, standard production period setting, management unit time setting, and processing capacity setting. Calculating the progress of the process in progress based on, and, from the result of the calculation, the step of obtaining the relationship between the number of completed products, the number of products arriving at the process in which the process is performed, and the process of the process in which the product is processed, And a step of predicting the number of completed products and the timing based on the relationship.

As a result, it is possible to predict the load progress in a short time with high accuracy and in accordance with the change.

Further, in the load progress prediction method according to the present invention, the in-process grasping information is not the actual in-process but the quantity of the kind to be produced in the future.

Thus, it is possible to know the state of the line when the product configuration changes.

Further, in the load progress prediction method of the present invention, the processing capacity setting is not an actual capacity but a fictitious capacity setting in consideration of future investment in a line or disposal of equipment.

Thus, the policy of increasing or decreasing the equipment can be easily and quickly evaluated.

In the load progress prediction method according to the present invention, the load progress prediction means predicts the progress of the product through the process based on the in-process grasp information of each process and the processing capacity setting. Have

Thus, the production period can be easily and quickly evaluated.

Further, the load progress prediction system according to the present invention has a first storage means for storing a process flow indicating an order in which products are processed, and a process in progress grasping a process in process for each type of each process. A second storage unit for storing information; a third storage unit for storing a reference production period setting which is data of a reference production period for each type of process; and a management unit time setting which is data for defining a management unit time. A fourth storage unit for storing processing capacity setting, which is data of a quantity that can be processed for each management unit time for each process and type, and a storage unit for storing the first to fifth storage units. Based on the stored data, the progress of the product through the process is calculated, and the load progress predicting means for predicting the number of completed products and the timing thereof, and outputting the result predicted by the load progress predicting means Prediction result output means , Process flow, processing capability setting,
Input means for inputting data such as reference production period setting and management unit time setting, and online connection with the load progress prediction means, collects the results of each process, and provides real-time in-process grasp information to the load progress prediction means. It consists of output result collection means.

As a result, it is possible to predict the load progress in a short time with high accuracy and according to the change.

As described above, accurate product inventory information can be appropriately provided, so that the efficiency of business activities and the like can be greatly improved.

[0022]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram of a load progress prediction system according to an embodiment of the present invention, FIG. 2 is an operation flowchart of the load progress prediction system according to an embodiment of the present invention, and FIG. Explanatory diagram showing the process,
FIG. 4 is an explanatory diagram of load progress prediction according to one embodiment of the present invention, and FIG. 5 is a diagram illustrating a display example of load progress prediction in a prediction result output unit according to one embodiment of the present invention.

In the present embodiment, the load progress prediction system 1 that performs inventory prediction accurately in a short time in accordance with a change in semiconductor device production and provides inventory information that can appropriately perform business activities is provided by a load progress prediction system 1 in which various users are provided. An input means 2 such as a keyboard for inputting data is provided.

Further, the load progress prediction system 1 is provided with a process flow registering means (first storing means) 3 for storing a process flow indicating through which process, from input to completion, each type is completed. I have.

Further, the load progress prediction system 1 has at least a line name, a process name, and a reference production period, and considers that it is necessary to pass through the process for predicting the progress of a product for each process in each line. Reference production period setting means (third storage means) 4 is provided for storing a reference production period (reference production period setting) which is data of a period for each product type.

The load progress prediction system 1 has a management unit time (management time) which is data of a minimum time length when considering the movement of a product in a line in order to predict the progress according to the actual situation of the line. Management unit time setting means (fourth storage means) 5 for storing the unit time setting is provided.

Further, the load progress prediction system 1 is provided with a processing capacity setting means (fifth storage means) 6 for storing a processing capacity (processing capacity setting) which is a quantity that can be processed within a management unit time for each type of process. Have been.

The user inputs data of the process flow, the reference production period, the management unit time, and the processing capacity by the input means 2.

Next, the load progress prediction system 1 is in-process information which includes at least the type and the number of products in process for each process collected by the existing performance collection system JS such as a workstation. In-process grasping means (second storage means) 7 for grasping and storing the grasping information in real time is provided.

The load progress predicting system 1 includes a load progress predicting means 8 for managing the process flow, the reference production period, the management unit time, the processing capacity and the information on the in-process grasp, and predicting the progress of the product. Is provided.

Further, the load progress prediction system 1 is provided with a prediction result output means 9 for displaying or printing out the result calculated by the load progress prediction means 8.

The process flow registration means 3, the reference production period setting means 4, the management unit time setting means 5, the processing capacity setting means 6, the in-process grasping means 7, and the load progress prediction means 8 constitute a server SV. I have.

Further, the load progress prediction system 1 includes an input means for inputting data to the server SV, a prediction result output means 9a for displaying or printing out various data such as a result calculated by the load progress prediction means 8, and the like. Is provided.

Then, the server SV and the result collection system J
S and the client subsystem CS are connected online, for example, by a line L such as a dedicated line or a telephone line.

Next, the operation of this embodiment will be described with reference to FIGS.
Using the flowchart in the load progress prediction system 1 of FIG. 2, the explanatory diagram showing the process of semiconductor production of FIG. 3, the explanatory diagram of the load progress prediction of FIG. 4, and the display example of the load progress prediction in the prediction result output means of FIG. explain.

First, the user inputs the data of the process flow, the reference production period, the management unit time, and the processing capacity from the input means 2 (step S101).

In the in-process grasp information, the result collection system is collected by the existing result collection system JS (step S101).

Further, the data of the process flow, the standard production period, the management unit time, the processing capacity and the in-process grasp information are stored in the process flow registration unit 3, the standard production period setting unit 4, the management unit time setting unit 5, It is stored in the capability setting means 6 and the in-process grasping means 7 (step S10).
2).

Next, the reference production period is a time value considered necessary from the start of the product in the process to the end thereof in the process.

In this case, the average value during the actual production period may be used, or a value obtained by adding a margin to the average value may be used. In any case, this value should be set according to the judgment of the line manager according to the actual condition of the line. It is assumed that the information includes at least a line name, a process name, a product type name, and a reference production period.

By the way, setting a reference production period for each process for each product type requires a great deal of labor.
Actually, the reference production period is set collectively for products having the same reference production period in terms of products having the same attribute, for example, package type, number of pins, memory capacity, and the like.

At this time, it is easy to manage by setting a product code according to the attribute for each product type and specifying the reference production period for each product code.

Next, the management unit time setting means 5 is data for setting the minimum length of time when considering the movement of a product in a line. This data is used for predicting the progress according to the actual situation of the line. It is.

For example, although the processing time of a product is in units of one day, when the progress is predicted, calculation in units of one second simply increases the calculation time unnecessarily. However, since this value is the minimum time value in calculation, an appropriate management unit time is set so that the deviation from the actual state is reduced when the reference production period is expressed by an integral multiple of the management unit time. .

It is to be noted that the setting can be common to all.
Alternatively, it can be set for each product type and each process.
Further, similarly to the reference production period, it is also possible to set a product code for each product type and set the product code for each product code.

Next, the processing capacity will be described.

The processing capacity is the quantity that can be processed within the unit time of management for each type of process. In actual lines, the ability is often captured on a daily basis.

In such a case, the user interface is set to the capacity on a daily basis, and the following equation 1 is converted inside the load progress prediction means 8.

Processing capacity = Processing capacity per day × Management unit time (Equation 1) Also, regarding the setting of the processing capacity, the product code is set for the product type and the product code is set in the same manner as in the reference production period. Is also possible.

Next, an outline of semiconductor production will be described with reference to FIG.

As shown in FIG. 3, in the semiconductor device, after forming an electric circuit on a semiconductor wafer such as a silicon substrate in a circuit forming step (step S201), a function test is performed in a test step (step S202). Package with resin for circuit protection in the assembly process (Step S20)
3) Inspection is performed to determine whether or not a desired function is performed in the sorting process, thereby completing the process (step S204).

In a test process or a sorting process, a semiconductor device determined to be defective is discarded without being passed to subsequent processes, and in the sorting process, products may be classified according to inspection results. is there.

Then, when the finished product is completed, it is stored in the finished product warehouse (step S205). Further, the sales department carries out an order receiving activity based on the stock in the warehouse and the future warehousing schedule (step S206).

Next, a description will be given here on the assumption that the load progress prediction means serving as a core of the storage prediction is of one type.

First, the basic concept will be described with reference to FIG.
This will be described with reference to FIG.

As shown in FIG. 4, the horizontal axis represents time, and the vertical axis represents the progress of processing in the process. The matrix indicates the amount of work in progress of the product, and moves from the upper left to the lower right as time passes.

However, it is assumed that the management unit time is one day or less. The time is represented by a combination of a day and a management unit time, such as a t2 management unit time on the t1 day.

Here, the in-process in the n-th step is defined as WIP (n,
For example, in FIG. 4, a portion indicated by a thick frame in FIG. 4 indicates a work in progress WIP (n, n) having a processing progress of 3 in two unit times on the first day in the n-th process.
1, 2, 3). Here, the processing progress is
It shows the progress of processing in the equipment.

Further, the completed quantity of the n-th step is represented by OUT (n,
For example, when expressed by (t1, t2), the shaded portion indicates the quantity OUT (n, 1, 2) completed in two unit times on the first day in the n-th process.

Further, when the quantity arriving from the (n-1) th process to the (n) th process is represented by IN (n, t1, t2), for example, the portion indicated by hatching is the two unit time of the first day in the (n) th process. Arrived at IN (n, 1,2)
Is shown.

Here, assuming that the transport time between the n-th step and the (n + 1) -th step can be ignored, the relationship between OUT and IN is expressed by the following equation.

IN (n + 1, t1, t2) = OUT (n, t1, t2) (Equation 2) Assuming that the maximum value of t2 is MAXt2, MAXt2 becomes
Is represented by

MAXt2 = 1 ÷ management unit time (Equation 3) As can be seen from Equation 3, the management unit time is preferably a value whose reciprocal is an integer.

Next, as to the degree of processing progress, 1 to 4 means that processing is being performed in the equipment, and 5 means that processing is waiting. Assuming that the maximum value of the processing progress during the processing in the equipment is MAXp ′, MAXp ′ is expressed by Expression 4.

MAXp ′ = reference production period / management unit time (Equation 4) Further, the processing progress of the processing waiting is always a value obtained by adding 1 to MAXp ′.

Next, the progress of the work in progress on the t1 day is calculated by the load progress estimating means 8 according to Equations 5 to 8 to calculate the degree of progress of the work in progress (Step S103). The relationship between the work in progress and completion is represented by Equation 5.

OUT (n, t1, t2 + 1) = WIP (n, t1, t2, 1) (Equation 5) where t2 = 1,..., MAXt2-1, and when p is less than MAXp ′, WIP (N, t1, t2, p)
Is the work in process of the equipment, and this work is paid out sequentially, and the maximum value of WIP (n, t1, t2, p) is
It is the quantity that can be processed within the management unit time in the process n, that is, the processing capacity.

Next, the transition of the in-process in the same process is expressed by the following equation (6).
And Equation 7:

When p = 1,..., MAXp′−1, WIP (n, t1, t2 + 1, p) = WIP (n, t1, t2, p + 1) (Equation 6) When p = MAXp ′, WIP (n, t1, t2 + 1, p) = min (WIP (n, t1, t2, p + 1), processing capacity) (Equation 7) Here, min (A, B) takes a smaller value between A and B. Define as a function.

Further, the relation between the arrival and the work in progress is expressed by Expression 8.

WIP (n, t1, t2 + 1, MAXp ′ + 1) = IN (n, t1, t2) + max (WIP (n, t1, t2, MAXp ′ + 1) −WIP (n, t1, t2 + 1, MAXp ′) ), 0) (Equation 8) Here, max (A, B) is defined as a function having a larger value between A and B.

By the way, when t2 = 1, WIP and O
UT and IN are represented by Equations 9 to 12.

OUT (n, t1, 1) = WIP (n, t1-1, MAXt2, 1) (Expression 9) When p = 1,..., MAXp′−1, WIP (n, t1, 1) , P) = WIP (n, t1-1, MAXt2, p + 1) (Equation 10) When p = MAXp ′, WIP (n, t1, 1, p) = min (WIP (n, t1-1, MAXt2, (p + 1), processing capacity) (Equation 11) Here, min (A, B) is defined as a function having a smaller value between A and B.

WIP (n, t1, 1, MAXp ′ + 1) = IN (n, t1, 1) + max (WIP (n, t1-1, MAXt2, MAXp ′ + 1) −WIP (n, t1, t2 + 1, MAXp) '), 0) (Equation 12) Here, max (A, B) is defined as a function having a larger value between A and B.

Then, based on the calculation results of these equations 5 to 8, the load progress prediction means 8 performs the calculations of the above equations 9 to 12 (step S104). Therefore, Equations 5 to 12
Can be used to calculate the progress of the work in progress on day t1.

By using these sequentially, it is possible to calculate the progress of the work in progress for the next step on the next day.

As described above, the case of one type has been described. In the case of multiple types, it is sufficient to calculate separately for each type within the same process.

When various setting values are specified by the attributes of the product, when proceeding to the next step, it is necessary to divide the work in accordance with the way of compiling the attributes in the next step. You need to know the breakdown.

[0109] Then, the warehouse prediction is performed by the above-described method, and the transition of the work in process of the finished product warehouse is obtained by the load progress prediction means 8 from this and the information on the warehouse according to the order (step S 105). As shown, the load progress prediction data is displayed on the prediction result output means 9 (FIG. 1) (step S106).

The load progress prediction data obtained by the load progress prediction means 8 is displayed on a prediction result output means 9a provided in the client subsystem CS by inputting a predetermined command from the client subsystem CS. You can also.

Thus, in the present embodiment, the load progress prediction system 1 accurately performs a warehouse prediction in the production of semiconductor devices in a short time in accordance with a change, and provides inventory information capable of appropriately conducting business activities. Can be.

Further, according to the present embodiment, the information on the in-process grasp is input as the actual in-process. However, for example, by inputting the quantity in progress of the kind to be produced in the future as the in-process grasp information, the product configuration can be improved. It is possible to predict the state of the line when it changes.

Further, although the processing capacity is input as the capacity to be actually processed, a fictitious capacity in consideration of investment in the future line and disposal of the equipment is input as the processing capacity, so that the policy of increasing or decreasing the equipment can be set. It can be evaluated easily and in a short time.

Further, the production period can be easily and quickly evaluated by estimating the progress of a product being processed and going through the process based on the in-process grasp information and the processing capability of each process.

Although the invention made by the inventor has been specifically described based on the embodiments of the present invention, the present invention is not limited to the above-described embodiments, and various modifications may be made without departing from the gist of the invention. Needless to say, it can be changed.

[0087]

Advantageous effects obtained by typical ones of the inventions disclosed by the present application will be briefly described as follows.
It is as follows.

(1) According to the present invention, the number of completed products and the time of completion can be predicted accurately and in a short time according to changes from the reference production period, processing capacity, and work-in-process information.

(2) Further, according to the present invention, according to the above (1), it is possible to provide appropriate warehousing prediction and inventory information.
Operating efficiency can be greatly improved.

[Brief description of the drawings]

FIG. 1 is a block diagram of a load progress prediction system according to an embodiment of the present invention.

FIG. 2 is an operation flowchart of the load progress prediction system according to the embodiment of the present invention;

FIG. 3 is an explanatory view showing a semiconductor production process.

FIG. 4 is an explanatory diagram of load progress prediction according to one embodiment of the present invention.

FIG. 5 is a diagram showing a display example of load progress prediction in a prediction result output unit according to one embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Load progress prediction system 2 Input means 3 Process flow registration means (first storage means) 4 Reference production period setting means (third storage means) 5 Management unit time setting means (fourth storage means) 6 Processing capacity setting Means (fifth storage means) 7 Work in progress grasping means (second storage means) 8 Load progress prediction means 8 9, 9a Prediction result output means JS result collection system SV server CS client subsystem L line

Claims (6)

[Claims] 1. A load progress prediction method for predicting a progress of a product on a production line in which a product is completed by sequentially performing a process in a plurality of processes, wherein a process flow indicating a sequence in which the products are processed is indicated. Storing in the first storage means, storing the in-process grasp information for grasping the in-process in each type of each process in the second storage means, and data on the reference production period for each in-process type. Storing the reference production period setting in the third storage means, storing the management unit time setting, which is data defining the management unit time, in the fourth storage means, Storing the processing capacity setting, which is the data of the quantity that can be processed in the fifth storage means, the stored process flow, work-in-process grasping information, standard production period setting, management unit time setting and processing A load progress prediction method, comprising: a step in which a load progress prediction unit calculates a progress of a product through a process to be processed based on a capability setting, and predicts the number of completed products and the timing of the completion. 2. The load progress estimating method according to claim 1, wherein the step of estimating the number of completed products and the timing thereof includes the steps of storing stored process flow, in-process grasp information, setting a reference production period, setting a management unit time, and processing. Calculating the progress of the process in process based on the capacity setting; and obtaining the relationship between the number of completed products, the number of products arriving at the process to be performed, and the process of the product being processed based on the calculation result. And a step of predicting the number of completed products and the timing based on the relationship. 3. The load progress prediction method according to claim 1, wherein the in-process grasp information is not an actual work-in-progress but a quantity of a variety to be produced in the future. 4. The load progress prediction method according to claim 1, wherein said processing capacity setting is not an actual capacity but an imaginary one considering investment in a future line and disposal of equipment. A load progress prediction method, characterized in that the load progress is set. 5. The load progress predicting method according to claim 1, wherein the load progress predicting means is configured to determine whether a product is processed based on the in-process grasp information of each step and the processing capacity setting. And a step of predicting progress through the process. 6. A load progress prediction system for predicting the progress of a product on a production line in which a product is completed by sequentially performing processes in a plurality of processes, wherein a process flow in which the order in which the products are processed is indicated. A first storage unit for storing; a second storage unit for storing in-process grasp information for grasping a work-in-process for each type of process in each process; and a reference production period setting which is data of a standard production period for each type of process. And storage means for storing a management unit time setting which is data defining the management unit time, and data of a quantity which can be processed for each management unit time for each process and type. Fifth storage means for storing the processing capacity setting, and based on the data stored in the first to fifth storage means, calculate the progress of the product through the process steps, and calculate the number of completed products and the Load to predict timing Progress prediction means; prediction result output means for outputting a result predicted by the load progress prediction means; input means for inputting data such as the process flow, processing capacity setting, reference production period setting, and management unit time setting. A load progress prediction system which is connected online to the load progress prediction means, collects results of each process, and outputs the in-process grasp information to the load progress prediction means in real time. .