Big data-based production efficiency diagnosis method, device, equipment and storage medium

Abstract

This invention relates to the field of big data technology, and discloses a production efficiency diagnosis method, apparatus, equipment, and storage medium based on big data, applicable to third-generation compound semiconductors. The production efficiency diagnosis method based on big data includes: acquiring diagnostic big data related to the production of third-generation compound semiconductors; constructing multiple performance evaluation indicators for evaluating production efficiency based on the diagnostic big data, wherein the performance evaluation indicators are used to evaluate product quality, product production cycle, resource input, and production output; and inputting each of the performance evaluation indicators into a preset production cycle model and a productivity model for production efficiency diagnosis. This invention realizes the use of big data to diagnose production efficiency, enabling factories to improve production efficiency and reduce product production cycles with limited resources, thereby increasing value for customers.

Description

Technical Field

[0001] This invention relates to the field of big data technology, and in particular to a method, apparatus, equipment and storage medium for diagnosing production efficiency based on big data. Background Technology

[0002] Production efficiency impacts a company's market share. Only by improving production efficiency and reducing costs can a company withstand price competition and avoid being replaced by domestic and international competitors. Currently, traditional methods for improving production efficiency include increasing manpower, adding equipment, improving machine availability, relaxing process restrictions, and delaying maintenance to shorten production cycle time. However, the effects are limited because they don't address the root causes. For example, adding equipment doesn't eliminate work-in-process inventory; or focusing solely on average machine availability can lead to variations in efficiency, with night shift availability consistently higher than day shift availability, causing bottleneck machines to be more prone to idleness. Traditional methods fail to monitor and accurately identify factors affecting production efficiency in real time, thus failing to address the root causes of production efficiency problems in a timely manner. Summary of the Invention

[0003] The main objective of this invention is to provide a production efficiency diagnosis method, apparatus, equipment, and storage medium based on big data, aiming to solve the technical problem that production efficiency is difficult to quantify due to the numerous and complex factors influencing it.

[0004] The first aspect of this invention provides a production efficiency diagnosis method based on big data, comprising:

[0005] Acquire diagnostic big data related to the production of third-generation compound semiconductors;

[0006] Based on the diagnostic big data, multiple performance evaluation indicators are constructed to evaluate production efficiency. These performance evaluation indicators are used to evaluate product quality, product production cycle, resource input and production output.

[0007] The performance evaluation indicators are input into the pre-set production cycle model and productivity model to diagnose production efficiency. The production cycle model is used to evaluate the relationship between product quality and product production cycle, and the productivity model is used to evaluate the relationship between resource input and production output.

[0008] Optionally, in a first implementation of the first aspect of the present invention, before acquiring diagnostic big data related to the production of third-generation compound semiconductors, the method further includes:

[0009] Acquire diagnostic sample data related to the production of third-generation compound semiconductors;

[0010] Based on the diagnostic sample data, multiple performance evaluation indicators are constructed to evaluate production efficiency.

[0011] A production efficiency evaluation model is constructed using a pre-defined production cycle function and a productivity function, and then trained using the performance evaluation indicators to obtain the corresponding production cycle model and productivity model.

[0012] Optionally, in a second implementation of the first aspect of the present invention, the performance evaluation indicators include product quality evaluation indicators, product production cycle evaluation indicators, resource input evaluation indicators, and production output evaluation indicators. The step of inputting each of the performance evaluation indicators into a pre-set production cycle model and productivity model for production efficiency diagnosis includes:

[0013] The product quality evaluation index and the product production cycle evaluation index are input into a preset production cycle model, and a genetic algorithm is used to diagnose the production efficiency of the production cycle model.

[0014] The resource input evaluation index and the production output evaluation index are input into a pre-set productivity model, and a genetic algorithm is used to diagnose the production efficiency of the productivity model.

[0015] Optionally, in a third implementation of the first aspect of the present invention, the product quality evaluation indicators include the number of qualified products and the total number of products, and the product production cycle evaluation includes the process production time and the equipment production time.

[0016] The step of inputting the product quality evaluation index and the product production cycle evaluation index into a preset production cycle model, and using a genetic algorithm to diagnose the production efficiency of the production cycle model, includes:

[0017] The qualified product quantity, the total number of products, the process production time, and the equipment production time are input into a preset production cycle model, and the production cycle model is subjected to multiple initial replication operations to obtain an initial combination of data for quantifying the production cycle using multiple indicators.

[0018] The initial combination of data quantified by each of the aforementioned indicators is input into the production cycle model, and multiple feasible cross operations are performed on the production cycle model to obtain multiple feasible combinations of data quantified by the production cycle for each indicator.

[0019] The feasible combination of data for quantifying the production cycle based on each of the aforementioned indicators is input into the production cycle model, and multiple target mutation operations are performed on the production cycle model to obtain the first optimal combination of data for quantifying the production cycle based on multiple indicators.

[0020] Optionally, in a fourth implementation of the first aspect of the present invention, the step of inputting the resource input evaluation index and the production output evaluation index into a preset productivity model, and using a genetic algorithm to diagnose the production efficiency of the productivity model, includes:

[0021] The number of devices, the capital investment, the number of workers and the output rate are input into the productivity model, and the productivity model is subjected to multiple initial replication operations to obtain an initial combination of multiple indicators for quantifying productivity.

[0022] The initial combination of the data for quantifying productivity of each of the aforementioned indicators is input into the productivity model, and multiple feasible cross operations are performed on the productivity model to obtain multiple feasible combinations of data for quantifying productivity of the indicators.

[0023] The feasible combinations of data for quantifying productivity based on each of the aforementioned indicators are input into the productivity model, and multiple target mutation operations are performed on the productivity model to obtain the second optimal combination of data for quantifying productivity based on multiple indicators.

[0024] Optionally, in a fifth implementation of the first aspect of the present invention, after inputting each of the performance evaluation indicators into a preset production cycle model and productivity model for production efficiency diagnosis, the method further includes:

[0025] Hypothesis analysis is performed on the first and second optimal data combinations to obtain multiple third optimal data combinations with different quantifications and priorities.

[0026] A Pareto analysis is performed on the third optimal data combination to obtain the fourth optimal data combination with priority quantification for each group for user reference.

[0027] Optionally, in a sixth implementation of the first aspect of the present invention, the diagnostic big data includes: machine status data, batch transaction data, and product management data.

[0028] A second aspect of the present invention provides a production efficiency diagnostic device based on big data, comprising:

[0029] The data acquisition module is used to acquire diagnostic big data related to the production of third-generation compound semiconductors, wherein the diagnostic big data includes: machine status data, batch transaction data and product management data;

[0030] The indicator construction module is used to construct multiple performance evaluation indicators for evaluating production efficiency based on the diagnostic big data. These performance evaluation indicators are used to evaluate product quality, product production cycle, resource input and production output.

[0031] The efficiency diagnosis module is used to input the performance evaluation indicators into a preset production cycle model and a productivity model to diagnose production efficiency. The production cycle model is used to evaluate the relationship between product quality and product production cycle, and the productivity model is used to evaluate the relationship between resource input and production output.

[0032] Optionally, in a first implementation of the second aspect of the present invention, the big data-based production efficiency diagnostic device further includes:

[0033] The model building module is used to acquire diagnostic sample data related to the production of third-generation compound semiconductors; based on the diagnostic sample data, multiple performance evaluation indicators for evaluating production efficiency are constructed; corresponding production efficiency evaluation models are constructed using preset production cycle functions and productivity functions, and trained using each of the performance evaluation indicators to obtain the corresponding production cycle model and productivity model.

[0034] Optionally, in a second implementation of the second aspect of the present invention, the efficiency diagnosis module includes:

[0035] The first efficiency diagnosis unit is used to input the product quality evaluation index and the product production cycle evaluation index into a preset production cycle model, and to use a genetic algorithm to diagnose the production efficiency of the production cycle model.

[0036] The second efficiency diagnosis unit is used to input the resource input evaluation index and the production output evaluation index into the preset productivity model, and to use a genetic algorithm to diagnose the production efficiency of the productivity model.

[0037] Optionally, in a third implementation of the second aspect of the present invention, the first efficiency diagnosis unit is specifically used for:

[0038] The qualified product quantity, the total number of products, the process production time, and the equipment production time are input into a preset production cycle model, and the production cycle model is subjected to multiple initial replication operations to obtain an initial combination of data for quantifying the production cycle using multiple indicators.

[0039] The initial combination of data quantified by each of the aforementioned indicators is input into the production cycle model, and multiple feasible cross operations are performed on the production cycle model to obtain multiple feasible combinations of data quantified by the production cycle for each indicator.

[0040] The feasible combination of data for quantifying the production cycle based on each of the aforementioned indicators is input into the production cycle model, and multiple target mutation operations are performed on the production cycle model to obtain the first optimal combination of data for quantifying the production cycle based on multiple indicators.

[0041] Optionally, in a fourth implementation of the second aspect of the present invention, the second efficiency diagnosis unit is specifically used for:

[0042] The number of devices, the capital investment, the number of workers and the output rate are input into the productivity model, and the productivity model is subjected to multiple initial replication operations to obtain an initial combination of multiple indicators for quantifying productivity.

[0043] The initial combination of the data for quantifying productivity of each of the aforementioned indicators is input into the productivity model, and multiple feasible cross operations are performed on the productivity model to obtain multiple feasible combinations of data for quantifying productivity of the indicators.

[0044] The feasible combinations of data for quantifying productivity based on each of the aforementioned indicators are input into the productivity model, and multiple target mutation operations are performed on the productivity model to obtain the second optimal combination of data for quantifying productivity based on multiple indicators.

[0045] Optionally, in a fifth implementation of the second aspect of the present invention, the big data-based production efficiency diagnostic device further includes:

[0046] The efficiency analysis module is used to perform hypothesis analysis on the first and second optimal data combinations, respectively obtaining multiple third optimal data combinations with different quantifications and priorities; and to perform Pareto analysis on the third optimal data combinations to obtain each group of fourth optimal data combinations with priority quantifications for user reference.

[0047] Optionally, in a sixth implementation of the second aspect of the present invention, the diagnostic big data includes: machine status data, batch transaction data, and product management data.

[0048] A third aspect of the present invention provides an electronic device, the electronic device comprising: a memory and at least one processor, wherein the memory stores instructions;

[0049] The at least one processor invokes the instructions in the memory to cause the electronic device to execute the above-described big data-based production efficiency diagnosis method.

[0050] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described production efficiency diagnosis method based on big data.

[0051] In the technical solution provided by this invention, users only need to provide machine status data, batch order data, and finished product management data. The system automatically constructs multiple performance evaluation indicators for assessing production efficiency. Then, based on these performance evaluation indicators, it establishes production cycle models and productivity models respectively. A genetic algorithm is then used to analyze the quantitative impact of each key performance indicator on the production cycle and productivity. This invention improves the accuracy of enterprises' diagnosis of production efficiency problems and, through hypothesis analysis to quantify the impact and Pareto analysis, obtains improvement plans for key factors affecting production efficiency. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of an embodiment of the production efficiency diagnosis method based on big data in this invention.

[0053] Figure 2 This is a schematic diagram of another embodiment of the production efficiency diagnosis method based on big data in this invention.

[0054] Figure 3 This is a flowchart illustrating the production efficiency diagnosis method based on big data in an embodiment of the present invention.

[0055] Figure 4 This is a schematic diagram of one embodiment of the device in this invention;

[0056] Figure 5 This is a schematic diagram of one embodiment of the electronic device in this invention. Detailed Implementation

[0057] This invention provides a production efficiency diagnosis method, apparatus, equipment, and storage medium based on big data. This invention improves the accuracy of enterprises' diagnosis of production efficiency problems and obtains improvement solutions for key factors affecting production efficiency through hypothesis analysis to quantify the impact and Pareto analysis.

[0058] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0059] To facilitate understanding of the present invention, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the production efficiency diagnosis method based on big data in this invention includes:

[0060] 101. Obtain diagnostic big data related to the production of third-generation compound semiconductors;

[0061] To improve production efficiency, a factory must first identify the factors affecting it and then develop improvement plans based on these factors. In this embodiment, the production efficiency diagnostic system and the manufacturing execution system need to...

[0062] To achieve production efficiency diagnostics for third-generation compound semiconductors, machine status data, batch posting data, and product management data are collected by the manufacturing execution system and uploaded to the production efficiency diagnostic system.

[0063] 102. Based on the diagnostic big data, construct multiple performance evaluation indicators for evaluating production efficiency. These performance evaluation indicators are used to evaluate product quality, product production cycle, resource input, and production output.

[0064] In this embodiment, the production efficiency diagnostic system extracts no fewer than 30 key performance indicators for efficiency improvement from diagnostic big data related to third-generation compound semiconductor production.

[0065] In this embodiment, extracting no fewer than 30 key performance indicators for efficiency improvement helps enterprises prioritize addressing key issues that impact improvement, thereby fundamentally improving production efficiency.

[0066] 103. Input the performance evaluation indicators into the preset production cycle model and productivity model to diagnose production efficiency. The production cycle model is used to evaluate the relationship between product quality and product production cycle, and the productivity model is used to evaluate the relationship between resource input and production output.

[0067] In this embodiment, the production efficiency diagnostic system establishes production cycle and productivity models based on multiple key performance indicators (KPIs) for efficiency improvement, respectively, to analyze the quantitative impact of KPIs on production cycle and productivity using a genetic algorithm. The company then submits the quantitative impact results of the KPIs on production cycle and productivity to professionals for hypothesis analysis to determine the improvement priorities of the indicators. Finally, Pareto analysis is used to obtain the optimal solution for improving the main efficiency-related issues.

[0068] In this embodiment, agile manufacturing production management goals are achieved through production efficiency diagnosis and networking of the Manufacturing Execution System (MES) and equipment. Real-time data visualization via dashboards enables accurate understanding of resource load status for various production processes, improving bottleneck resource utilization, enhancing the accuracy of raw material and tooling delivery, and improving production responsiveness. Visualized workshop management improves efficiency by making raw material preparation, production planning information, and processing transparent and real-time, improving delivery accuracy, monitoring production equipment operation, ensuring timely production and delivery, and increasing efficiency. Visualized equipment information enhances equipment informatization, improves efficiency, reduces defect rates, and increases user satisfaction. Visualized product lifecycle analysis analyzes process bottlenecks, effectively promotes production coordination, reduces production execution error rates, and improves departmental collaboration efficiency. Visualized process content strengthens employee autonomy, improves data sharing efficiency, and increases production line responsiveness. Visualized anomaly information allows for rapid response and effective resolution, avoiding line downtime and reducing risks. Transparent and real-time inventory information avoids material entry and exit errors caused by information transmission delays, reducing warehousing costs.

[0069] Please see Figure 2 , Figure 2 Another embodiment of the production efficiency diagnosis method based on big data of the present invention includes:

[0070] 201. Obtain diagnostic sample data related to the production of third-generation compound semiconductors;

[0071] 202. Based on the diagnostic sample data, construct multiple performance evaluation indicators for evaluating production efficiency;

[0072] In this embodiment, the Manufacturing Execution System (MES) connects to the local area network of the production workshop. It collects order priority data, delivery date data, inventory data, processing path data, product characteristic data, processing procedure data, equipment load data, resource constraint data, equipment fault diagnosis data, equipment maintenance data, and equipment operation statistics related to third-generation compound semiconductor production from the software on the network server as diagnostic sample data. Based on this diagnostic sample data, at least 30 key performance indicators (KPIs) are extracted. Irrelevant factors are further eliminated to identify key efficiency-influencing factors, thus clarifying the direction for efficiency improvement.

[0073] 203. Construct corresponding production efficiency evaluation models using pre-set production cycle functions and productivity functions, and train them using the performance evaluation indicators to obtain the corresponding production cycle model and productivity model.

[0074] In this embodiment, a production cycle efficiency evaluation model is established based on the production cycle function, and a production efficiency evaluation model is established based on the productivity function. By training the aforementioned performance evaluation indicators, the production cycle model and the productivity model are obtained.

[0075] 204. Obtain diagnostic big data related to the production of third-generation compound semiconductors;

[0076] 205. Based on the diagnostic big data, construct multiple performance evaluation indicators for evaluating production efficiency, wherein the performance evaluation indicators are used to evaluate product quality, product production cycle, resource input and production output;

[0077] In this embodiment, the descriptions of steps 204 and 205 above refer to the first embodiment, and will not be repeated here. The following is a detailed description of the production efficiency diagnosis process.

[0078] 206. Input the product quality evaluation index and the product production cycle evaluation index into the preset production cycle model, and use a genetic algorithm to diagnose the production efficiency of the production cycle model;

[0079] Optionally, step 206 above includes:

[0080] The qualified product quantity, the total number of products, the process production time, and the equipment production time are input into a preset production cycle model, and the production cycle model is subjected to multiple initial replication operations to obtain an initial combination of data for quantifying the production cycle using multiple indicators.

[0081] The initial combination of data quantified by each of the aforementioned indicators is input into the production cycle model, and multiple feasible cross operations are performed on the production cycle model to obtain multiple feasible combinations of data quantified by the production cycle for each indicator.

[0082] The feasible combination of data for quantifying the production cycle based on each of the aforementioned indicators is input into the production cycle model, and multiple target mutation operations are performed on the production cycle model to obtain the first optimal combination of data for quantifying the production cycle based on multiple indicators.

[0083] Optionally, after step 206 above, the following steps are also included:

[0084] Hypothesis analysis is performed on the first and second optimal data combinations to obtain multiple third optimal data combinations with different quantifications and priorities.

[0085] A Pareto analysis is performed on the third optimal data combination to obtain the fourth optimal data combination with priority quantification for each group for user reference.

[0086] In this embodiment, the product quality evaluation indicators include the number of qualified products. Factors affecting product qualification can include operator factors, material factors, process method factors, measurement factors, and environmental factors. The product production cycle evaluation indicators include the total number of products, the production time of the process, and the production time of the equipment. Based on the performance evaluation indicators, an initial population of the performance evaluation indicators is constructed, wherein the initial population contains multiple individuals, and each individual is a performance evaluation indicator. Based on a preset fitness function, the fitness of individuals in the population is calculated to obtain the fitness value. Based on the fitness value, multiple individuals with lower fitness are selected and replicated to obtain a new population. Based on the new population, individuals are randomly selected with a preset crossover probability to perform crossover operations to obtain a feasible population. Based on the feasible population, individuals are randomly selected with a preset mutation probability to perform mutation operations to obtain a target population. Based on the iteration rule, if multiple iterations do not generate a new population, the first optimal combination of multiple indicators for quantifying the production cycle is obtained.

[0087] In this embodiment, an intelligent optimization solution developed using a genetic algorithm as its engine offers fast operation and relatively low cost. The production efficiency diagnostic system specifically considers the factory's unique production constraints and adjusts its operating patterns accordingly. Through optimization calculations performed by the genetic algorithm engine, a convergent near-optimal solution can be obtained in a short time. This allows enterprises to quickly identify key factors affecting production efficiency and obtain the optimal solution through Pareto analysis. It enables accurate monitoring of resource load conditions for various production processes, improves bottleneck resource utilization, and enhances production responsiveness.

[0088] 207. Input the resource input evaluation index and the production output evaluation index into the preset productivity model, and use a genetic algorithm to diagnose the production efficiency of the productivity model.

[0089] Optionally, step 207 above includes:

[0090] The step of inputting the resource input evaluation indicators and the production output evaluation indicators into a pre-set productivity model, and then using a genetic algorithm to diagnose the production efficiency of the productivity model, includes:

[0091] The number of devices, the capital investment, the number of workers and the output rate are input into the productivity model, and the productivity model is subjected to multiple initial replication operations to obtain an initial combination of multiple indicators for quantifying productivity.

[0092] The initial combination of the data for quantifying productivity of each of the aforementioned indicators is input into the productivity model, and multiple feasible cross operations are performed on the productivity model to obtain multiple feasible combinations of data for quantifying productivity of the indicators.

[0093] The feasible combinations of data for quantifying productivity based on each of the aforementioned indicators are input into the productivity model, and multiple target mutation operations are performed on the productivity model to obtain the second optimal combination of data for quantifying productivity based on multiple indicators.

[0094] Optionally, after step 207 above, the following steps are also included:

[0095] Hypothesis analysis is performed on the first and second optimal data combinations to obtain multiple third optimal data combinations with different quantifications and priorities.

[0096] A Pareto analysis is performed on the third optimal data combination to obtain the fourth optimal data combination with priority quantification for each group for user reference.

[0097] In this embodiment, the resource input evaluation indicators include the number of equipment, capital investment, and number of production workers, and the production output evaluation indicators include the output rate. Based on the performance evaluation indicators, an initial population of the performance evaluation indicators is constructed, wherein the initial population contains multiple individuals, each of which represents a performance evaluation indicator. Based on a preset fitness function, the fitness of individuals in the population is calculated to obtain fitness values. Based on the fitness values, individuals with lower fitness are selected from the multiple individuals and a replication operation is performed to obtain a new population. Based on the new population, individuals are randomly selected with a preset crossover probability to perform a crossover operation to obtain a feasible population. Based on the feasible population, individuals are randomly selected with a preset mutation probability to perform a mutation operation to obtain a target population. Based on the iteration rule, if multiple iterations do not generate a new population, an initial combination of data for quantifying productivity using multiple indicators is obtained.

[0098] In this embodiment, an intelligent optimization solution developed using a genetic algorithm as its engine offers fast operation and relatively low cost. The production efficiency diagnostic system specifically considers the factory's unique production constraints and adjusts its operating patterns accordingly. Through optimization calculations performed by the genetic algorithm engine, a convergent near-optimal solution can be obtained in a short time. This allows enterprises to quickly identify key factors affecting production efficiency and obtain the optimal solution through Pareto analysis. It enables accurate monitoring of resource load conditions for various production processes, improves bottleneck resource utilization, and enhances production responsiveness.

[0099] Please see Figure 3 , Figure 3This invention provides a production efficiency diagnosis method based on big data: First, production-related data is collected through a manufacturing execution system to obtain diagnostic big data, and data extraction is performed on the diagnostic big data to obtain no less than 30 evaluation indicators. A production cycle model and a productivity model are constructed using these evaluation indicators. Simultaneously, the quantitative impact of the evaluation indicators on the production cycle and productivity is analyzed using a genetic algorithm involving replication, crossover, and mutation. This yields a first optimal combination of indicators for quantifying the production cycle and a second optimal combination of indicators for quantifying productivity. These first and second optimal combinations of indicators are then given to professionals for hypothesis analysis to obtain multiple third optimal combinations of indicators with different quantifications and priorities. Pareto analysis is performed on these third optimal combinations to obtain a fourth optimal combination of indicators with priority quantification. Multiple Pareto analyses are conducted to obtain multiple optimal combinations of data.

[0100] In this embodiment, the production management goal of agile manufacturing is achieved through production efficiency diagnosis and networking of the manufacturing execution system and equipment. Information visualization is achieved by displaying data in real time through a dashboard, which can accurately grasp the resource load status of various production processes, improve the utilization rate of bottleneck resources, and enhance production responsiveness.

[0101] The above describes the production efficiency diagnosis method based on big data in the embodiments of the present invention. The following describes the production efficiency diagnosis device based on big data in the embodiments of the present invention. Please refer to [link / reference]. Figure 4 The first embodiment of the production efficiency diagnostic device based on big data in this invention includes:

[0102] Data acquisition module 301 is used to acquire diagnostic big data related to the production of third-generation compound semiconductors;

[0103] The indicator construction module 302 is used to construct multiple performance evaluation indicators for evaluating production efficiency based on the diagnostic big data. The performance evaluation indicators are used to evaluate product quality, product production cycle, resource input and production output.

[0104] The efficiency diagnosis module 303 is used to input the performance evaluation indicators into a preset production cycle model and a productivity model to diagnose production efficiency. The production cycle model is used to evaluate the relationship between product quality and product production cycle, and the productivity model is used to evaluate the relationship between resource input and production output.

[0105] Optionally, in one embodiment, the big data-based production efficiency diagnostic device further includes:

[0106] The model building module is used to acquire diagnostic sample data related to the production of third-generation compound semiconductors; based on the diagnostic sample data, multiple performance evaluation indicators for evaluating production efficiency are constructed; corresponding production efficiency evaluation models are constructed using preset production cycle functions and productivity functions, and trained using each of the performance evaluation indicators to obtain the corresponding production cycle model and productivity model.

[0107] Optionally, in one embodiment, the efficiency diagnosis module includes:

[0108] The first efficiency diagnosis unit is used to input the product quality evaluation index and the product production cycle evaluation index into a preset production cycle model, and to use a genetic algorithm to diagnose the production efficiency of the production cycle model.

[0109] The second efficiency diagnosis unit is used to input the resource input evaluation index and the production output evaluation index into the preset productivity model, and to use a genetic algorithm to diagnose the production efficiency of the productivity model.

[0110] Optionally, in one embodiment, the first efficiency diagnostic unit is specifically used for:

[0111] The qualified product quantity, the total number of products, the process production time, and the equipment production time are input into a preset production cycle model, and the production cycle model is subjected to multiple initial replication operations to obtain an initial combination of data for quantifying the production cycle using multiple indicators.

[0112] The initial combination of data quantified by each of the aforementioned indicators is input into the production cycle model, and multiple feasible cross operations are performed on the production cycle model to obtain multiple feasible combinations of data quantified by the production cycle for each indicator.

[0113] The feasible combination of data for quantifying the production cycle based on each of the aforementioned indicators is input into the production cycle model, and multiple target mutation operations are performed on the production cycle model to obtain the first optimal combination of data for quantifying the production cycle based on multiple indicators.

[0114] Optionally, in one embodiment, the second efficiency diagnostic unit is specifically used for:

[0115] The number of devices, the capital investment, the number of workers and the output rate are input into the productivity model, and the productivity model is subjected to multiple initial replication operations to obtain an initial combination of multiple indicators for quantifying productivity.

[0116] The initial combination of the data for quantifying productivity of each of the aforementioned indicators is input into the productivity model, and multiple feasible cross operations are performed on the productivity model to obtain multiple feasible combinations of data for quantifying productivity of the indicators.

[0117] The feasible combinations of data for quantifying productivity based on each of the aforementioned indicators are input into the productivity model, and multiple target mutation operations are performed on the productivity model to obtain the second optimal combination of data for quantifying productivity based on multiple indicators.

[0118] Optionally, in one embodiment, the big data-based production efficiency diagnostic device further includes:

[0119] The efficiency analysis module is used to perform hypothesis analysis on the first and second optimal data combinations, respectively obtaining multiple third optimal data combinations with different quantifications and priorities; and to perform Pareto analysis on the third optimal data combinations to obtain each group of fourth optimal data combinations with priority quantifications for user reference.

[0120] Optionally, in one embodiment, the diagnostic big data includes: machine status data, batch transaction data, and product management data.

[0121] In this embodiment, users only need to provide machine status data, batch order data, and finished product management data. The system automatically constructs multiple performance evaluation indicators for assessing production efficiency. Then, based on these performance evaluation indicators, it establishes production cycle models and productivity models respectively. A genetic algorithm is then used to analyze the quantitative impact of each key performance indicator on the production cycle and productivity. This invention improves the accuracy of enterprises' diagnosis of production efficiency problems and, through hypothesis analysis to quantify the impact and Pareto analysis, obtains improvement plans for key factors affecting production efficiency.

[0122] above Figure 4 The production efficiency diagnostic device based on big data in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The electronic device in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0123] Figure 5This is a schematic diagram of the structure of an electronic device 500 provided in an embodiment of the present invention. The electronic device 500 can vary significantly due to differences in configuration or performance, and may include one or more central processing units (CPUs) 510 (e.g., one or more processors) and a memory 520, and one or more storage media 530 (e.g., one or more mass storage devices) for storing application programs 533 or data 532. The memory 520 and storage media 530 can be temporary or persistent storage. The program stored in the storage media 530 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the electronic device 500. Furthermore, the processor 510 may be configured to communicate with the storage media 530 and execute the series of instruction operations in the storage media 530 on the electronic device 500.

[0124] Electronic device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input / output interfaces 560, and / or one or more operating systems 531, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 4 The illustrated electronic device structure does not constitute a limitation on the electronic device and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0125] The present invention also provides an electronic device, which includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the production efficiency diagnosis method based on big data in the above embodiments.

[0126] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of a big data-based production efficiency diagnosis method.

[0127] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0128] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0129] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A production efficiency diagnosis method based on big data, applied to third-generation compound semiconductors, characterized in that, The big data-based production efficiency diagnosis method includes: Acquire diagnostic sample data related to the production of third-generation compound semiconductors; Based on the diagnostic sample data, multiple performance evaluation indicators are constructed to evaluate production efficiency. A production efficiency evaluation model is constructed by using a pre-set production cycle function and a productivity function, and then trained using the performance evaluation indicators to obtain the production cycle model and the productivity model. The manufacturing execution system collects machine status data, batch processing data, and product management data and uploads them to the production efficiency diagnostic system to obtain diagnostic big data related to the production of third-generation compound semiconductors. Based on the diagnostic big data, multiple performance evaluation indicators are constructed to evaluate production efficiency. These performance evaluation indicators are used to evaluate product quality, product production cycle, resource input, and production output. The performance evaluation indicators include product quality evaluation indicators, product production cycle evaluation indicators, resource input evaluation indicators, and production output evaluation indicators. The product quality evaluation index and the product production cycle evaluation index are input into a preset production cycle model, and a genetic algorithm is used to diagnose the production efficiency of the production cycle model to obtain the first optimal combination of data. The resource input evaluation index and the production output evaluation index are input into a pre-set productivity model, and a genetic algorithm is used to diagnose the production efficiency of the productivity model to obtain a second optimal combination of data; wherein, the production cycle model is used to evaluate the relationship between product quality and product production cycle, and the productivity model is used to evaluate the relationship between resource input and production output. Hypothesis analysis is performed on the first and second optimal data combinations to obtain multiple third optimal data combinations with different quantifications and priorities. A Pareto analysis is performed on the optimal combination of the third data to obtain a final improvement scheme with priority quantification for user reference.

2. The production efficiency diagnosis method based on big data according to claim 1, characterized in that, The product quality evaluation indicators include the number of qualified products and the total number of products; the product production cycle evaluation includes the process production time and the equipment production time. The step of inputting the product quality evaluation index and the product production cycle evaluation index into a preset production cycle model, and using a genetic algorithm to diagnose the production efficiency of the production cycle model, includes: The qualified product quantity, the total number of products, the process production time, and the equipment production time are input into a preset production cycle model, and the production cycle model is subjected to multiple initial replication operations to obtain an initial combination of data for quantifying the production cycle using multiple indicators. The initial combination of data quantified by each of the aforementioned indicators is input into the production cycle model, and multiple feasible cross operations are performed on the production cycle model to obtain multiple feasible combinations of data quantified by the production cycle for each indicator. The feasible combination of data for quantifying the production cycle based on each of the aforementioned indicators is input into the production cycle model, and multiple target mutation operations are performed on the production cycle model to obtain the first optimal combination of data for quantifying the production cycle based on multiple indicators.

3. The production efficiency diagnosis method based on big data according to claim 1, characterized in that, The resource input evaluation indicators include the number of equipment, capital investment, and number of production workers; the production output evaluation indicators include the output rate. The step of inputting the resource input evaluation indicators and the production output evaluation indicators into a pre-set productivity model, and then using a genetic algorithm to diagnose the production efficiency of the productivity model, includes: The number of devices, the capital investment, the number of workers and the output rate are input into the productivity model, and the productivity model is subjected to multiple initial replication operations to obtain an initial combination of multiple indicators for quantifying productivity. The initial combination of the data for quantifying productivity of each of the aforementioned indicators is input into the productivity model, and multiple feasible cross operations are performed on the productivity model to obtain multiple feasible combinations of data for quantifying productivity of the indicators. The feasible combinations of data for quantifying productivity based on each of the aforementioned indicators are input into the productivity model, and multiple target mutation operations are performed on the productivity model to obtain the second optimal combination of data for quantifying productivity based on multiple indicators.

4. The production efficiency diagnosis method based on big data according to claim 1, characterized in that, The diagnostic big data includes: machine status data, batch transaction data, and product management data.

5. A production efficiency diagnostic device based on big data, applied to third-generation compound semiconductors, wherein the big data production efficiency diagnostic device comprises: The data acquisition module is used to acquire diagnostic big data related to the production of third-generation compound semiconductors; The data acquisition module is further used to acquire diagnostic sample data related to the production of third-generation compound semiconductors; based on the diagnostic sample data, multiple performance evaluation indicators for evaluating production efficiency are constructed; corresponding production efficiency evaluation models are constructed using preset production cycle functions and productivity functions, and trained using each of the performance evaluation indicators to obtain the corresponding production cycle model and productivity model; machine status data, batch posting data, and product management data are collected through the manufacturing execution system and uploaded to the production efficiency diagnostic system to obtain diagnostic big data related to the production of third-generation compound semiconductors; The indicator construction module is used to construct multiple performance evaluation indicators for evaluating production efficiency based on the diagnostic big data. These performance evaluation indicators are used to evaluate product quality, product production cycle, resource input and production output. The indicator construction module is further used to input the product quality evaluation indicators and the product production cycle evaluation indicators into a preset production cycle model, and to use a genetic algorithm to diagnose the production efficiency of the production cycle model to obtain the first optimal combination of data. The efficiency diagnosis module is used to input the performance evaluation indicators into the preset production cycle model and productivity model to diagnose production efficiency and obtain the second optimal combination of data. The production cycle model is used to evaluate the relationship between product quality and product production cycle, and the productivity model is used to evaluate the relationship between resource input and production output. The efficiency diagnosis module is further used to perform hypothesis analysis on the first and second optimal data combinations, respectively obtaining multiple third optimal data combinations with different quantifications and priorities; and to perform Pareto analysis on the third optimal data combinations to obtain a final improvement scheme with priority quantification for user reference.

6. An electronic device applied to third-generation compound semiconductors, characterized in that, The big data production efficiency diagnostic device includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the electronic device to execute the big data-based production efficiency diagnosis method as described in any one of claims 1-4.

7. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements the big data-based production efficiency diagnosis method as described in any one of claims 1-4.

Images