WO2021102902A1 - 在制产品数量上限推荐的系统和方法、计算机可读介质 - Google Patents

在制产品数量上限推荐的系统和方法、计算机可读介质 Download PDF

Info

Publication number
WO2021102902A1
WO2021102902A1 PCT/CN2019/121940 CN2019121940W WO2021102902A1 WO 2021102902 A1 WO2021102902 A1 WO 2021102902A1 CN 2019121940 W CN2019121940 W CN 2019121940W WO 2021102902 A1 WO2021102902 A1 WO 2021102902A1
Authority
WO
WIPO (PCT)
Prior art keywords
time
products
preferred
production
record
Prior art date
Application number
PCT/CN2019/121940
Other languages
English (en)
French (fr)
Inventor
沈国梁
柴栋
吴昊晗
兰天
刘伟赫
Original Assignee
京东方科技集团股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 京东方科技集团股份有限公司 filed Critical 京东方科技集团股份有限公司
Priority to US17/273,177 priority Critical patent/US11703837B2/en
Priority to CN201980002698.1A priority patent/CN113454660A/zh
Priority to EP19945405.9A priority patent/EP4068174A4/en
Priority to PCT/CN2019/121940 priority patent/WO2021102902A1/zh
Publication of WO2021102902A1 publication Critical patent/WO2021102902A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/4183—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by data acquisition, e.g. workpiece identification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Definitions

  • the embodiments of the present disclosure relate to the field of production process control, and in particular, to a system and method for recommending the upper limit of the number of products in production, and a computer-readable medium.
  • the products waiting to be processed and being processed at each process site are collectively referred to as the "work in process (WIP, Work In Process)" of the process site.
  • WIP Work In Process
  • the embodiments of the present disclosure provide a system and method for recommending the upper limit of the number of products in production, and a computer-readable medium.
  • the embodiments of the present disclosure provide a system for recommending the upper limit of the number of products in production, which includes a distributed storage device, an analysis device, and a display device, where:
  • the distributed storage device is configured to store production data generated by factory equipment
  • the analysis device includes one or more processors, and the one or more processors are configured to perform the following operations of determining the upper limit of the number of products in production:
  • the production data includes the quantity record and time-consuming record of the production line in multiple time periods, and the quantity record of each time period is included in the time period
  • the number of products in production at each process site of the production line, and the time-consuming record of each time period includes the process time of each process site of the production line in that time period;
  • each of the quantity records Clustering each of the quantity records to obtain a plurality of initial classifications; wherein, each of the initial classifications includes at least one quantity record;
  • the display device is configured to display the upper limit of the number of products in production of each process site determined by the analysis device.
  • determining that the initial classification of the part as a preferred classification includes:
  • the determining that a part of the initial classification is a preferred classification according to the time-consuming records corresponding to the respective number records of each initial classification includes:
  • the ratio of the classification time of the initial classification to the system time is determined as the time-consuming score of the initial classification; where the classification time of each initial classification is all of the initial classification
  • the average value of the process time in the time-consuming record corresponding to the quantity record, the system time-consuming is all the time-consuming records, after filtering the predetermined proportion of the largest process time and the smallest process time, the remaining process time average value;
  • the determining the upper limit of the number of products in production at each process site according to at least part of the number records of the preferred classification includes:
  • the determining a partial number record as a preferred number record according to the distance between each number record of the preferred classification and the cluster center of the preferred classification where it is located includes:
  • determining that the partial quantity record is a preferred quantity record includes:
  • the determining the proportion coefficient of each process site according to the ratio of the number of products in production of each process site to the total number of products of the production line recorded by each preferred quantity includes:
  • For each of the multiple preferred quantity records determine the ratio of the number of products in production at each process site to the total number of products on the production line as the proportion of the preferred quantity record corresponding to the process site;
  • the determining the upper limit of the number of products in production at each process site according to the current total number of products on the production line and the proportion coefficient of each process site includes:
  • the magnification factor is the predetermined Set a number greater than 1.
  • the cluster is a neighbor propagation cluster.
  • one or more processors included in the analysis device are configured to perform the operation of determining the upper limit of the number of products in production every predetermined time.
  • At least part of the production data stored in the distributed storage device obtained by the analysis device includes:
  • the production line is a display panel production line.
  • the embodiments of the present disclosure provide a method for recommending the upper limit of the number of products in production, including:
  • the time-consuming records corresponding to each quantity record of each initial classification determine part of the initial classification as the preferred classification; wherein, the time-consuming record of each time period includes the process time of each process site of the production line in that time period;
  • determining that the initial classification of the part as a preferred classification includes:
  • the determining that a part of the initial classification is a preferred classification according to the time-consuming records corresponding to the respective number records of each initial classification includes:
  • the ratio of the classification time of the initial classification to the system time is determined as the time-consuming score of the initial classification; where the classification time of each initial classification is all of the initial classification
  • the average value of the process time in the time-consuming record corresponding to the quantity record, the system time-consuming is all the time-consuming records, after filtering the predetermined proportion of the largest process time and the smallest process time, the remaining process time average value;
  • the determining the upper limit of the number of products in production at each process site according to at least part of the number records of the preferred classification includes:
  • the determining a partial number record as a preferred number record according to the distance between each number record of the preferred classification and the cluster center of the preferred classification where it is located includes:
  • determining that the partial quantity record is a preferred quantity record includes:
  • the determining the proportion coefficient of each process site according to the ratio of the number of products in production of each process site to the total number of products of the production line recorded by each preferred quantity includes:
  • For each of the multiple preferred quantity records determine the ratio of the number of products in production at each process site to the total number of products on the production line as the proportion of the preferred quantity record corresponding to the process site;
  • the determining the upper limit of the number of products in production at each process site according to the current total number of products on the production line and the proportion coefficient of each process site includes:
  • the magnification factor is the predetermined Set a number greater than 1.
  • the cluster is a neighbor propagation cluster.
  • the number records of the multiple time periods include:
  • the number of scheduled adjacent time periods before the current time is recorded.
  • the production line is a display panel production line.
  • embodiments of the present disclosure provide a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the above-mentioned method for recommending the upper limit of the number of products in production is implemented.
  • the distributed storage device can efficiently realize the collection and preliminary processing of the raw data of multiple factory equipment through big data, and the analysis device can easily obtain the required data from the distributed storage device.
  • the upper limit of the number of products in production at each process site of the production line is calculated by calculation and displayed by the display device.
  • the embodiments of the present disclosure can automatically recommend the upper limit of the number of products in production for the process sites in each production line, so that users can monitor and schedule the production process according to the upper limit of the number of products in production, for example, when a certain process site
  • timely adjustments can be made (such as reducing the number of products put into the production line, or moving some products to other production lines for processing, etc.) to avoid affecting production capacity and production effectiveness.
  • Fig. 1 is a block diagram of a system for recommending the upper limit of the number of products in production according to an embodiment of the present disclosure
  • FIG. 2 is a block diagram of the analysis device of a system for recommending the upper limit of the number of products in production according to an embodiment of the present disclosure
  • FIG. 3 is a flow chart of the operation performed by the analysis device in the system for recommending the upper limit of the number of manufactured products according to an embodiment of the present disclosure
  • FIG. 4 is a flowchart of the operation performed by the analysis equipment in another system for recommending the upper limit of the number of manufactured products provided by the embodiment of the present disclosure
  • FIG. 5 is a schematic diagram of data flow in a system for recommending the upper limit of the number of products in production according to an embodiment of the present disclosure
  • FIG. 6 is a schematic diagram of the upper limit of the number of products in production obtained by the system for recommending the upper limit of the number of products in production according to an embodiment of the disclosure
  • FIG. 7 is a flowchart of a method for recommending the upper limit of the number of products in production according to an embodiment of the present disclosure
  • FIG. 8 is a flowchart of another method for recommending the upper limit of the number of products in production according to an embodiment of the present disclosure
  • Fig. 9 is a block diagram of the composition of a computer-readable medium provided by an embodiment of the present disclosure.
  • a product such as a display panel
  • the product including semi-finished products
  • Each process site includes one or more process equipment.
  • the process equipment is used to perform certain products on the product. Processing (such as deposition, exposure, etching, inspection, etc.).
  • each process site may have part of the products being processed in it, and some products are waiting to be processed by it.
  • the total quantity of these two types of products It is called the number of products in process (WIP, Work In Process) of the process site.
  • the "number of products” refers to the number of basic counting units of products commonly used in the corresponding technical field.
  • a stack can be used as the basic counting unit, that is, each stack is a product, and each stack usually includes multiple (such as 20) substrates (glass), each The substrate corresponds to a display panel; alternatively, the substrate can also be used as the basic counting unit, that is, each substrate (glass) is used as a product.
  • the product quantity only changes in specific values, and does not affect the implementation of the embodiments of the present disclosure.
  • process site refers to a site used to perform relatively independent processing processes.
  • the specific process sites obtained by division may also be different.
  • the site where a certain main process (such as deposition, exposure, etching, etc.) and the inspection process of the main process are performed can be regarded as one process site, or they can be regarded as two process sites. It should be understood that when the processing divisions are different, only the number of specific process sites and the specific number of products in production at each process site are different, and does not affect the implementation of the embodiments of the present disclosure.
  • cycle time is also called cycle time, which refers to the length of time between when a product enters the process site and leaves the process site for a certain process site, that is, when the product is in the process site. The sum of the length of time the craft station is waiting to be processed and the length of time actually being processed. It should be understood that because the work efficiency of the process site, the number of products waiting to be processed, etc. are different at different times, the process time-consuming of different products processed at the same process site may be different.
  • an embodiment of the present disclosure provides a system for recommending the upper limit of the number of products in production.
  • the system of the embodiment of the present disclosure is used to determine the upper limit of the number of products in production at each process site in the production line, that is, to determine the maximum number of products in production allowed by each process site without causing serious accumulation of products. After determining the upper limit of the number of products in production allowed by each process site, the production process can be controlled according to the upper limit of the number of products in production to avoid serious accumulation of products.
  • the system for recommending the upper limit of the number of products in production includes a distributed storage device, an analysis device, and a display device.
  • Distributed storage devices are configured to store production data generated by factory equipment.
  • the analysis device includes one or more processors, and the one or more processors are configured to perform operations that determine the upper limit of the number of products in production.
  • the display device is configured to display the upper limit of the number of products in production at each process site determined by the analysis device.
  • the distributed storage device stores production data from factory equipment.
  • factory equipment refers to any equipment in each factory, which can include the process equipment in each process site, and can also include the management equipment used to manage the production line in the factory; and the production data refers to any information related to production. Including which products are produced by each production line, the number of products in production at each process site at each time, the cycle time of each product at each process site, product information and bad information of each product, etc.
  • the analysis device includes a processor (such as a CPU) with data processing capabilities, and a memory (such as a hard disk) storing required programs.
  • the processor and the memory are connected through I/O to achieve information interaction.
  • This processor can perform required operations according to the program stored in the memory.
  • the analysis device can extract part of the production data stored in the distributed storage device, and calculate the upper limit of the number of products in production at each process site of the production line (such as a production line) based on the extracted data.
  • the display device has a display function, which is used to display the upper limit of the number of products in production calculated by the analysis device for the user to monitor the production status according to it.
  • the distributed storage device stores relatively complete data (such as a database), and the distributed storage device includes multiple hardware memories, and different hardware memories are distributed in different physical locations (such as in different factories or in Different production lines), and realize the transfer of information between each other through the network, so that the data is distributed, but logically constitute a database based on big data technology.
  • data such as a database
  • the distributed storage device includes multiple hardware memories, and different hardware memories are distributed in different physical locations (such as in different factories or in Different production lines), and realize the transfer of information between each other through the network, so that the data is distributed, but logically constitute a database based on big data technology.
  • the raw data of a large number of different factory equipment are stored in the corresponding manufacturing system, such as YMS (Yield Management System), FDC (Fault Detection & Classification), MES (Manufacturing Execution System, Manufacturing Execution System) and other relational databases (such as Oracle, Mysql, etc.), and these raw data can be extracted from the original table through data extraction tools (such as Sqoop, kettle, etc.) to be transmitted to distributed storage devices (such as Hadoop Distributed).
  • File System, HDFSHadoop Distributed File System, HDFS to reduce the load on factory equipment and manufacturing systems, and facilitate subsequent data reading of analysis equipment.
  • the data in the distributed storage device can be stored in Hive tool or Hbase database format.
  • the above raw data is first stored in the data lake; after that, data cleaning, data conversion, etc. can be preprocessed in the Hive tool according to the application theme and scene of the data to obtain different topics (such as production history Themes, inspection data themes, equipment data themes) data warehouses, and data marts with different scenarios (such as equipment analysis scenarios, parameter analysis scenarios).
  • the above data mart can be connected to display devices, analysis devices, etc. through different API interfaces to realize data interaction with these devices.
  • the data volume of the above raw data is very large.
  • the raw data generated by all factory equipment every day may be several hundred G, and the data generated every hour may also be tens of G.
  • RDBMS relational database management Relational Database Management System
  • DFS distributed File System
  • the grid computing of RDBMS divides the problem that requires very huge computing power into many small parts, and then distributes these parts to many computers for separate processing, and finally combines these calculation results.
  • Oracle RAC Real Application Cluster
  • Oracle RAC is the core technology of grid computing supported by the Oracle database, in which all servers can directly access all data in the database.
  • the RDBMS grid computing application system cannot meet user requirements when the amount of data is large. For example, due to the limited expansion space of the hardware, when the data is increased to a large enough order of magnitude, the input/output bottleneck of the hard disk will cause The efficiency of processing data is very low.
  • the Hive tool is a data warehouse tool based on Hadoop, which can be used for data extraction, transformation and loading (ETL).
  • the Hive tool defines a simple SQL-like query language, and also allows custom MapReduce mappers and reducers to be used by default tools.
  • the Hive tool does not have a special data storage format, nor does it create an index for the data. Users can freely organize the tables in it and process the data in the database. It can be seen that the parallel processing of distributed file management can meet the storage and processing requirements of massive data. Users can process simple data through SQL queries, and use custom functions for complex processing. Therefore, when analyzing the massive data of the factory, it is necessary to extract the data of the factory database into the distributed file system. On the one hand, it will not damage the original data, and on the other hand, it will improve the efficiency of data analysis.
  • the display device may include one or more displays, including one or more terminals with display functions, so that the analysis device can send the upper limit of the number of products in production obtained by its analysis to the display device, and the display device will then display it.
  • the display device can also be used to display an "interactive interface", the interactive interface can include a sub-interface displaying the calculated upper limit of the number of products in production, and is used to control the system for recommending the upper limit of the number of products in production.
  • the user can fully interact with the system recommended for the maximum number of products in production (control and receive results).
  • the distributed storage device can efficiently realize the collection and preliminary processing of the raw data of multiple factory equipment through big data, and the analysis device can easily obtain the required data from the distributed storage device.
  • the upper limit of the number of products in production at each process site of the production line is calculated by calculation and displayed by the display device.
  • the embodiments of the present disclosure can automatically recommend the upper limit of the number of products in production for the process sites in each production line, so that users can monitor and schedule the production process according to the upper limit of the number of products in production, for example, when a certain process site
  • timely adjustments can be made (such as reducing the number of products put into the production line, or moving some products to other production lines for processing, etc.) to avoid affecting production capacity and production effectiveness.
  • the production line is a display panel production line.
  • the embodiments of the present disclosure can be used in the production process of display panels (such as liquid crystal display panels, organic light-emitting diode display panels, etc.) to determine the upper limit of the number of products in production at each process site of the display panel production line.
  • display panels such as liquid crystal display panels, organic light-emitting diode display panels, etc.
  • one or more processors included in the analysis device are configured to perform the operation of determining the upper limit of the number of products in production every predetermined time.
  • the above analysis equipment can calculate the upper limit of the number of products in production periodically (for example, with a period of 1 hour), so that after the upper limit of the number of products in production is calculated each time, the new number of products in production can be calculated next time Between the upper limit, the production process can be monitored and scheduled according to the upper limit of the number of products in production calculated this time.
  • the operation of determining the upper limit of the number of products in production can also be performed in other ways, for example, when the user feels it is necessary to update the upper limit of the number of products in production.
  • the operation of determining the upper limit of the number of products in production performed by one or more processors of the above analysis device may include the following steps:
  • the production data includes the quantity record and time-consuming record of the production line in multiple time periods.
  • the quantity record of each time period includes the number of products in process (WIP) of each process site of the production line in the time period.
  • WIP products in process
  • Each time period The time-consuming record includes the process time-consuming of each process site of the production line during the time period.
  • the analysis device extracts part of the required production data from the above distributed storage device (specifically, the data warehouse) for subsequent calculations.
  • the production data required for each specific calculation (or extraction each time) includes the data for the same production line in multiple different time periods, specifically the number records of multiple time periods (the data of each process site of the production line) The number of products in production) and time-consuming records (the process time-consuming of each process site of the production line).
  • each time period refers to a time period for performing statistics, for example, one hour, and the data (quantity record and time-consuming record) of each time period are obtained by statistics in the time period. Therefore, the quantity recording and the time-consuming recording in different time periods are different.
  • the value of the data in a time period can be the average value of the corresponding data in the time period of the time period. For example, in one hour, for a process site, the instantaneous number of products in production can be counted every 10 minutes, and the average of the number of products in process counted for multiple times is used as the number of products in the process site in the time period. The number of products in production. For another example, in one hour, a certain process site can actually process (process completed) multiple products, and each of these products has a corresponding process time (the process time of different products may be the same or different) , And the average of these process time can be regarded as the process time of the process site in the time period.
  • At least part of the production data stored in the distributed storage device acquired by the analysis device includes: a number record and a time-consuming record of a predetermined number of adjacent time periods before the current time.
  • the data used in the calculation of the upper limit of the number of products in production each time may be the data of all time periods within a predetermined time period before the time point at which the calculation is performed. For example, if you want to calculate the upper limit of the number of products in production at 10 o'clock on a certain day, you can use all the time periods from 10 o'clock on the third day to the current (10 o'clock today) (for example, each time period is 1 hour) data.
  • time period is selected in other ways (such as using multiple discontinuous time periods), it is also feasible.
  • the quantity records and time-consuming records extracted above may also be pre-processed in advance.
  • data preprocessing can specifically include one-hot encoding, data fusion, processing discrete values (such as box-plot method), redundant data deletion, null value processing (such as deletion, filling in, etc.), etc., to eliminate Irregular data to facilitate the use of data in subsequent calculations.
  • the above data preprocessing process may be performed after the analysis device extracts the data, or it may be performed on the data in the data mart by the distributed storage device.
  • each initial classification includes at least one quantity record.
  • each quantity can be recorded as a "point", and all points are clustered according to the spatial position of each point to classify different points (quantity records) into different initial classifications.
  • the clustering is specifically neighbor propagation clustering (AP clustering).
  • Neighbor propagation clustering is also called AP (Affinity Propagation) clustering, which is used to classify multiple points in a multi-dimensional space into multiple categories according to their positions, and each category includes multiple points that are relatively concentrated in a multi-dimensional space.
  • AP Affinity Propagation
  • the logic of the AP clustering algorithm is to treat all points as potential cluster centers, and then analyze the relationship between different points in an iterative manner to find points that are actually suitable as cluster centers and those that are actually suitable for each category. Points to get multiple categories. For example, for any two points i and k to be clustered, the following definitions can be made:
  • the attractiveness matrix is R(i,k), which represents the degree to which k is suitable as the cluster center of i;
  • the attribution degree matrix is A(i,k), which represents the degree to which i is suitable for taking k as the clustering center (or the degree to which i is suitable for belonging to the classification taking k as the clustering center);
  • the similarity matrix is S(i,k), which represents the degree of similarity between i and k;
  • R t+1 (i,k) (1- ⁇ ) ⁇ R t+1 (i,k)+ ⁇ R t (i,k);
  • a t+1 (i,k) (1- ⁇ ) ⁇ A t+1 (i,k)+ ⁇ A t (i,k);
  • the above iterative operation is equivalent to sequentially calculating the suitability of each point as a cluster center, and the suitability of each point belonging to different classifications, so that it can be determined that all points have multiple cluster centers (that is, all points should be divided into How many categories), and which cluster center each point should correspond to (that is, which points are in each category).
  • each quantity record is data in a certain time period, and in this time period, there should also be a time-consuming record. Therefore, the time-consuming record and the quantity record in the same time period correspond to each other. Therefore, the corresponding time-consuming records can be found according to the number records in each initial classification, and based on these time-consuming records, some initial classifications can be determined as preferred classifications for subsequent calculations.
  • the above determining that a part of the initial classification is a preferred classification includes: determining an initial classification as a preferred classification.
  • step S103 only one initial classification can be selected as the preferred classification.
  • the above determining that a part of the initial classification is a preferred classification according to the time-consuming records corresponding to each number record of each initial classification includes:
  • the classification time of each initial classification is the average of the process time in the time-consuming records corresponding to all the quantity records of the initial classification
  • the system time-consuming is the largest process in all the time-consuming records filtered out of the predetermined proportion The average value of the remaining process time after the time-consuming and minimum process time-consuming.
  • ct is the classification time of the initial classification, that is, the average process time (cycle time) corresponding to all process sites in the points (quantity records) belonging to the initial classification; for example, if an initial classification includes 10 There are 10 quantity records, the 10 quantity records belong to 10 time periods, and there are 10 time-consuming records in these 10 time periods. Each time-consuming record includes the process time of 200 process sites. Therefore, the classified consumption Hour ct is the average value of the 10*200 total of 2000 process time-consuming.
  • Ect is the system time, which means that among all the time-consuming records (that is, the acquired time-consuming records of all time periods) of the process time of all process sites (cycle time), the predetermined proportion of the largest process time is removed , And remove the predetermined proportion of the minimum process time, the remaining process time average; for example, if a total of 50 time periods of time-consuming records are obtained, and each time-consuming record includes 200 process sites The process time is 50*200, a total of 10000 process time is counted. If the reservation ratio is 10%, the largest 1000 and the smallest 1000 of the 10000 process time should be removed, and the remaining 8000 processes The average time-consuming Ect is the system time-consuming Ect.
  • the first predetermined initial classification with the largest time-consuming score (such as the largest one) can be selected as the preferred classification, or the time-consuming score can be selected to exceed a predetermined value (such as 0.8, 0.9, etc.)
  • the initial classification of is the preferred classification.
  • S104 Determine the upper limit of the number of products in production at each process site according to at least part of the number records of the preferred classification.
  • step S104 determining the upper limit of the number of products in production at each process site based on at least part of the number records of the preferred classification (step S104) includes:
  • this step (S1041) includes: for each of the preferred categories, determining the number record of the first second predetermined position in the preferred category with the smallest Euclidean distance from the cluster center of the preferred category Or the number whose Euclidean distance is less than the second predetermined value is recorded as the preferred number record.
  • the Euclidean distance between each point (quantity record) and the cluster center can be calculated, and the predetermined number (such as 3) with the smallest Euclidean distance is recorded as the preferred number record, or the Euclidean distance is less than the predetermined number.
  • the number of values is recorded as the preferred number.
  • the above Euclidean distance can be "weighted Euclidean distance".
  • the weighted Euclidean distance in each category is affected by a specific attenuation coefficient. This is because in the process of AP clustering, each point (quantity record ) Is the weighted Euclidean distance from the corresponding cluster center, so the weighted Euclidean distance can be used for evaluation to reduce the amount of calculation.
  • this step (S1041) includes: determining a plurality of quantity records as preferred quantity records.
  • one preferred category can be selected from multiple initial categories (such as the initial category with the largest time-consuming score), and multiple (such as 3) preferred categories can be selected from this preferred category (such as the 3 with the smallest Euclidean distance) Initial classification).
  • the number of products in production at each process site must have a certain proportional relationship with the total number of products in process at all process sites in the preferred quantity record (that is, the total number of products on the production line) ( Percentage factor), so that the upper limit of the number of products in production at each process site can be calculated based on the percentage factor.
  • the proportion coefficient of the process site represents the proportion of the number of products in production at the process site to the number of all products in the production line in a relatively reasonable process flow.
  • this step (S1042) includes: for each of the multiple preferred quantity records, determining the number of products in production at each process site and the products of the production line The ratio of the total quantity is used as the proportion component of the optimal quantity record corresponding to the process site; for each of the multiple process sites, the average value of the proportion component corresponding to each preferential quantity record is determined as the proportion component of the process site Ratio coefficient.
  • the ratio of the number of products in production at each process site to the total number of products of the corresponding production line can be obtained, and the same process
  • the average value of the proportion components of the stations in the multiple preferred quantity records is the proportion coefficient of the process stations.
  • its proportion coefficient SCEa can be calculated by the following formula:
  • Wba is the number of products in production of process site a in the preferred quantity record b
  • Wbs is the total number of products of the production line in the preferred quantity record b
  • S1043. Determine the upper limit of the number of products in production at each process site according to the current total number of products on the production line and the proportion coefficient of each process site.
  • the upper limit of the number of products in production at each process site can be finally determined according to the current total number of products on the production line and the percentage factor of each process site.
  • the current total number of products on the production line means the total number of products that the production line should (or plan to) produce in the current time.
  • the planned output of the production line in the current one hour that is, the number of products planned to be put into the production line
  • the current total product quantity of the production line can be used as the current total product quantity of the production line.
  • this step (S1041) includes: for each of the multiple process sites, determining the proportion of the process site, the product of the current total number of products on the production line, and the magnification factor, as the product of the process site The upper limit of the number of products to be manufactured; among them, the magnification factor is a preset number greater than 1.
  • the percentage factor of each process site can be multiplied by the total number of products that the production line should produce in the current time, and then multiplied by a magnification factor greater than 1, and the result is the number of products in production at the process site Upper limit.
  • the result of multiplying the proportion coefficient of each process site by the current total product quantity of the production line indicates that according to the total number of products currently required to be produced on the production line, the process site preferably has the number of products in production; obviously, the largest product in production
  • the quantity should be greater than the quantity of the above-mentioned preferred products in production, so it should be multiplied by the magnification factor to get the upper limit of the number of products in production at the process site.
  • the amplification factor can be set according to needs, but must be greater than 1, for example, it can take a value between 80 and 120, such as 100.
  • the upper limit of the number of products in process (WIP) of some process sites and the actual number of products in process of some process sites can be referred to Figure 6. It can be seen that the current number of products in production at most process sites in Figure 6 is within the corresponding upper limit of the number of products in production, so corresponding adjustments are needed.
  • an embodiment of the present disclosure provides a method for recommending the upper limit of the number of products in production, including:
  • S201 Perform clustering on the quantity records of the production line in multiple time periods to obtain multiple initial classifications.
  • each initial classification includes at least one quantity record
  • the quantity record for each time period includes the quantity of products in production at each process site of the production line in that time period.
  • S202 Determine a part of the initial classification as a preferred classification according to the time-consuming records corresponding to each quantity record of each initial classification.
  • the time-consuming record of each time period includes the process time of each process site of the production line in that time period.
  • S203 Determine the upper limit of the number of products in production at each process site according to at least part of the number records of the preferred classification.
  • the method of the embodiment of the present disclosure is used to determine the number of products in production at each process site of the production line in a factory, so that the user can monitor and schedule the production process according to the upper limit of the number of products in production.
  • determining that part of the initial classification is a preferred classification includes:
  • determining a part of the initial classification as a preferred classification according to the time-consuming records corresponding to each number record of each initial classification includes:
  • the system time-consuming is all the time-consuming records, after filtering the predetermined proportion of the largest process time and the smallest process time, the remaining process time average value.
  • determining the upper limit of the number of products in production at each process site according to at least part of the number records of the preferred classification includes:
  • S2033. Determine the upper limit of the number of products in production at each process site according to the current total number of products on the production line and the proportion coefficient of each process site.
  • determining the partial number records as the preferred number records according to the distance between each number record of the preferred category and the cluster center of the preferred category where it is located includes:
  • determining that the partial quantity record is the preferred quantity record includes:
  • determining the proportion coefficient of each process site based on the ratio of the number of products in production at each process site to the total number of products on the production line recorded by each preferred quantity includes:
  • For each of the multiple preferred quantity records determine the ratio of the number of products in production at each process site to the total number of products on the production line as the proportion of the preferred quantity record corresponding to the process site;
  • determining the upper limit of the number of products in production at each process site includes:
  • the cluster is a neighbor propagation cluster.
  • the number records of the multiple time periods include:
  • the number of scheduled adjacent time periods before the current time is recorded.
  • the production line is a display panel production line.
  • an embodiment of the present disclosure provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, any one of the above-mentioned methods for recommending the upper limit of the number of products in production is implemented.
  • the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may consist of several physical components. The components are executed cooperatively.
  • Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit (CPU), a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as Application specific integrated circuit.
  • a processor such as a central processing unit (CPU), a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as Application specific integrated circuit.
  • Such software may be distributed on a computer-readable medium, and the computer-readable medium may include a computer storage medium (or a non-transitory medium) and a communication medium (or a transitory medium).
  • the term computer storage medium includes volatile and non-volatile memory implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Sexual, removable and non-removable media.
  • Computer storage media include, but are not limited to, random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), charged erasable programmable read-only memory (EEPROM), flash memory (FLASH) or other disk storage ; CD-ROM, digital versatile disk (DVD) or other optical disk storage; magnetic cassette, tape, disk storage or other magnetic storage; any other that can be used to store desired information and that can be accessed by a computer medium.
  • communication media usually contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as carrier waves or other transmission mechanisms, and may include any information delivery media. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Manufacturing & Machinery (AREA)
  • Quality & Reliability (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • General Factory Administration (AREA)

Abstract

一种在制产品数量上限推荐的系统，包括分布式存储设备、分析设备、显示设备，分布式存储设备的一个或多个处理器被配置为执行以下操作：获取分布式存储设备中存储的至少部分生产数据（S101），生产数据包括生产线在多个时间周期的数量记录和耗时记录，每个时间周期的耗时记录包括在该时间周期生产线的各工艺站点的工艺耗时；对各数量记录进行聚类得到多个初始分类（S102）；每个初始分类包括至少一个数量记录；根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类（S103）；根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限（S104）；显示设备，被配置为显示分析设备确定的各工艺站点的在制产品数量上限。

Description

在制产品数量上限推荐的系统和方法、计算机可读介质 技术领域
本公开实施例涉及生产过程调控领域，特别涉及在制产品数量上限推荐的系统和方法、计算机可读介质。
背景技术
在显示面板(如液晶显示面板、有机发光二极管显示面板等)等产品的生产线中，产品(包括半成品)需依次经过多个工艺站点，每个工艺站点用于进行一定的处理(如沉积、曝光、刻蚀、检测等)。
其中，在每个工艺站点等待处理和正在处理的产品，统称为该工艺站点的“在制产品(WIP,Work In Process)”。
当某个工艺站点的在制产品数量过大时，可能导致产品严重堆积，进而降低产能和生产效率。为此，确定各工艺站点能允许的最大在制产品数量(即在制产品数量上限)，对生产过程调控有重要意义。
发明内容
本公开实施例提供一种在制产品数量上限推荐的系统和方法、计算机可读介质。
第一方面，本公开实施例提供一种在制产品数量上限推荐的系统，包括：分布式存储设备、分析设备、显示设备，其中，
所述分布式存储设备，被配置为存储工厂设备产生的生产数据；
所述分析设备包括一个或多个处理器，所述一个或多个处理器被配置为执行以下确定在制产品数量上限的操作：
获取所述分布式存储设备中存储的至少部分所述生产数据，其中，所述生产数据包括生产线在多个时间周期的数量记录和耗时记录，每个时间周期的数量记录包括在该时间周期生产线的各工艺站点的在制 产品数量，每个时间周期的耗时记录包括在该时间周期生产线的各工艺站点的工艺耗时；
对各所述数量记录进行聚类得到多个初始分类；其中，每个所述初始分类包括至少一个数量记录；
根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类；
根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限；
所述显示设备，被配置为显示所述分析设备确定的所述各工艺站点的在制产品数量上限。
在一些实施例中，所述确定部分初始分类为优选分类包括：
确定一个初始分类为优选分类。
在一些实施例中，所述根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类包括：
对多个初始分类中的每一个，确定该初始分类的分类耗时与系统耗时的比，作为该初始分类的耗时评分；其中，每个初始分类的分类耗时为该初始分类的所有数量记录对应的耗时记录中的工艺耗时的平均值，所述系统耗时为所有耗时记录中，滤除预定比例的最大的工艺耗时和最小的工艺耗时后，剩余工艺耗时的平均值；
确定耗时评分最大的前第一预定位的初始分类或耗时评分大于第一预定值的初始分类为优选分类。
在一些实施例中，所述根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限包括：
根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录；
根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数；
根据生产线当前的产品总数量和各工艺站点的占比系数，确定各 工艺站点的在制产品数量上限。
在一些实施例中，所述根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录包括：
对优选分类中的每一个，确定该优选分类中的、与该优选分类的聚类中心间的欧式距离最小的前第二预定位的数量记录或欧氏距离小于第二预定值的数量记录为优选数量记录。
在一些实施例中，所述确定部分数量记录为优选数量记录包括：
确定多个数量记录为优选数量记录。
在一些实施例中，所述根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数包括：
对多个优选数量记录中的每一个，确定其中每个工艺站点的在制产品数量与生产线的产品总数量的比，作为该工艺站点对应该优选数量记录的占比分量；
对多个工艺站点中的每一个，确定其对应各优选数量记录的占比分量的平均值，作为该工艺站点的占比系数。
在一些实施例中，所述根据生产线当前的产品总数量和各工艺站点的占比系数，确定各工艺站点的在制产品数量上限包括：
对多个工艺站点中的每一个，确定该工艺站点的占比系数、生产线当前的产品总数量、放大系数的乘积，作为该工艺站点的在制产品数量上限；其中，所述放大系数为预设的、大于1的数。
在一些实施例中，所述聚类为近邻传播聚类。
在一些实施例中，所述分析设备包括的一个或多个处理器被配置为每隔预定时间执行所述确定在制产品数量上限的操作。
在一些实施例中，所述分析设备获取的分布式存储设备中存储的至少部分所述生产数据包括：
当前时间前的预定个临近的时间周期的数量记录和耗时记录。
在一些实施例中，所述生产线为显示面板生产线。
第二方面，本公开实施例提供一种在制产品数量上限推荐的方法，包括：
对生产线在多个时间周期的数量记录进行聚类得到多个初始分类；其中，每个所述初始分类包括至少一个数量记录，每个时间周期的数量记录包括在该时间周期生产线的各工艺站点的在制产品数量；
根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类；其中，每个时间周期的耗时记录包括在该时间周期生产线的各工艺站点的工艺耗时；
根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限。
在一些实施例中，所述确定部分初始分类为优选分类包括：
确定一个初始分类为优选分类。
在一些实施例中，所述根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类包括：
对多个初始分类中的每一个，确定该初始分类的分类耗时与系统耗时的比，作为该初始分类的耗时评分；其中，每个初始分类的分类耗时为该初始分类的所有数量记录对应的耗时记录中的工艺耗时的平均值，所述系统耗时为所有耗时记录中，滤除预定比例的最大的工艺耗时和最小的工艺耗时后，剩余工艺耗时的平均值；
确定耗时评分最大的前第一预定位的初始分类或耗时评分大于第一预定值的初始分类为优选分类。
在一些实施例中，所述根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限包括：
根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录；
根据各优选数量记录的各工艺站点的在制产品数量与生产线的产 品总数量的比，确定各工艺站点的占比系数；
根据生产线当前的产品总数量和各工艺站点的占比系数，确定各工艺站点的在制产品数量上限。
在一些实施例中，所述根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录包括：
对优选分类中的每一个，确定该优选分类中的、与该优选分类的聚类中心间的欧式距离最小的前第二预定位的数量记录或欧氏距离小于第二预定值的数量记录为优选数量记录。
在一些实施例中，所述确定部分数量记录为优选数量记录包括：
确定多个数量记录为优选数量记录。
在一些实施例中，所述根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数包括：
对多个优选数量记录中的每一个，确定其中每个工艺站点的在制产品数量与生产线的产品总数量的比，作为该工艺站点对应该优选数量记录的占比分量；
对多个工艺站点中的每一个，确定其对应各优选数量记录的占比分量的平均值，作为该工艺站点的占比系数。
在一些实施例中，所述根据生产线当前的产品总数量和各工艺站点的占比系数，确定各工艺站点的在制产品数量上限包括：
对多个工艺站点中的每一个，确定该工艺站点的占比系数、生产线当前的产品总数量、放大系数的乘积，作为该工艺站点的在制产品数量上限；其中，所述放大系数为预设的、大于1的数。
在一些实施例中，所述聚类为近邻传播聚类。
在一些实施例中，所述对生产线在多个时间周期的数量记录进行聚类得到多个初始分类中，所述多个时间周期的数量记录包括：
当前时间前的预定个临近的时间周期的数量记录。
在一些实施例中，所述生产线为显示面板生产线。
第三方面，本公开实施例提供一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现上述的在制产品数量上限推荐的方法。
本公开实施例中，分布式存储设备可通过大数据方式高效率的实现对多个工厂设备的原始数据的收集和初步处理，分析设备则可从分布式存储设备方便的获取所需的数据，以计算得到生产线的各工艺站点的在制产品数量上限，并供显示设备显示。由此，本公开实施例可自动化的为各生产线中工艺站点推荐在制产品数量上限，以供用户根据该在制产品数量上限对生产过程进行监控和排产，例如，当某个工艺站点的在制产品数量接近、达到或超过其在制产品数量上限时，可及时的进行调整(如降低向生产线投放的产品数量，或将部分产品移出到其它生产线处理等)，以避免影响产能和生产效率。
附图说明
附图用来提供对本公开实施例的进一步理解，并且构成说明书的一部分，与本公开实施例一起用于解释本公开，并不构成对本公开的限制。通过参考附图对详细示例实施例进行描述，以上和其它特征和优点对本领域技术人员将变得更加显而易见，在附图中：
图1本公开实施例提供的一种在制产品数量上限推荐的系统的组成框图；
图2本公开实施例提供的一种在制产品数量上限推荐的系统的分析设备的组成框图；
图3本公开实施例提供的一种在制产品数量上限推荐的系统中分析设备进行的操作的流程图；
图4本公开实施例提供的另一种在制产品数量上限推荐的系统中 分析设备进行的操作的流程图
图5为本公开实施例提供的一种在制产品数量上限推荐的系统中数据流向的示意图；
图6为本公开实施例提供的一种在制产品数量上限推荐的系统得到的部分工艺站点的在制产品数量上限的示意图；
图7本公开实施例提供的一种在制产品数量上限推荐的方法的流程图；
图8本公开实施例提供的另一种在制产品数量上限推荐的方法的流程图；
图9本公开实施例提供的一种计算机可读介质的组成框图。
具体实施方式
为使本领域的技术人员更好地理解本公开实施例的技术方案，下面结合附图对本公开实施例提供的在制产品数量上限推荐的系统和方法、计算机可读介质进行详细描述。
在下文中将参考附图更充分地描述本公开实施例，但是所示的实施例可以以不同形式来体现，且不应当被解释为限于本公开阐述的实施例。反之，提供这些实施例的目的在于使本公开透彻和完整，并将使本领域技术人员充分理解本公开的范围。
本公开实施例可借助本公开的理想示意图而参考平面图和/或截面图进行描述。因此，可根据制造技术和/或容限来修改示例图示。
在不冲突的情况下，本公开各实施例及实施例中的各特征可相互组合。
本公开所使用的术语仅用于描述特定实施例，且不意欲限制本公开。如本公开所使用的术语“和/或”包括一个或多个相关列举条目的任何和所有组合。如本公开所使用的单数形式“一个”和“该”也意欲包括复数形式，除非上下文另外清楚指出。如本公开所使用的术语“包括”、“由……制成”，指定存在所述特征、整体、步骤、操作、 元件和/或组件，但不排除存在或添加一个或多个其它特征、整体、步骤、操作、元件、组件和/或其群组。
除非另外限定，否则本公开所用的所有术语(包括技术和科学术语)的含义与本领域普通技术人员通常理解的含义相同。还将理解，诸如那些在常用字典中限定的那些术语应当被解释为具有与其在相关技术以及本公开的背景下的含义一致的含义，且将不解释为具有理想化或过度形式上的含义，除非本公开明确如此限定。
本公开实施例不限于附图中所示的实施例，而是包括基于制造工艺而形成的配置的修改。因此，附图中例示的区具有示意性属性，并且图中所示区的形状例示了元件的区的具体形状，但并不是旨在限制性的。
在工厂中，通过生产线生产某种产品(如显示面板)时，产品(包括半成品)需依次经过多个工艺站点，每个工艺站点包括一个或多个工艺设备，工艺设备用于对产品进行一定的处理(如沉积、曝光、刻蚀、检测等)。
由于不同工艺站点的处理速度、处理能力不同，故在生产过程中，每个工艺站点都可能有一部分产品正在其中进行处理，还有一部分产品正在等待由其进行处理，这两类产品的总数量即称为该工艺站点的在制产品(WIP,Work In Process)数量。
当某个工艺站点的在制产品数量过大时，可能导致其无法及时完成产品处理，造成产品严重堆积，进而降低产能和生产效率。
本公开实施例中，“产品数量”是指在相应技术领域中，惯用的产品的基本计数单位的数量。例如，对显示面板的生产线而言，可用堆叠(lot)作为基本计数单位，即以每个堆叠为一个产品，而每个堆叠中通常包括多个(如20个)基板(玻璃)，每个基板对应一个显示面板；或者，也可用基板作为基本计数单位，即以每个基板(玻璃)作为一个产品。应当理解，当依据不同的基本计数单位时，产品数量仅仅是具体数值有变化，而并不影响本公开实施例的实现。
本公开实施例中，“工艺站点”是指用于进行相对独立的处理工艺的站点，依照对处理划分的不同，具体划分得到的工艺站点也可有不同。例如，可将进行某个主工艺(如沉积、曝光、刻蚀等)的站点和对该主工艺的检测工艺的站点视为一个工艺站点，也可将它们视为两个工艺站点。应当理解，当处理划分不同时，仅仅是具体的工艺站点数量和每个工艺站点的具体在制产品数量有所不同，而并不影响本公开实施例的实现。
本公开实施例中，“工艺耗时(cycle time)”也称循环耗时，是指对某个工艺站点，产品从进入该工艺站点到离开该工艺站点之间的时长，也就是产品在该工艺站点等待处理的时长和实际被处理的时长的和。应当理解，由于工艺站点的工作效率、等待处理的产品数量等在不同时间是不同的，故同一个工艺站点处理的不同产品的工艺耗时可能不同。
第一方面，参照图1，本公开实施例提供一种在制产品数量上限推荐的系统。
本公开实施例的系统用于确定生产线中的各工艺站点的在制产品数量上限，也就是确定在不会引起产品严重堆积的情况下，各工艺站点允许的最大的在制产品数量。在确定各工艺站点允许的在制产品数量上限后，即可根据该在制产品数量上限对生产过程进行调控，避免引起产品严重堆积。
本公开实施例提供的在制产品数量上限推荐的系统包括分布式存储设备、分析设备、显示设备。
分布式存储设备，被配置为存储工厂设备产生的生产数据。
分析设备包括一个或多个处理器，一个或多个处理器被配置为执行确定在制产品数量上限的操作。
显示设备，被配置为显示分析设备确定的各工艺站点的在制产品数量上限。
分布式存储设备中存储有来自工厂设备的生产数据。其中，工厂设备是指各工厂中的任何设备，其可包括各工艺站点中的工艺设备，也可包括工厂中用于管理生产线的管理设备等；而生产数据是指与生产相关的任何信息，包括各生产线生产了哪些产品，各工艺站点在各时间的在制产品数量，各产品在各工艺站点的工艺耗时(cycle time)，各产品的产品信息、不良信息等。
参照图2，分析设备包括具有数据处理能力的处理器(如CPU)，还可具有存储有所需程序的存储器(如硬盘)，处理器与存储器通过I/O连接从而能实现信息交互，由此处理器可根据存储器中存储的程序进行所需运算。本公开实施例中，分析设备能提取分布式存储设备中存储的部分生产数据，并根据提取的数据计算得到生产线(如一条生产线)的各工艺站点的在制产品数量上限。
显示设备具有显示功能，用于将分析设备计算得到的在制产品数量上限显示出来，供用户根据其监控生产状况。
其中，分布式存储设备中存储有相对完整的数据(如一个数据库)，而且，分布式存储设备包括多个硬件的存储器，且不同的硬件存储器分布在不同物理位置(如在不同工厂，或在不同生产线)，并通过网络实现相互之间信息的传递，从而其数据是分布式关系的，但在逻辑上构成一个基于大数据技术的数据库。
参照图5，大量不同工厂设备的原始数据存储在相应的生产制造系统中，如YMS(Yield Management System，收益管理系统)、FDC(Fault Detection&Classification，错误侦测及分类)、MES(Manufacturing Execution System，制造执行系统)等系统的关系型数据库(如Oracle、Mysql等)中，而这些原始数据可通过数据抽取工具(如Sqoop、kettle等)进行原表抽取以传输给分布式存储设备(如Hadoop Distributed File System，HDFSHadoop Distributed File System，HDFS)，以降低对工厂设备和生产制造系统的负载，便于后续分析设备的数据读取。
分布式存储设备中的数据可采用Hive工具或Hbase数据库格式存 储。例如，根据Hive工具，以上原始数据先存储在数据湖中；之后，可继续在Hive工具中按照数据的应用主题、场景等进行数据清洗、数据转换等预处理，得到具有不同主题(如生产履历主题、检测数据主题、设备数据主题)的数据仓库，以及具有不同场景(如设备分析场景、参数分析场景)的数据集市。以上数据集市可再通过不同的API接口，与显示设备、分析设备等连接，以实现与这些设备间的数据交互。
其中，由于涉及多个工厂的多个工厂设备，故以上原始数据的数据量是很大的。例如，所有工厂设备每天产生的原始数据可能有几百G，每小时产生的数据也可能有几十G。
对海量结构化数据实现存储与计算主要有两种方案：RDBMS关系型数据库管理(Relational Database Management System,RDBMS)的网格计算方案；分布式文件管理系统(Distributed File System,DFS)的大数据方案。
其中，RDBMS的网格计算是把需要非常巨大的计算能力的问题分成许多小部分，然后把这些部分分配给许多计算机分别处理，最后把这些计算结果综合起来。例如，作为一种具体例子，Oracle RAC(真正应用集群)是Oracle数据库支持的网格计算的核心技术，其中所有服务器都可直接访问数据库中的所有数据。但是，RDBMS的网格计算的应用系统在数据量很大时无法满足用户要求，例如，由于硬件的扩展空间有限，故数据增加到足够大的数量级后，会因为硬盘的输入/输出的瓶颈使得处理数据的效率非常低。
分布式文件管理为基础的大数据技术，则允许采用多个廉价硬件设备构建大型集群，以对海量数据进行处理。如Hive工具是基于Hadoop的数据仓库工具，可用来进行数据提取转化加载(ETL)，Hive工具定义了简单的类SQL查询语言，同时也允许通过自定义的MapReduce的mapper和reducer来默认工具无法完成的复杂的分析工作。Hive工具没有专门的数据存储格式，也没有为数据建立索引，用户可以自由的组织其中的表，对数据库中的数据进行处理。可见，分布式文件管理的并行处理可满足海量数据的存储和处理要求，用户可通过SQL查询处理简单 数据，而复杂处理时可采用自定义函数来实现。因此，在对工厂的海量数据分析时，需要将工厂数据库的数据抽取到分布式文件系统中，一方面不会对原始数据造成破坏，另一方面提高了数据分析效率。
其中，显示设备可包括一个或多个显示器，包括一个或多个具有显示功能的终端，从而分析设备可将其分析得到的在制产品数量上限发送给显示设备，显示设备再将其显示出来。
在一些实施例中，显示设备还可用于显示“交互界面”，该交互界面可包括显示计算得到的在制产品数量上限的子界面，用于控制该在制产品数量上限推荐的系统进行所需工作(如任务设定)的子界面，以及显示当前各工艺站点的实际在制产品数量的子界面等。
也就是说，通过该显示设备的“交互界面”，可实现用户与在制产品数量上限推荐的系统的完全交互(控制和接收结果)。
本公开实施例中，分布式存储设备可通过大数据方式高效率的实现对多个工厂设备的原始数据的收集和初步处理，分析设备则可从分布式存储设备方便的获取所需的数据，以计算得到生产线的各工艺站点的在制产品数量上限，并供显示设备显示。由此，本公开实施例可自动化的为各生产线中工艺站点推荐在制产品数量上限，以供用户根据该在制产品数量上限对生产过程进行监控和排产，例如，当某个工艺站点的在制产品数量接近、达到或超过其在制产品数量上限时，可及时的进行调整(如降低向生产线投放的产品数量，或将部分产品移出到其它生产线处理等)，以避免影响产能和生产效率。
在一些实施例中，生产线为显示面板生产线。
本公开实施例可用于在显示面板(如液晶显示面板、有机发光二极管显示面板等)的生产过程中，确定显示面板生产线的各工艺站点的在制产品数量上限。
当然，本公开实施例也可用于其它产品的生产线。
在一些实施例中，分析设备包括的一个或多个处理器被配置为每隔预定时间执行确定在制产品数量上限的操作。
也就是说，以上分析设备可周期性(如以1个小时为周期)计算在制产品数量上限，从而在每次计算得到在制产品数量上限之后，到下一次计算出新的在制产品数量上限之间，可根据本次计算得到的在制产品数量上限对生产过程进行监控和排产。
当然，确定在制产品数量上限的操作也可按照其它方式进行，例如在用户觉得有必要更新在制产品数量上限时进行。
参照图3，在一些实施例中，以上分析设备的一个或多个处理器执行的确定在制产品数量上限的操作可包括以下步骤：
S101、获取分布式存储设备中存储的至少部分生产数据。
其中，生产数据包括生产线在多个时间周期的数量记录和耗时记录，每个时间周期的数量记录包括在该时间周期内生产线的各工艺站点的在制产品(WIP)数量，每个时间周期的耗时记录包括在该时间周期内生产线的各工艺站点的工艺耗时。
也就是说，分析设备从以上分布式存储设备中(具体是数据仓库中)提取部分所需的生产数据，以用于进行后续计算。其中，每次具体计算所需(或者说每次提取)的生产数据，包括针对同一条生产线的、在多个不同时间周期的数据，具体是多个时间周期的数量记录(生产线各工艺站点的在制产品数量)和耗时记录(生产线各工艺站点的工艺耗时)。
其中，每个时间周期是指一个用于进行统计的时间段，例如是1个小时，而每个时间周期的数据(数量记录和耗时记录)，都是在该时间周期内统计得到的。由此，不同时间周期内的数量记录和耗时记录是不同的。
一个时间周期的数据的值，可为在该时间周期的时间段内，相应数 据的平均值。例如，在1个小时中，对某个工艺站点，可每隔10分钟统计一次其瞬时的在制产品数量，并以多次统计的在制产品数量的平均作为该工艺站点在该时间周期的在制产品数量。再如，在1个小时中，某个工艺站点可总计实际处理(处理完成)了多个产品，而这些产品每个都具有相应的工艺耗时(不同产品的工艺耗时可能相同或不同)，而这些工艺耗时的平均可作为该工艺站点在该时间周期的工艺耗时。
当然，如果采用时间周期内的数据的最大值、最小值、最接近某个时间点的值等其它数值作为该时间周期的数据的值，也是可行的。
在一些实施例中，分析设备获取的分布式存储设备中存储的至少部分生产数据包括：当前时间前的预定个临近的时间周期的数量记录和耗时记录。
也就是说，每次在制产品数量上限的计算中所用的数据，可以是在进行计算的时间点之前，预定时间段内的所有时间周期的数据。例如，若某天10点要计算在制产品数量上限，则可采用之前第三天的10点开始，到当前(今天10点)为止的时间段内的全部时间周期(如每个时间周期为1小时)的数据。
当然，如果以上时间周期的选取采用其它方式(如采用多个不连续的时间周期)，也是可行的。
在一些实施例中，以上提取出的数量记录和耗时记录，还可预先经过数据预处理。
其中，数据预处理(数据清洗)具体可包括独热编码、数据融合、处理离散值(如箱型图法)、冗余数据删除、空值处理(如删除、补缺等)等，用于消除不规范的数据，以利于数据用于后续计算。
具体的，以上数据预处理过程，可以是在分析设备提取数据后进行，或者也可由分布式存储设备对其数据集市中的数据进行。
S102、对各数量记录进行聚类得到多个初始分类。
其中，每个初始分类包括至少一个数量记录。
如前，每个数量记录包括生产线中的多个(n个)工艺站点的在制产品数量，故其相当于一个n维空间的“点”，该点的每一维的坐标为一个工艺站点的在制产品数量。例如，若生产线包括200个工艺站点(n＝200)，则每个数量记录包括200个工艺站点的在制产品数量，对应一个200维空间中的点。
由此，可以每个数量记录为一个“点”，根据各点的空间位置，对所有的点进行聚类，以将不同的点(数量记录)分入不同的初始分类中，每类中有一个或多个位置相对接近的数量记录。
可选的，聚类具体为近邻传播聚类(AP聚类)。
近邻传播聚类也称AP(Affinity Propagation)聚类，其用于根据多维空间中多个点的位置将它们分为多个分类，每个分类包括多个在多维空间中相对集中分布的点。
AP聚类算法的逻辑是将所有的点都看成潜在的聚类中心，再通过迭代的方式分析不同点间的关系，找到实际适于作为聚类中心的点和实际适于属于各分类的点，从而得到多个分类。例如，对于任意两个待进行聚类的点i和k，可进行如下定义：
吸引度矩阵为R(i,k)，其表示k适于作为i的聚类中心的程度；
归属度矩阵为A(i,k)，其表示i适于以k作为聚类中心的程度(或者说i适于属于以k为聚类中心的分类的程度)；
相似度矩阵为S(i,k)，其表示i与k之间的相似程度；
之后，可通过以下一组公式进行迭代运算：
R t+1(i,k)＝(1-λ)·R t+1(i,k)+λ·R t(i,k)；
Figure PCTCN2019121940-appb-000001
A t+1(i,k)＝(1-λ)·A t+1(i,k)+λ·A t(i,k)；
Figure PCTCN2019121940-appb-000002
以上迭代运算，相当于依次计算各点作为聚类中心的合适程度，以及各点属于不同分类的合适程度，从而可确定出所有的点中，有多个聚类中心(即所有点应分为多少个分类)，以及每个点应对应哪个聚类中心(即每个分类中有哪些点)。
S103、根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类。
在对数量记录进行聚类后，每个初始分类中有多个数量记录。显然，每个数量记录是在某个时间周期的数据，而在该时间周期，还应具有一个耗时记录，由此，同一个时间周期的耗时记录和数量记录，就是相互对应的。因此，可根据每个初始分类中的数量记录，找到对应的耗时记录，并根据这些耗时记录，确定确定出部分初始分类为优选分类，用于后续计算。
在一些实施例中，以上确定部分初始分类为优选分类(步骤S103)包括：确定一个初始分类为优选分类。
为简化计算过程，以上S103步骤中，可仅选取一个初始分类作为优选分类。
当然，应当理解，本步骤中选取出多个优选分类，也是可行的。
在一些实施例中，参照图4，以上根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类(步骤S103)包括：
S1031、对多个初始分类中的每一个，确定该初始分类的分类耗时与系统耗时的比，作为该初始分类的耗时评分。
其中，每个初始分类的分类耗时为该初始分类的所有数量记录对应的耗时记录中的工艺耗时的平均值，系统耗时为所有耗时记录中，滤除预定比例的最大的工艺耗时和最小的工艺耗时后，剩余工艺耗时的平均 值。
为确定优选分类，首先计算每个初始分类的耗时评分，每个初始分类的耗时评分sc通过以下公式计算：
Figure PCTCN2019121940-appb-000003
其中，ct为该初始分类的分类耗时，也就是所有属于该初始分类的点(数量记录)中的工艺站点对应的工艺耗时(cycle time)的平均值；例如，若某初始分类包括10个数量记录，则该10个数量记录分别属于10个时间周期，而这10个时间周期又有10个耗时记录，每个耗时记录包括200个工艺站点的工艺耗时，因此，分类耗时ct为该10*200共2000个工艺耗时的平均值。
其中，Ect为系统耗时，表示在所有的耗时记录(即获取的所有时间周期的耗时记录)的所有工艺站点的工艺耗时(cycle time)中，去掉预定比例的最大的工艺耗时，并去除预定比例的最小的工艺耗时后，剩下的工艺耗时的平均值；例如，若共获取了50个时间周期的耗时记录，且每个耗时记录包括200个工艺站点的工艺耗时，则计有50*200共10000个工艺耗时，若预订比例取10％，则应将10000个工艺耗时中最大的1000个和最小的1000个去掉，以剩余的8000个工艺耗时的平均值为系统耗时Ect。
S1032、确定耗时评分最大的前第一预定位的初始分类或耗时评分大于第一预定值的初始分类为优选分类。
在得到每个初始分类的耗时评分后，可选取耗时评分最大的前预定个(如最大的1个)初始分类为优选分类，或选取耗时评分超过预定值(如0.8、0.9等)的初始分类为优选分类。
S104、根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限。
在得到优选分类后，继续从优选分类中选出部分数量记录作为优选数量记录，并根据这些优选数量记录得出各工艺站点的在制产品数量上 限。
在一些实施例中，参照图4，以上根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限(S104步骤)包括：
S1041、根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录。
为确定在制产品数量上限，首先根据数量记录与其所在的优选分类的聚类中心间的距离，从优选分类的多个数量记录中选出部分作为优选数量记录用于后续计算。
在一些实施例中，本步骤(S1041)包括：对优选分类中的每一个，确定该优选分类中的、与该优选分类的聚类中心间的欧式距离最小的前第二预定位的数量记录或欧氏距离小于第二预定值的数量记录为优选数量记录。
对每个对优选分类，可计算其中各点(数量记录)与聚类中心的欧式距离，并以欧式距离最小的预定个(如3个)数量记录作为优选数量记录，或者以欧式距离小于预定值的数量记录作为优选数量记录。
其中，以上欧式距离可为“加权欧式距离”，每个分类中的加权欧式距离都受特定的衰减系数的影响，这是因为在AP聚类的过程中，通常可得出各点(数量记录)与相应聚类中心的加权欧式距离，故用加权欧式距离进行评价可降低运算量。
当然，应当理解，由于本步骤中进行欧式距离比较的多个数量记录必然属于一个优选分类，故其是否“加权”对最终选出的优选数量记录并无影响。
在一些实施例中，本步骤(S1041)包括：确定多个数量记录为优选数量记录。
为降低偶然因素的影响，本步骤中可选出多个优选数量记录用于后续计算。例如，可从多个初始分类中选出一个优选分类(如耗时评分最大的初始分类)，并从该优选分类中选出多个(如3个)优选分类(如欧式距离最小的3个初始分类)。
当然，以上的多个优选数量记录，也可以是从多个优选分类中得到的。
其中，对于作为聚类中心的“点(数量记录)”，其是否可作为优选数量记录，可以根据需要设定，在此不再详细描述。
S1042、根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数。
对每个优选数量记录，其中每个工艺站点的在制产品数量，与该优选数量记录中所有工艺站点的在制产品的总数量(即生产线的产品总数量)必然有一个确定的比例关系(占比系数)，从而可根据该占比系数计算各工艺站点的在制产品数量上限。
其中，工艺站点的占比系数表示在相对合理的工艺流程中，工艺站点的在制产品数量占生产线中所有产品的数量的比例。
在一些实施例中，当优选数量记录的数量为多个时，本步骤(S1042)包括：对多个优选数量记录中的每一个，确定其中每个工艺站点的在制产品数量与生产线的产品总数量的比，作为该工艺站点对应该优选数量记录的占比分量；对多个工艺站点中的每一个，确定其对应各优选数量记录的占比分量的平均值，作为该工艺站点的占比系数。
当优选数量记录的数量为多个时，对每个优选数量记录，可求出其中每个工艺站点的在制产品数量与相应生产线的产品总数量的比(占比分量)，而同一个工艺站点在多个优选数量记录中的占比分量的平均值，即为该工艺站点的占比系数。例如，对任意工艺站点a，其占比系数SCEa可通过以下公式计算：
Figure PCTCN2019121940-appb-000004
其中，Wba为优选数量记录b中工艺站点a的在制产品数量，Wbs为优选数量记录b中生产线的产品总数量，m为优选数量记录的总数量(如m＝3)。
S1043、根据生产线当前的产品总数量和各工艺站点的占比系数，确 定各工艺站点的在制产品数量上限。
在确定以上占比系数后，可根据生产线当前的产品总数量和各工艺站点的占比系数，最终确定各工艺站点的在制产品数量上限。
其中，“生产线当前的产品总数量”表示，在当前时间内生产线应当(或者说计划)生产的产品的总数量。例如，若以1个小时为间隔进行本公开实施例的方法，则可用生产线的在当前1个小时内的计划产量(即计划投入生产线的产品的数量)作为该生产线当前的产品总数量。
在一些实施例中，本步骤(S1041)包括：对多个工艺站点中的每一个，确定该工艺站点的占比系数、生产线当前的产品总数量、放大系数的乘积，作为该工艺站点的在制产品数量上限；其中，放大系数为预设的、大于1的数。
也就是说，可用每个工艺站点的占比系数，乘以当前时间内生产线应当生产的产品的总数量，再乘以一个大于1的放大系数，以所得结果为该工艺站点的在制产品数量上限。
其中，每个工艺站点的占比系数乘以生产线当前的产品总数量的结果，表示根据生产线当前需要生产的产品总数量，该工艺站点优选应有在制产品数量；显然，最大的在制产品数量应大于以上优选在制产品数量，故还应将其乘以放大系数，才能得到该工艺站点的在制产品数量上限。
其中，放大系数可根据需要设定，但必然是大于1的，例如其可取80～120之间的值，例如为100。
例如，某次计算过程中，得到的部分工艺站点的在制产品(WIP)数量上限，以及部分工艺站点当前实际的在制产品数量可参照图6。可见，图6中多数工艺站点的当前在制产品数量，均在对应的在制产品数量上限，故需要进行相应的调整。
第二方面，参照图7，本公开实施例提供一种在制产品数量上限推荐的方法，包括：
S201、对生产线在多个时间周期的数量记录进行聚类得到多个初始分类。
其中，每个初始分类包括至少一个数量记录，每个时间周期的数量记录包括在该时间周期生产线的各工艺站点的在制产品数量。
S202、根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类。
其中，每个时间周期的耗时记录包括在该时间周期生产线的各工艺站点的工艺耗时。
S203、根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限。
本公开实施例的方法用于确定工厂中生产线的各工艺站点的在制产品数量上，以供用户根据该在制产品数量上限对生产过程进行监控和排产。
在一些实施例中，确定部分初始分类为优选分类包括：
确定一个初始分类为优选分类。
参照图8，在一些实施例中，根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类(步骤S202)包括：
S2021、对多个初始分类中的每一个，确定该初始分类的分类耗时与系统耗时的比，作为该初始分类的耗时评分；其中，每个初始分类的分类耗时为该初始分类的所有数量记录对应的耗时记录中的工艺耗时的平均值，系统耗时为所有耗时记录中，滤除预定比例的最大的工艺耗时和最小的工艺耗时后，剩余工艺耗时的平均值。
S2022、确定耗时评分最大的前第一预定位的初始分类或耗时评分大于第一预定值的初始分类为优选分类。
参照图8，在一些实施例中，根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限(步骤S203)包括：
S2031、根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录。
S2032、根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数。
S2033、根据生产线当前的产品总数量和各工艺站点的占比系数，确定各工艺站点的在制产品数量上限。
在一些实施例中，根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录包括：
对优选分类中的每一个，确定该优选分类中的、与该优选分类的聚类中心间的欧式距离最小的前第二预定位的数量记录或欧氏距离小于第二预定值的数量记录为优选数量记录。
在一些实施例中，确定部分数量记录为优选数量记录包括：
确定多个数量记录为优选数量记录。
在一些实施例中，根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数包括：
对多个优选数量记录中的每一个，确定其中每个工艺站点的在制产品数量与生产线的产品总数量的比，作为该工艺站点对应该优选数量记录的占比分量；
对多个工艺站点中的每一个，确定其对应各优选数量记录的占比分量的平均值，作为该工艺站点的占比系数。
在一些实施例中，根据生产线当前的产品总数量和各工艺站点的占比系数，确定各工艺站点的在制产品数量上限包括：
对多个工艺站点中的每一个，确定该工艺站点的占比系数、生产线当前的产品总数量、放大系数的乘积，作为该工艺站点的在制产品数量上限；其中，放大系数为预设的、大于1的数。
在一些实施例中，聚类为近邻传播聚类。
在一些实施例中，对生产线在多个时间周期的数量记录进行聚类得到多个初始分类中，多个时间周期的数量记录包括：
当前时间前的预定个临近的时间周期的数量记录。
在一些实施例中，生产线为显示面板生产线。
第三方面，参照图9，本公开实施例提供一种计算机可读介质，其上存储有计算机程序，程序被处理器执行时实现上述任意一种在制产品数量上限推荐的方法。
本领域普通技术人员可以理解，上文中所公开的全部或某些步骤、系统、装置中的功能模块/单元可以被实施为软件、固件、硬件及其适当的组合。
在硬件实施方式中，在以上描述中提及的功能模块/单元之间的划分不一定对应于物理组件的划分；例如，一个物理组件可以具有多个功能，或者一个功能或步骤可以由若干物理组件合作执行。
某些物理组件或所有物理组件可以被实施为由处理器，如中央处理器(CPU)、数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、程序模块或其它数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于随机存取存储器(RAM，更具体如SDRAM、DDR等)、只读存储器(ROM)、带电可擦可编程只读存储器(EEPROM)、闪存(FLASH)或其它磁盘存储器；只读光盘(CD-ROM)、数字多功能盘(DVD)或其它光盘存储器；磁盒、磁带、磁盘存储或其它磁存储器；可以用于存储期望的信息并且可以被计算机访问的任何其它的介质。此外，本领域普通技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其它传输机制之类的调制数据信号中的其它数据，并且可包括任何信息递送介质。
本公开已经公开了示例实施例，并且虽然采用了具体术语，但它们 仅用于并仅应当被解释为一般说明性含义，并且不用于限制的目的。在一些实例中，对本领域技术人员显而易见的是，除非另外明确指出，否则可单独使用与特定实施例相结合描述的特征、特性和/或元素，或可与其它实施例相结合描述的特征、特性和/或元件组合使用。因此，本领域技术人员将理解，在不脱离由所附的权利要求阐明的本公开的范围的情况下，可进行各种形式和细节上的改变。

Claims (20)

  1. 一种在制产品数量上限推荐的系统，包括：分布式存储设备、分析设备、显示设备，其中，
    所述分布式存储设备，被配置为存储工厂设备产生的生产数据；
    所述分析设备包括一个或多个处理器，所述一个或多个处理器被配置为执行以下确定在制产品数量上限的操作：
    获取所述分布式存储设备中存储的至少部分所述生产数据，其中，所述生产数据包括生产线在多个时间周期的数量记录和耗时记录，每个时间周期的数量记录包括在该时间周期生产线的各工艺站点的在制产品数量，每个时间周期的耗时记录包括在该时间周期生产线的各工艺站点的工艺耗时；
    对各所述数量记录进行聚类得到多个初始分类；其中，每个所述初始分类包括至少一个数量记录；
    根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类；
    根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限；
    所述显示设备，被配置为显示所述分析设备确定的所述各工艺站点的在制产品数量上限。
  2. 根据权利要求1所述的系统，其中，所述确定部分初始分类为优选分类包括：
    确定一个初始分类为优选分类。
  3. 根据权利要求1所述的系统，其中，所述根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类包括：
    对多个初始分类中的每一个，确定该初始分类的分类耗时与系统耗时的比，作为该初始分类的耗时评分；其中，每个初始分类的分类耗时为该初始分类的所有数量记录对应的耗时记录中的工艺耗时的平均值，所述系统耗时为所有耗时记录中，滤除预定比例的最大的工艺耗时和最小的工艺耗时后，剩余工艺耗时的平均值；
    确定耗时评分最大的前第一预定位的初始分类或耗时评分大于第一预定值的初始分类为优选分类。
  4. 根据权利要求1所述的系统，其中，所述根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限包括：
    根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录；
    根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数；
    根据生产线当前的产品总数量和各工艺站点的占比系数，确定各工艺站点的在制产品数量上限。
  5. 根据权利要求4所述的系统，其中，所述根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录包括：
    对优选分类中的每一个，确定该优选分类中的、与该优选分类的聚类中心间的欧式距离最小的前第二预定位的数量记录或欧氏距离小于第二预定值的数量记录为优选数量记录。
  6. 根据权利要求4所述的系统，其中，所述确定部分数量记录为优选数量记录包括：
    确定多个数量记录为优选数量记录。
  7. 根据权利要求6所述的系统，其中，所述根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数包括：
    对多个优选数量记录中的每一个，确定其中每个工艺站点的在制产品数量与生产线的产品总数量的比，作为该工艺站点对应该优选数量记录的占比分量；
    对多个工艺站点中的每一个，确定其对应各优选数量记录的占比分量的平均值，作为该工艺站点的占比系数。
  8. 根据权利要求4所述的系统，其中，所述根据生产线当前的产品总数量和各工艺站点的占比系数，确定各工艺站点的在制产品数量上限包括：
    对多个工艺站点中的每一个，确定该工艺站点的占比系数、生产线当前的产品总数量、放大系数的乘积，作为该工艺站点的在制产品数量上限；其中，所述放大系数为预设的、大于1的数。
  9. 根据权利要求1所述的系统，其中，
    所述聚类为近邻传播聚类。
  10. 根据权利要求1所述的系统，其中，
    所述分析设备包括的一个或多个处理器被配置为每隔预定时间执行所述确定在制产品数量上限的操作。
  11. 根据权利要求1所述的系统，其中，所述分析设备获取的分布式存储设备中存储的至少部分所述生产数据包括：
    当前时间前的预定个临近的时间周期的数量记录和耗时记录。
  12. 根据权利要求1所述的系统，其中，
    所述生产线为显示面板生产线。
  13. 一种在制产品数量上限推荐的方法，包括：
    对生产线在多个时间周期的数量记录进行聚类得到多个初始分类；其中，每个所述初始分类包括至少一个数量记录，每个时间周期的数量记录包括在该时间周期生产线的各工艺站点的在制产品数量；
    根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类；其中，每个时间周期的耗时记录包括在该时间周期生产线的各工艺站点的工艺耗时；
    根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限。
  14. 根据权利要求13所述的方法，其中，所述根据每个初始分类的各数量记录对应的耗时记录，确定部分初始分类为优选分类包括：
    对多个初始分类中的每一个，确定该初始分类的分类耗时与系统耗时的比，作为该初始分类的耗时评分；其中，每个初始分类的分类耗时为该初始分类的所有数量记录对应的耗时记录中的工艺耗时的平均值，所述系统耗时为所有耗时记录中，滤除预定比例的最大的工艺耗时和最小的工艺耗时后，剩余工艺耗时的平均值；
    确定耗时评分最大的前第一预定位的初始分类或耗时评分大于第一预定值的初始分类为优选分类。
  15. 根据权利要求13所述的方法，其中，所述根据优选分类的至少部分数量记录，确定各工艺站点的在制产品数量上限包括：
    根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录；
    根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数；
    根据生产线当前的产品总数量和各工艺站点的占比系数，确定各工艺站点的在制产品数量上限。
  16. 根据权利要求15所述的方法，其中，所述根据优选分类的各数量记录与其所在优选分类的聚类中心间的距离，确定部分数量记录为优选数量记录包括：
    对优选分类中的每一个，确定该优选分类中的、与该优选分类的聚类中心间的欧式距离最小的前第二预定位的数量记录或欧氏距离小于第二预定值的数量记录为优选数量记录。
  17. 根据权利要求15所述的方法，其中，所述确定部分数量记录为优选数量记录包括：
    确定多个数量记录为优选数量记录。
  18. 根据权利要求17所述的方法，其中，所述根据各优选数量记录的各工艺站点的在制产品数量与生产线的产品总数量的比，确定各工艺站点的占比系数包括：
    对多个优选数量记录中的每一个，确定其中每个工艺站点的在制产品数量与生产线的产品总数量的比，作为该工艺站点对应该优选数量记录的占比分量；
    对多个工艺站点中的每一个，确定其对应各优选数量记录的占比分量的平均值，作为该工艺站点的占比系数。
  19. 根据权利要求13所述的方法，其中，
    所述聚类为近邻传播聚类。
  20. 一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现根据权利要求13至19中任意一项所述的在制产品数量上限推荐的方法。
PCT/CN2019/121940 2019-11-29 2019-11-29 在制产品数量上限推荐的系统和方法、计算机可读介质 WO2021102902A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US17/273,177 US11703837B2 (en) 2019-11-29 2019-11-29 System and method for recommending maximum quantity of work in process, and computer readable medium
CN201980002698.1A CN113454660A (zh) 2019-11-29 2019-11-29 在制产品数量上限推荐的系统和方法、计算机可读介质
EP19945405.9A EP4068174A4 (en) 2019-11-29 2019-11-29 SYSTEM TO RECOMMEND A MAXIMUM QUANTITY OF PRODUCTS IN A METHOD, PROCEDURE AND COMPUTER READABLE MEDIA
PCT/CN2019/121940 WO2021102902A1 (zh) 2019-11-29 2019-11-29 在制产品数量上限推荐的系统和方法、计算机可读介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/121940 WO2021102902A1 (zh) 2019-11-29 2019-11-29 在制产品数量上限推荐的系统和方法、计算机可读介质

Publications (1)

Publication Number Publication Date
WO2021102902A1 true WO2021102902A1 (zh) 2021-06-03

Family

ID=76129867

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/121940 WO2021102902A1 (zh) 2019-11-29 2019-11-29 在制产品数量上限推荐的系统和方法、计算机可读介质

Country Status (4)

Country Link
US (1) US11703837B2 (zh)
EP (1) EP4068174A4 (zh)
CN (1) CN113454660A (zh)
WO (1) WO2021102902A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070129978A1 (en) * 2005-11-09 2007-06-07 Yoshinori Shirasu Production plan apparatus
CN101329746A (zh) * 2007-06-22 2008-12-24 鸿富锦精密工业（深圳）有限公司 生产计划自动排程系统及方法
CN101655715A (zh) * 2009-09-25 2010-02-24 金蝶软件（中国）有限公司 一种自动分配生产线任务的方法和装置
CN101950387A (zh) * 2010-09-08 2011-01-19 合肥工业大学 一种机械产品装配过程实时物料配送方法
CN106997514A (zh) * 2017-04-12 2017-08-01 杭州纳戒科技有限公司 生产数据管理方法及系统
CN109165851A (zh) * 2018-08-28 2019-01-08 武汉易普优科技有限公司 一种用于智能制造的滚动计划排程方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5255197A (en) * 1990-07-06 1993-10-19 Honda Giken Kogyo Kabushiki Kaisha Line production management system
US5838565A (en) * 1997-05-15 1998-11-17 Vanguard International Semiconductor Corporation Manufacturing control method for IC plant batch sequential machine
US6473721B1 (en) * 1998-10-26 2002-10-29 Sony Corporation Factory traffic monitoring and analysis apparatus and method
CN102622667B (zh) * 2012-02-13 2016-08-03 浪潮通用软件有限公司 一种基于多产品多生产线生产模式下的均衡排产方法
KR20130124208A (ko) * 2012-05-04 2013-11-13 주식회사 썬스타 봉제 생산 공정 컨설팅 장치와 방법 및 컴퓨터 판독 가능한 기록 매체
CN105631554B (zh) * 2016-02-22 2019-11-26 渤海大学 一种基于时间序列的油井油液含水率多模型预测方法
US11619927B2 (en) 2017-11-03 2023-04-04 Drishti Technologies, Inc. Automatic analysis of real time conditions in an activity space
CN112825161A (zh) * 2019-11-21 2021-05-21 鸿富锦精密电子（成都）有限公司 产品良率预测方法、装置、计算机装置及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070129978A1 (en) * 2005-11-09 2007-06-07 Yoshinori Shirasu Production plan apparatus
CN101329746A (zh) * 2007-06-22 2008-12-24 鸿富锦精密工业（深圳）有限公司 生产计划自动排程系统及方法
CN101655715A (zh) * 2009-09-25 2010-02-24 金蝶软件（中国）有限公司 一种自动分配生产线任务的方法和装置
CN101950387A (zh) * 2010-09-08 2011-01-19 合肥工业大学 一种机械产品装配过程实时物料配送方法
CN106997514A (zh) * 2017-04-12 2017-08-01 杭州纳戒科技有限公司 生产数据管理方法及系统
CN109165851A (zh) * 2018-08-28 2019-01-08 武汉易普优科技有限公司 一种用于智能制造的滚动计划排程方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4068174A4 *

Also Published As

Publication number Publication date
US20220113710A1 (en) 2022-04-14
US11703837B2 (en) 2023-07-18
EP4068174A4 (en) 2022-11-16
CN113454660A (zh) 2021-09-28
EP4068174A1 (en) 2022-10-05

Similar Documents

Publication Publication Date Title
KR102627690B1 (ko) Ｓｑｌ 질의 플랜들을 최적화하기 위한 차원 콘텍스트 전파 기술들
US8650181B2 (en) OLAP execution model using relational operations
US20220284014A1 (en) Deriving metrics from queries
US11822545B2 (en) Search integration
US10409650B2 (en) Efficient access scheduling for super scaled stream processing systems
US20230259525A1 (en) Low-Latency Predictive Database Analysis
US9767427B2 (en) Modeling multi-dimensional sequence data over streams
US9892178B2 (en) Systems and methods for interest-driven business intelligence systems including event-oriented data
US10540363B2 (en) Systems and methods for providing performance metadata in interest-driven business intelligence systems
US10824614B2 (en) Custom query parameters in a database system
JP6434154B2 (ja) トランザクションアクセスパターンに基づいた結合関係の識別
US20210364999A1 (en) System and method for analyzing cause of product defect, computer readable medium
US20220358563A1 (en) System and method for a real-time egocentric collaborative filter on large datasets
US9600559B2 (en) Data processing for database aggregation operation
WO2015041731A1 (en) Interest-driven business intelligence systems including segment data
CN107515899A (zh) 数据库联合分片方法、装置以及存储介质
CN107729500B (zh) 一种联机分析处理的数据处理方法、装置及后台设备
WO2021142622A1 (zh) 确定不良原因的方法、电子设备、存储介质及系统
US11734286B2 (en) Automatic database insight analysis
CN102902614A (zh) 一种动态监测及智能引导方法
WO2021102902A1 (zh) 在制产品数量上限推荐的系统和方法、计算机可读介质
CN111143328A (zh) 一种敏捷商业智能数据构建方法、系统、设备、存储介质
Salem et al. A Cloud-Based Data Integration Framework for E-Government Service.
Siddesh et al. Driving big data with hadoop technologies
WO2021168783A1 (zh) 基于非线性规划模型的投产规划系统，投产规划方法和计算机可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19945405

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019945405

Country of ref document: EP

Effective date: 20220629