TW491802B - Automated material handling method basing on adaptability control - Google Patents

Abstract

The invention is used to elevate the efficiency of automated handling of material basing on the adaptability control in the semiconductor factory. The method in the invention includes: continuously recording the automated material handling data in the factory; analyzing the data within a specified period of time at each specified interval to create a probability table about the delivery destinations of the material; inputting the probability table about the delivery destinations of the material to the material distribution table in the automated material handling system (AMHS); and determining the following delivery destinations of the material by the AMHS after the AMHS references the renewed material distribution table. The method in the invention increases the accuracy in delivering the material to the destinations, so the frequencies occurred in the erroneous delivery and the jammed stock flow can be reduced to effectively raise up the delivery efficiency of the material in the field and further the production efficiency of the material.

Description

491802

V. Description of the invention (1) Description of the invention The things in the plant of efficiency The present invention relates to a method for improving the automatic transportation of materials', which is particularly suitable for a method for improving the automatic transportation efficiency of semiconductors based on adaptive control. BACKGROUND OF THE INVENTION The material handling of semiconductor manufacturing plants known by S is by hand, but as the size of the wafer increases to 12 pairs, plus the consideration of factors such as the cleanliness of work and transportation, automated material handling: Yield and Material + Automated

Ml Handling Systera (hereinafter referred to as AMHs) has recently become one of the necessary equipment for returning to the factory. The automated transportation of materials in the factory has been built in the factory. The materials are transported and stored on the self / provider, and the amount of a is reduced by manually moving the materials = = same shelf and time. For the extra labor and AMHS points consumed by the corpse, the transportation of the forecasted transportation must take care of the original user and progress at the same time. However, the system does not have a general table for material allocation. The requirement is correctness. Sending points is not expected. The storage output by the user is also delayed. And in an AMHS that often follows the AMHS rules, the material is managed to place the material in the most correct Shi Dan {Xiangyi, Fuzhengshi transported to the appropriate place: the transportation time and materials = = transported to the wrong location, this; ;; lose: way 'return from the wrong inventory to: Μ Processing ^ 'and must set up the main ^ ^ material processing time, external personnel to carry out the general semiconductor factory, and affect the close connection of product manufacturing, the main I ^ material manufacturing distribution of a variety of production -t 要疋 Because of those in the semiconductor factory: J receives early products and changes rapidly. The static allocation table (static dagger, material allocation table

491802 V. Description of the invention (2) The gf rules are usually preset values, and cannot be updated in real time to correspond to changes in production and operation modes of Ji. Therefore, they cannot quickly respond to changes in on-site production and lack flexibility. Ability. Therefore, in semiconductor manufacturing plants, the chance of AMHS erroneous shipping also increases with the complexity and diversification of manufactured products. μ From the perspective of management, from the perspective of management, the production of semiconductor factories generally has one: i: 5 朿 2 ί holds the capacity of the equipment to operate at full load ', but the actual operation: ί: Γ is quickly changing according to the different needs of the manufactured products . In the case of the conductor factory, 'Because of the diversified products produced by the order, the production is rapidly changing as the product is not good'. Therefore, the hall in the factory building must also be shipped in accordance with the manufacturing requirements. Change fast. i? Online workers with different values may have specific or preferred processing modes for the distribution of materials. The AMHS preset distribution method may not be integrated with factory production as an example. Generally, only materials or yellow light areas are designated in AMHS. However, the machine in this area is: 'Indicate the arrangement of a semiconductor factory in the region, there may be multiple machines in a yellow light area, and there are 16 knives (bays) for setting materials. ) 12, there may be a difference. F: As for the work, Ding Xi Zhi Shang Zuo Lu, although AMHS will deliver the material, it may not just deliver the material to the online personnel expected 2 shelves, at this time, the wrong transportation occurs. Online: Ai "Manually transport this batch of materials from the wrong shelf to mm 0503-6148TW; tsmc20〇〇.〇9i9; peggy ~ p ^ m page 5 ^ 1802 V. Description of the invention (3) Stored on the shelf on. In addition, in general, Aos cannot predict the plutonium on a specific shelf, so it may cause too much material on some shelves, forming a congestion situation, and some shelves may not have any materials on the shelf at all, resulting in equipment resource upload rate ^ ^ or unallocated The situation of congestion is more likely to lead to product manufacturing efficiency. Brief description of the invention In order to solve the above-mentioned general AMHS encountered in semiconductor manufacturing plants, an object of the invention is to propose an improved control based on adaptive control. ΞΠϊShipping method 'can simulate the online workers in a semiconductor factory to predict the delivery point of the material. The other purpose of material from ί = is to propose an accurate improvement method based on adaptive control, which can improve the automation of material transportation in the semiconductor factory + accuracy and reduce the occurrence of material return. The conductor is re-introduced for a purpose—improving semi-automated and adaptive control based on adaptive control. The carrier can balance the distribution of materialized transportation in the semiconductor factory and avoid congestion. Applicable i ί = i: ϊ Should ί control the improved method of automatic material transportation, analyze the data of automatic material transportation every-fixed time; transport the above-mentioned data to produce a material when the landing point base ^ r; and ϋ ^ Import material drop point probability table is imported into the latest material delivery position of amhs system. The new material drop dal point probability table determines the material transportation. The information to be sent can include: each batch, the mother; the area of the bucket (material area), and the material manufacturing stage 〇〇〇〇3.6l48TW; tsm ^^:

Peggy.ptd Page 6 491802 V. Description of the invention (4) 'Stage, manufacturing capability, next capability, and / or designated destination. The advantage of the method for improving the automatic material transportation based on adaptive control of the present invention is that it can increase the accuracy of AMHS system's prediction of material transportation, reduce the probability of material misfeed, reduce the time of material return, and balance the load of each inventory. Avoid stockpiles. In addition, the method of the present invention can automatically update the jtatic table of the assigned material transportation in the AMHS system, so that the material distribution rules in the material distribution table are more in line with the material transportation requirements of the semiconductor manufacturing plant with product diversification. . ▲ In order to make the above-mentioned objects, features, and advantages of the present invention more obvious and easy to understand, the following is a detailed description of specific embodiments in conjunction with the accompanying drawings, which are briefly described in the first figure. Schematic diagram of location distribution in the zone. Figure 2 shows the schematic control of a system architecture based on the automated transportation of materials in a good semiconductor according to an embodiment of the present invention. = 3. Figure 3 does not show a method for improving the automated transportation of materials in a semiconductor factory based on adaptive control according to an embodiment of the present invention. flow chart. The figure shows the material transportation data analysis diagram of the embodiment of the present invention. Zhu Shuming 101420 Yuguang District Machine AMHS Host 121622 Separated Block Shelf Storage Device

Page 7 491802 V. Description of the invention (5) 24: Data analysis host 28: Shelf C1 to C3: Manufacturing capacity S1 to S3: Manufacturing stage Example 26: AMHS system

Ml and M2: Analysis mode C ’1 to C’ 5: Next manufacturing capability P1 to P9: Probability

In the following, an AMHS system of a semiconductor factory is used for detailed description. See FIG. 2 'The operation of the amhs system 26 of a semiconductor factory is usually controlled by an AMHS mainframe 20, and an AMHS mainframe 20 is stored in the AMHS mainframe 20 to distribute material transportation rules. The AMHS mainframe of the present invention is, but is not limited to, generally, such as: each batch (area), example (capabi 1 washing or exposure of some materials), such as: the pre-cleaning step is in the yellow light area, and is mainly caused by In the embodiment of the step-by-step definition, a storage device 22 can be used to continuously record the material transportation data. The storage device 22 can be stored in a database to record the above data. Semiconductor factory Various materials may be generated during the transportation of materials, and the time of material transportation; if the landing area of the material transportation is from the CVD area to the yellow light area; the manufacturing capability of the machine at the landing location ^ y) There may be multiple types of machine ports Manufacturing capabilities, such as: next capabiiity, a certain processing step may only be a pre-process step, pre-cleaning, so the next major process is to produce important data; material manufacturing Stage (stage), especially this = stage may be a set of several steps in a specific order :: The manufacturing capabilities of the machine in the light zone are all exposed and developed, so the waste and special collections represent the batch. Manufacturing process making the material to distinguish particular stage in this area

491802 V. Description of the invention (6); In addition, the material transportation information may include (destination) ^ ^ a ,,,, and again :: to a specific location, so that the designated transportation end point is generated. Relevant talents generated when working for 4 people can be analyzed by one source: two to the material transport to produce a material transport landing point :: analysis; the machine 24 can not add the shelves regularly under various conditions Distribution of landing points ^ The probability table of the material transportation landing points shown in this probability table is loaded into the transportation process of the job host = 物 Π ::. + Knife allocation table is the material to be prepared. The method of the present invention will be described in detail with reference to the present drawing in FIG. 2. Carry the first 4 ^, and then use the 3rd automated material transportation data. In order to analyze the second step 32, the recording mode ′ needs to continuously collect the best material about the material automatic = sending. Therefore, a storage device 22 can be used. Hours: the time when the related materials are transported, the areas at both ends of the transport: each batch of material (capability), the next stage of manufacturing capacity (manufacturing power material manufacturing stage (stage), and also, = apabl, lity), (destination). The recorded data can be selected and recorded according to the characteristics of the end-of-conveyance of 22 =, which is based on the manufactured data of the manufacturing plant, and is not limited to the analysis of the material in the field transportation mode. Then proceed to step 34: fix the above every other The data is used to generate a material transport drop point machine ^ ^ knife analysis point probability table within a certain time. After continuous collection

491802

V. Description of the invention (7) The data in the time after the data generated in the AMHS system. For example: Time data. However, the present invention does not depend on the change of the manufacturing process of the analysis to be performed. The above steps can be based on a data and analysis of a certain period of time in the storage device 2 2 to produce a material drop point probability. 'The analysis will be performed every 6 hours after a certain period of time, and the analysis is currently limited to the first 12 hours. The time range for the analysis can be performed according to the manufacturing analysis host 24 of the manufacturer to periodically start the interval. Data mode. The data score table, the machine for estimating the location of the falling point of materials, and then referring to FIG. 4, is a schematic diagram showing the analysis of the material transportation data in an embodiment according to the present invention. The analysis of the data transmitted in a general semiconductor factory can be divided into two modes. The main analysis of the manufacturing capability (capability, c) of the material is to analyze the distribution of the next manufacturing capability (next CaPability, (:,) ^ rate (possibiiity, P). In the M2 mode, it is mainly in the half The yellow light area of the machine's machine has the same manufacturing capabilities. If it is analyzed by manufacturing (capabUHy, C), it will not show the change of the online knife. Therefore, the material manufacturing stage (s) is used as the data analysis & In order to produce a meaningful table of the probability of falling points of materials. Here the probability is displayed again and again ^ Under various conditions, the distribution of the falling points generated by the materials delivered to each shelf ^ is limited to the stage shown in the fourth figure, etc. The data is divided into various specific materials. However, the data analysis of the above steps is not only about manufacturing capability (capability) or manufacturing analysis, but also according to the manufacturing characteristics of each manufacturer.

491802 V. Description of the invention (8) Material delivery landing point machine for sending parameter data Then proceed to the main machine. After the fund is acquired, the newly allocated material among the AMHS mainframes is analyzed. The main analysis unit is 24. This point allocation table is near the current line. The last is the probability table. After determining the maximum point probability table, the updated reference is used to continue to follow the manufacturing. condition. The present invention can point the position, such as the advancement of the landing point of the special storage rack, so it will, and further improve. Although the present invention is invented, any skilled person can 'appropriate analysis' to obtain the most efficient table. Step 36: After the material is analyzed by the material analysis host, the probability of drop point of material transportation, and the drop point allocation table of material transportation is to determine the probability of completion of the analysis. Step 38: AMHS system When new materials are delivered, the AMHS host 24 specifies the entry and exit point allocation table, so that the AMHS system process is updated, because the machine and storage rack are determined according to the material manufacturing process, and the accuracy is measured. At the same time, the operating efficiency in the factory can also be used to make the overall effect in the factory. It is disclosed in the preferred embodiment that those skilled in the art are improperly altered or retouched, and the table of near-workers' expectations for the drop-off point is imported into AMHS 刖 1 2 The hourly material delivery data table is returned to the AMHS host 20, with a static table. The rules of material transportation, so the data table is imported into the drop point allocation table. The drop point probability table is updated to the point closest to the reference update. When the automation of loading and unloading of materials is generally dropped, it is closer to the factory's predicted materials. Therefore, the increase is due to the predicted shipping. The reduction rate is as above. However, it does not depart from the present invention. The operation within the sub-point distribution may be carried out when an area is delivered to the landing point by mistake. Table, the exact mechanism of the current storage and storage is not used to limit the spirit of the spirit and protect the fan house

0503.6148TW; tsmc2000-0919; peggy.ptd Page 11 491802 V. Description of the invention (9) Subject to the scope of the attached patent application. (nil p. 12 0503-6148TWF; t smc2000-0919; peggy.p t d

Claims (1)

1 · A law system based on adaptive control applicable to a semiconductor manufacturing plant includes: a method for improving the automated transportation of materials-an automated transportation system for materials, the transportation data generated by the party's system; The automatic control system continuously records the material transportation drop point probability table at a fixed time and analyzes the material transportation at a fixed time; the material transportation drop point probability table is compiled; and < after the determination: ί 二 Ξ: b ί system Refer to the material drop point of the material to vote for the location of the drop. Including · 2 · According to the method described in item 1 of the scope of patent application, which includes one of the drop points in the drop point allocation table of the drop point of the material, the automatic transfer of the material determines the drop point of the subsequent material 3 · According to the patent application The data is the transportation of each batch of materials 4 · The area of the landing area according to the patent application is the ion implantation area 5 · The area of the delivery of each batch of material is based on the patent application-6 · The area of the landing area is the ion implantation area according to the patent application Record the exchange, _ to update the point allocation table; and the system refers to the updated point allocation table, point position. The method described in item 1, wherein the machine manufacturing capacity of the transportation landing area. The method according to item 3, wherein the transport, the chemical vapor deposition zone, and the yellow light zone are one of. The method described in item 1, wherein the material manufacturing stage of the transportation landing area. The method according to item 5, wherein the transport, the chemical vapor deposition region, and the yellow light region are one of. i〇yj ^ Scope of patent application ^ t According to the method described in item 5 of the scope of patent application, where the material is white and is a specific number of manufacturing steps. Combination 0 times Method, in which the time for the transportation of the material in the domain i11, the landing point area, the landing point force, the manufacturing capacity of the gate, and the next stage manufacturing capability of the machine in the landing point area 9 ^ The manufacturing stage and designation of the machine Any combination of end points. The method described in item 8 of the manufacturing stage, wherein the material is a specific combination of several manufacturing steps. The landing point area is based on the method described in item 8 of the patent scope of V Sheshi, wherein the transport is one of an ion implantation area, a chemical vapor deposition area, and a yellow light area. 0503 -6148TW; t smc2000 -0919; peggy. P t d p. 14