An artificial intelligence analysis method and system based on digital factory operation data

Abstract

The present application discloses an artificial intelligence analysis method and system based on digital factory operation data, which relates to the field of smart factory technology, in order to at least solve the problem of high production costs in the factory. The analysis method includes: obtaining the operation data of the digital factory, the operation data includes production equipment parameters, quality inspection system parameters and historical repair data; obtaining abnormal information of the current abnormal product, the abnormal information includes abnormal product labels, abnormal feature parameters and abnormal process node information; determining the repair cost and scrap cost of the current abnormal product according to the operation data and abnormal information; at least when the repair cost is less than the scrap cost, according to the abnormal process node information, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair; updating the historical repair data corresponding to the production equipment parameters. The analysis method provided by the present application can prevent abnormal products from increasing the production costs of enterprises.

Description

Technical Field

[0001] The present application relates to the field of smart factory technology, and in particular to an artificial intelligence analysis method and system based on digital factory operation data. Background Art

[0002] The specific content of the digital factory is to establish an integrated management platform based on a three-dimensional virtual refinery, integrating various dynamic and static data such as engineering design, production operation, mobile equipment, environmental protection and safety, and using two-dimensional visualization technology to provide enterprise asset information full life cycle management. The system has highly configurable applications and is an expanded application of the enterprise information management system. In the entire process of physical assets from generation, operation to retirement, it provides asset information engineering design, digital asset transfer, integration and life cycle management, technical changes, operation and maintenance, dynamic display, information query and simulation demonstration, which can provide a management environment with accurate data support for planning, design, construction, operation and other departments.

[0003] Nowadays, with the increasing popularization of digital factories, artificial intelligence analysis methods and systems for digital factory operation data are particularly important. Summary of the invention

[0004] The embodiments of the present application provide an artificial intelligence analysis method and system based on digital factory operation data to at least solve the problem of high production costs in the factory.

[0005] In order to solve the above technical problems, the embodiments of the present application adopt the following technical solutions:

[0006] According to a first aspect of an embodiment of the present application, an artificial intelligence analysis method based on digital factory operation data is provided, the analysis method comprising:

[0007] Acquire operation data of the digital factory, the operation data including production equipment parameters, quality inspection system parameters, and historical repair data corresponding to the production equipment parameters and the quality inspection system parameters respectively;

[0008] Obtaining abnormal information of the current abnormal product, the abnormal information includes abnormal product labels, abnormal characteristic parameters and abnormal process node information, wherein the abnormal characteristic parameters correspond to the abnormal process node information;

[0009] Determine the rework cost and scrap cost of the current abnormal product based on the operation data and abnormal information;

[0010] At least when the repair cost is less than the scrap cost, according to the abnormal process node information, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair;

[0011] Update historical repair data corresponding to production equipment parameters.

[0012] In some embodiments, the historical repair data includes: abnormal information of historical repair products and cost information of historical repair products;

[0013] Determine the repair cost of the current abnormal product based on the operation data and abnormal information, including:

[0014] Fitting the abnormal information of historical repaired products and the cost information of historical repaired products to obtain a repair cost fitting function;

[0015] According to the abnormal information and the rework cost fitting function, the rework cost of the current abnormal product is determined.

[0016] In some embodiments, the abnormal information of the historical reworked products includes abnormal characteristic parameters of the historical reworked products and historical rework equipment occupation costs, rework result parameters and rework product scrapping costs respectively corresponding to the abnormal characteristic parameters of the historical reworked products;

[0017] The abnormal information of historical repaired products and the cost information of historical repaired products are fitted to obtain the repair cost fitting function, including:

[0018] Fitting the abnormal characteristic parameters of historical repaired products and the historical repair equipment occupancy costs corresponding to the abnormal characteristic parameters of historical repaired products to obtain a repair product equipment occupancy cost fitting function;

[0019] Fitting the abnormal characteristic parameters of the historical repaired products and the repair result parameters corresponding to the abnormal characteristic parameters of the historical repaired products to obtain a fitting function of the scrap rate of the historical repaired products corresponding to the abnormal characteristic parameters of the historical repaired products;

[0020] Obtaining a repair product scrap cost fitting function based on a scrap rate fitting function of historical repair products corresponding to the abnormal characteristic parameters of historical repair products and a repair product scrap cost corresponding to the abnormal characteristic parameters of historical repair products;

[0021] The repair cost fitting function is obtained according to the repair product equipment occupation cost fitting function and the repair product scrap cost fitting function.

[0022] In some embodiments, the operating data also includes historical data of unrepaired products corresponding to production equipment parameters.

[0023] Before returning the current abnormal product to the process node corresponding to the abnormal process node information for repair, the analysis method further includes:

[0024] According to the abnormal information and the scrap rate fitting function of the historical repaired products, the yield rate of the current abnormal product after repair is determined;

[0025] Determine the yield rate of the unrepaired products of the production equipment corresponding to the abnormal information according to the historical data of the unrepaired products corresponding to the production equipment parameters;

[0026] Determine the change in the production efficiency of the digital factory after the repair of the current abnormal product according to the yield rate of the current abnormal product after repair and the yield rate of the unrepaired products of the production equipment corresponding to the abnormal information;

[0027] At least when the change in production efficiency is less than zero and the rework cost is less than the scrap cost, the current abnormal product will be temporarily stored.

[0028] In some embodiments, the operation data further includes: order management system parameters and production system parameters, wherein the order management system includes sales quantity information within multiple historical time periods, and the production system parameters include current product production rate;

[0029] Before returning the current abnormal product to the process node corresponding to the abnormal process node information for repair, the analysis method further includes:

[0030] Determine multiple sales quantity information sequences based on sales quantity information within multiple historical time periods;

[0031] Based on the sales quantity information sequence, predict the future sales rate within a preset time period;

[0032] At least when the change in production efficiency is less than zero and the repair cost is less than the scrap cost, the current abnormal products are temporarily stored, including:

[0033] At least when the future sales rate is greater than or equal to the current product production rate, the change in production efficiency is less than zero, and the rework cost is less than the scrap cost, the current abnormal product will be temporarily stored.

[0034] In some embodiments, at least when the repair cost is less than the scrap cost, according to the abnormal process node information, returning the current abnormal product to the process node corresponding to the abnormal process node information for repair includes:

[0035] At least when the change in production efficiency is greater than or equal to zero, and the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for rework; and / or, at least when the future sales rate is greater than or equal to the current product production rate, the change in production efficiency is greater than or equal to zero, and the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for rework.

[0036] In some embodiments, the operation data further includes product inventory parameters, and the product inventory parameters include current inventory quantity and optimal inventory quantity.

[0037] The analysis method includes: temporarily storing the current abnormal product at least when the previous inventory number is greater than or equal to the optimal inventory number, the change in production efficiency is greater than zero, and the rework cost is less than the scrap cost;

[0038] In some embodiments, at least when the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair based on the abnormal process node information, including: at least when the previous inventory number is less than the optimal inventory number, the change in production efficiency is greater than or equal to zero, and the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair.

[0039] In some embodiments, the production equipment parameters include: the working state of the production equipment and the rated circuit information of the production equipment corresponding to the working state of the production equipment;

[0040] Before updating the historical repair data corresponding to the production equipment parameters, the analysis method further includes:

[0041] Obtain the working status of the production equipment and the actual circuit information of the production equipment during the repair process of the current abnormal product;

[0042] When the actual circuit information is outside the rated circuit information, the repair data of the current abnormal product is eliminated;

[0043] Update historical repair data corresponding to production equipment parameters, including:

[0044] When the actual circuit information is within the rated circuit information, the historical repair data corresponding to the production equipment parameters is updated according to the repair data of the current abnormal product.

[0045] According to the second aspect of the embodiment of the present application, there is provided an artificial intelligence analysis system based on the operation data of a digital factory, which is applied to a digital factory, and the intelligent analysis system includes: a data acquisition module, a data storage module and a data processing module. Among them, the data acquisition module is used to collect the operation data of the digital factory from the equipment, production system and quality inspection system of the digital factory, and is used to collect the abnormal information of the current abnormal product, and the abnormal information includes the abnormal product label, abnormal characteristic parameters and abnormal process node information, wherein the abnormal characteristic parameters correspond to the abnormal process node information. The data storage module is used to store the operation data of the digital factory and the abnormal information of the abnormal product. The data processing module is used to determine the rework cost and scrap cost of the current abnormal product according to the operation data and the abnormal information. The control module is used to control the production equipment to return the current abnormal product to the process node corresponding to the abnormal process node information for rework when the rework cost is less than the scrap cost, according to the abnormal process node information, and is used to control the data storage module to update the historical rework data corresponding to the production equipment parameters.

[0046] In some embodiments, the artificial intelligence analysis system also includes a terminal system, which is connected to the data storage module, and the terminal system is used to digitally display the operating data of the digital factory.

[0047] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0048] The artificial intelligence analysis method based on digital factory operation data provided by the embodiment of the present application can quickly analyze the size of the repair cost and scrap cost based on historical repair data when an abnormality occurs at any process node in the production process of the target product, and only implement the repair operation when the repair cost of the abnormal product is less than the protection cost. In other words, not all repairable products are subject to repair operations. This method can eliminate abnormal products with high repair costs, avoid the repaired products from further increasing the average preparation cost of each product, and is beneficial to reducing the total cost of factory production. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0050] Figure 1 A flowchart diagram of an artificial intelligence analysis method based on digital factory operation data provided in an embodiment of the present application;

[0051] Figure 2A schematic diagram of the structure of an artificial intelligence analysis system based on digital factory operation data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0053] As described in the background art, in the digital factory in the related art, the abnormal products that can be repaired are not evaluated, but all the abnormal products that can be repaired are directly repaired, which leads to high repair costs and increases the production costs of the factory.

[0054] In view of the above situation, this application provides an artificial intelligence analysis method based on digital factory operation data. Figure 1 As shown, the analysis method includes:

[0055] Step 101, obtaining operation data of the digital factory. The operation data includes production equipment parameters, quality inspection system parameters, and historical repair data corresponding to the production equipment parameters and the quality inspection system parameters respectively.

[0056] Exemplarily, the production equipment parameters may be various parameters of the operation of the production equipment, such as the feed speed, feed amount, rotation speed, etc. of the cutting equipment. The quality inspection system parameters may be incoming material inspection data, warehouse inspection data, process inspection plan data, process inspection data, completion inspection data, graded quality inspection data, quality inspection statistical data, quality inspection analysis data, graded statistical data, incoming material inspection standard data, warehouse inspection standard data, product inspection standard data, graded quality inspection template data, quality report plan data, inspection item data, etc.

[0057] Historical repair data can be all the data of the repair of the production products during the production process. Specifically, it can be the label information of the repaired products, the process points where the abnormalities occurred, the abnormal features and the deviations from the standard features, as well as the time required for various abnormal repairs and the cost information of the materials required for the repairs.

[0058] For example, a corresponding database may be established to store the production equipment parameters, quality inspection system parameters, and historical repair data of the digital factory.

[0059] Step 103, obtaining abnormal information of the current abnormal product. The abnormal information includes abnormal product labels, abnormal characteristic parameters and abnormal process node information, wherein the abnormal characteristic parameters correspond to the abnormal process node information.

[0060] Exemplarily, the abnormal product label can be an abnormal product number or an abnormal product QR code, etc., which can be used to identify and distinguish each abnormal product. The abnormal characteristic parameter can be the amount of product shape, color or performance that exceeds the range of product standards due to the abnormal process node of the product. Exemplarily, if a hole in the product is abnormal, the abnormal characteristic parameter can be the aperture size of the hole or the difference between the aperture of the abnormal product and the standard aperture of the product. Abnormal process node information, that is, the process node information corresponding to the abnormal characteristic parameter, is information that can characterize at which process node the abnormal characteristic parameter is abnormal.

[0061] In some embodiments, when it is detected that a product is abnormal after a certain process node is completed, an abnormal information label corresponding to the process node may be generated, and abnormal characteristic parameters corresponding to the abnormal information expression may be recorded.

[0062] In some embodiments, as needed, repair evaluation can be performed on abnormal products with different abnormal characteristic parameters to obtain initial data of historical repair data.

[0063] Step 105, determining the repair cost and scrap cost of the current abnormal product according to the operation data and the abnormal information.

[0064] Exemplarily, the abnormal information of the current abnormal product can be compared with the historical repair data, and the repair cost of a group of historical repair data closest to the abnormal information of the current abnormal product can be used as the repair cost of the current abnormal product.

[0065] Exemplarily, the scrapping cost of the current abnormal product can be generated based on historical production data. Exemplarily, the scrapping cost of the current abnormal product can be the cost required for the raw material to go through each process node and complete the process node corresponding to the abnormal information. In some embodiments, the scrapping cost includes raw material cost and processing cost.

[0066] In some embodiments, the scrap cost of the product protected after the process node is completed can be calculated according to the process node, that is, the scrap cost of a specific product after an abnormality occurs at a specific process node is always the same. Exemplarily, the scrap cost corresponding to each process node can be pre-set.

[0067] In some embodiments, when an abnormal product has an unrepairable defect, the repair cost of the abnormal product is a preset value. The preset value is greater than or equal to the cost required to complete the production of a qualified product. In this way, it can be ensured that some abnormal products that cannot be repaired can be scrapped in time. Among them, an abnormal product that cannot be repaired to a qualified product by repairing it based on the existing process equipment.

[0068] Step 107 , at least when the repair cost is less than the scrap cost, according to the abnormal process node information, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair.

[0069] Exemplarily, a normal product branch, a scrapped product branch, and a repaired product branch are provided after each production equipment. Among them, the normal product branch is used to transport or store qualified products. The scrapped product branch is used to transport or store scrapped products. The repaired product branch is used to transport or store abnormal products that need to be repaired.

[0070] In the related art, when an abnormal product appears, whether the product can be repaired is used as a criterion to determine whether to repair the abnormal product. However, some abnormal products are difficult to repair and costly to repair. This will not only result in high repair costs, a large number of repairs, and occupy the internal logistics of the factory.

[0071] In the artificial intelligence analysis method based on digital factory operation data provided by the present application, when a product is abnormal, those abnormal products with high rework costs can be eliminated, and those abnormal products that are returned to the local area can also be retained, which is beneficial to reducing production costs.

[0072] Step 109, updating the historical repair data corresponding to the production equipment parameters.

[0073] Specifically, after the current abnormal product is repaired, the data of the current abnormal product repair can be stored, and the historical repair data corresponding to the production equipment parameters can be updated to make the estimation of the repair cost of the abnormal product more accurate, thereby helping to reduce the production cost of the digital factory.

[0074] In the process of product preparation, even if anomalies occur at the same process node, the generated anomaly information is not completely consistent. Specifically, the cause of this anomaly may be inconsistent wear of production equipment in different periods. Of course, it may also be caused by changes in environmental conditions. For example, it may be changes in ambient temperature in different seasons.

[0075] In some embodiments, the artificial intelligence analysis method based on digital factory operation data also includes: step 10, scrapping the current abnormal product at least when the rework cost is greater than the scrap cost.

[0076] In some embodiments, the historical repair data includes: abnormal information of historical repair products and cost information of historical repair products. Step 105, determining the repair cost of the current abnormal product according to the operation data and the abnormal information includes:

[0077] Step 1051, fitting the abnormal information of historical repaired products and the cost information of historical repaired products to obtain a repair cost fitting function.

[0078] In some embodiments, the abnormal information may be classified, and different fitting methods may be used for different abnormal information to fit the abnormal information and the cost information of historical repaired products.

[0079] For example, for some historical repair data with known abnormal information and repair cost being linearly correlated, a linear function fitting method can be used to fit the abnormal information and cost information of this type. For some historical repair data with known abnormal information and repair cost being nonlinearly correlated, a nonlinear least squares fitting method can be used to fit.

[0080] Step 1053, determining the rework cost of the current abnormal product according to the abnormal information and the rework cost fitting function.

[0081] Exemplarily, the abnormal information of the current abnormal product can be directly brought into the repair cost fitting function to obtain the cost required for repairing the current abnormal product.

[0082] The above-mentioned artificial intelligence analysis method based on digital factory operation data can estimate the rework cost of current abnormal products through historical rework data and abnormal information of current abnormal products, so as to eliminate abnormal products with high rework costs, which is beneficial to reduce rework costs.

[0083] The repair cost of abnormal product repair is related to the abnormal physical parameters of the abnormal product. For example, in the process of processing a hole structure of the product structure, the difference between the size of the hole structure and the standard size directly determines the processing amount during the product repair process, which will make the equipment repair time longer and the cost higher.

[0084] In some embodiments, the abnormal information of historical repaired products includes abnormal characteristic parameters of the historical repaired products and historical repair equipment occupancy costs, repair result parameters and repaired product scrap costs corresponding to the abnormal characteristic parameters of the historical repaired products, respectively.

[0085] In some embodiments, step 1051 in the artificial intelligence analysis method based on digital factory operation data is to fit the abnormal information of historical rework products and the cost information of historical rework products to obtain a rework cost fitting function, including:

[0086] Step 10511, fitting the abnormal characteristic parameters of historical repaired products and the historical repair equipment occupancy costs corresponding to the abnormal characteristic parameters of historical repaired products, to obtain a repair product equipment occupancy cost fitting function.

[0087] In some embodiments, when there are multiple abnormal parameter features in the abnormal information of historical repaired products, each abnormal parameter feature in the historical repaired products and the cost of repairing and correcting each abnormal parameter feature can be fitted separately, and then the multiple fitted results are added together.

[0088] Exemplarily, the equipment occupation cost may include but is not limited to equipment depreciation cost and equipment opportunity cost. The equipment occupation cost may include but is not limited to the occupation cost of the production equipment for processing the corresponding process node. In some embodiments, the equipment occupation cost also includes the equipment occupation cost of the handling equipment used to transport the reworked products.

[0089] Step 10513, fitting the abnormal characteristic parameters of the historical repaired products and the repair result parameters corresponding to the abnormal characteristic parameters of the historical repaired products, to obtain a fitting function of the scrap rate of the historical repaired products corresponding to the abnormal characteristic parameters of the historical repaired products.

[0090] Exemplarily, the repair result parameters corresponding to the abnormal characteristic parameters may include, but are not limited to, the quality assessment results of the product after the abnormal product is repaired. Exemplarily, the quality assessment results may include, but are not limited to, needing repair, needing scrapping, and qualified.

[0091] Step 10515, obtaining a repair product scrap cost fitting function based on the scrap rate fitting function of the historical repair products corresponding to the abnormal characteristic parameters of the historical repair products and the scrap cost of the repair products corresponding to the abnormal characteristic parameters of the historical repair products.

[0092] In some embodiments, the cost of processing the raw materials of the product to the process node corresponding to the abnormal characteristic parameters can be used as the scrap cost.

[0093] In some embodiments, the scrapping cost also includes, but is not limited to, the transportation cost, storage cost, and cost required for the disposal of the scrapped products.

[0094] Step 10517, obtaining a repair cost fitting function according to the repair product equipment occupancy cost fitting function and the repair product scrap cost fitting function.

[0095] Exemplarily, the repair cost fitting function may be obtained by adding the repair product equipment occupancy cost fitting function to the repair product scrap cost fitting function.

[0096] In some embodiments, the operation data also includes historical data of unrepaired products corresponding to the production equipment parameters. In step 107, before returning the current abnormal product to the process node corresponding to the abnormal process node information for repair, the artificial intelligence analysis method based on the digital factory operation data also includes:

[0097] Step 111, determining the post-repair yield rate of the current abnormal product according to the abnormal information and the scrap rate fitting function of the historical re-repaired products.

[0098] Step 113, determining the yield rate of the non-repaired products of the production equipment corresponding to the abnormal information according to the historical data of the non-repaired products corresponding to the production equipment parameters.

[0099] Step 115 , determining the change in the production efficiency of the digital factory after the repair of the current abnormal product according to the yield rate of the current abnormal product after repair and the yield rate of the unrepaired products of the production equipment corresponding to the abnormal information.

[0100] Among them, the production efficiency of the digital factory is the number of qualified products produced by the digital factory per unit time.

[0101] Step 117, temporarily storing the current abnormal product at least when the change in production efficiency is less than zero and the rework cost is less than the scrap cost.

[0102] For example, when the yield rate of the current abnormal product after repair is lower than the yield rate of the unrepaired product of the production equipment corresponding to the abnormal information, the change in production efficiency is less than zero. That is, the current abnormal product repair will reduce the production efficiency of the digital factory.

[0103] When the yield rate of the current abnormal product after repair is equal to the yield rate of the unrepaired product of the production equipment corresponding to the abnormal information, the change in production efficiency is zero. That is, the current abnormal product repair will not change the production efficiency of the digital factory.

[0104] For example, when the yield rate of the current abnormal product after repair is greater than the yield rate of the unrepaired product of the production equipment corresponding to the abnormal information, the change in production efficiency is less than zero. That is, the current abnormal product repair will improve the production efficiency of the digital factory.

[0105] The above-mentioned embodiments can prevent abnormal products from being returned for repair and affecting the production efficiency of the digital factory, thereby ensuring the stability of the output.

[0106] In some embodiments, when the change in production efficiency is less than zero and the rework cost is less than the scrap cost, the current abnormal products are temporarily stored. In this way, when the market demand for the product decreases, the temporarily stored abnormal products can be reworked. This not only reduces the production cost of the product, but also is beneficial to the production regulation of the digital factory.

[0107] In some embodiments, when the raw material process of the digital factory lags behind, the temporarily stored abnormal products can be returned for repair, which helps to reduce the impact of raw material shortages on factory output.

[0108] In some embodiments, the market demand for the products produced by the digital factory has a time periodicity. That is, the market demand for the products produced by the digital factory is different in different time periods. For example, if the production object is summer clothing, the demand increases in summer, and the demand decreases in spring, autumn and winter. Therefore, the digital factory needs to adjust the output in order to adapt to market demand.

[0109] In some embodiments, the operation data further includes: order management system parameters and production system parameters, wherein the order management system includes sales quantity information within multiple historical time periods, and the production system parameters include the current product production rate. In step 107, before returning the current abnormal product to the process node corresponding to the abnormal process node information for repair according to the abnormal process node information, the artificial intelligence analysis method based on the digital factory operation data also includes:

[0110] Step 119, determining a plurality of sales quantity information sequences based on the sales quantity information within a plurality of historical time periods.

[0111] For example, product sales data can be counted according to the periodicity of product demand to obtain sales quantity information in multiple historical time periods. For example, for a digital factory that produces clothing, sales data of production products can be counted according to different seasons.

[0112] Step 121, predicting the future sales rate within a preset time period based on the sales quantity information sequence.

[0113] Exemplarily, time series analysis, regression analysis, weighted average, moving average and exponential smoothing methods may be used but are not limited to them.

[0114] Step 117, at least when the change in production efficiency is less than zero and the repair cost is less than the scrap cost, temporarily storing the current abnormal product, includes:

[0115] Step 1171, at least when the future sales rate is greater than or equal to the current product production rate, the change in production efficiency is less than zero, and the repair cost is less than the scrap cost, the current abnormal product is temporarily stored. This can avoid the current abnormal product repair reducing the output of the digital factory, and can maximize production efficiency by repairing the abnormal product.

[0116] In some embodiments, the artificial intelligence analysis method based on the operation data of the digital factory further includes: Step 12, when the change in production efficiency is less than zero and the rework cost is greater than or equal to the scrap cost, the current abnormal product is temporarily stored. Exemplarily, in some cases, in order to achieve the production quantity target within a preset time, and when the production efficiency of the current digital factory is not enough to complete it, the production efficiency of the digital factory can be ensured by temporarily storing the abnormal products.

[0117] In some embodiments, in step 12, when the change in production efficiency is less than zero and the rework cost is greater than or equal to the scrap cost, after temporarily storing the current abnormal product, the artificial intelligence analysis method based on the digital factory operation data further includes:

[0118] When the future sales rate is lower than the current product production rate, the change in production efficiency is less than zero, and the repair cost is lower than the scrap cost, the abnormal products temporarily stored in step 12 are repaired, which is beneficial to achieving production and sales balance, reducing preparation costs and inventory costs.

[0119] In some embodiments, when the future sales rate is greater than the current product production rate, the delivery delay cost is calculated as the scrap cost. In some embodiments, the minimum value of the delivery delay cost is greater than the production cost of a qualified product. In this way, when the output is less than the sales volume, it is beneficial to increase the output of the digital factory.

[0120] In some embodiments, when the future sales rate is less than the current product production rate, the delivery delay cost is deducted from the scrap cost. Exemplarily, when the future sales rate is less than the current product production rate, the delivery delay cost is zero.

[0121] In some embodiments, step 107, at least when the repair cost is less than the scrap cost, returning the current abnormal product to the process node corresponding to the abnormal process node information for repair according to the abnormal process node information, includes:

[0122] Step 1071, at least when the change in production efficiency is greater than or equal to zero and the repair cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair. This embodiment can improve production efficiency by repairing and is also beneficial to reducing production costs.

[0123] In some embodiments, step 107, at least when the repair cost is less than the scrap cost, returning the current abnormal product to the process node corresponding to the abnormal process node information for repair according to the abnormal process node information, includes:

[0124] Step 1073, at least when the future sales rate is greater than or equal to the current product production rate, the change in production efficiency is greater than or equal to zero, and the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for rework.

[0125] In the above implementation, effectiveness can be improved by returning abnormal products for repair, thereby benefiting the balance between production and sales.

[0126] In some embodiments, in order to meet daily supply and sales, the digital factory generally needs to set an optimal inventory quantity to avoid repeated adjustments of production equipment, so that the production equipment is always in a relatively stable working state.

[0127] In some embodiments, the operation data also includes product inventory parameters. The product inventory parameters include the current inventory number and the optimal inventory number. The artificial intelligence analysis method based on the digital factory operation data also includes:

[0128] Step 123, temporarily storing the current abnormal product at least when the previous inventory number is greater than or equal to the optimal inventory number, the change in production efficiency is greater than zero, and the rework cost is less than the scrap cost.

[0129] In some embodiments, the operation data further includes product inventory parameters, and the product inventory parameters include current inventory and optimal inventory. Step 107, at least when the repair cost is less than the scrap cost, returns the current abnormal product to the process node corresponding to the abnormal process node information for repair according to the abnormal process node information, including:

[0130] Step 1075, at least when the previous inventory number is less than the optimal inventory number, the change in production efficiency is greater than or equal to zero, and the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for rework.

[0131] In the above embodiment, the inventory level of the digital factory can be regulated by controlling the repair of abnormal products, which is beneficial for making the inventory level closer to the optimal inventory level.

[0132] In some embodiments, the production equipment parameters include: a working state of the production equipment and rated circuit information of the production equipment corresponding to the working state of the production equipment.

[0133] Step 109, before updating the historical repair data corresponding to the production equipment parameters, the artificial intelligence analysis method based on the digital factory operation data further includes:

[0134] Step 125, obtaining the working status of the production equipment and the actual circuit information of the production equipment during the repair of the current abnormal product; wherein the circuit information includes but is not limited to current information, voltage information, and temperature information of each part of the circuit.

[0135] Step 127, when the actual circuit information is outside the rated circuit information, remove the repair data of the current abnormal product. Exemplarily, when the actual circuit information is outside the rated circuit information, the removal can be achieved by selecting not to store. In some embodiments, when the actual circuit information is outside the rated circuit information, the part of the data can be stored in a separate group.

[0136] The above embodiment can obtain the working status of the production equipment by analyzing the actual circuit information of the production equipment, so as to judge whether there is equipment abnormality during the repair process of the abnormal product. It is further helpful to propose some data that cause the cost of the repaired product to increase or the yield rate to decrease due to equipment abnormality. In this way, some accidental events can be avoided from affecting the accuracy of the repair data. Among them, accidental events may include but are not limited to equipment damage and power supply abnormality. Therefore, this embodiment is helpful to accurately evaluate the repair cost and scrap cost of the current abnormal product.

[0137] Step 109, updating the historical repair data corresponding to the production equipment parameters, including:

[0138] Step 1091, when the actual circuit information is within the rated circuit information, the historical repair data corresponding to the production equipment parameters is updated according to the repair data of the current abnormal product, which is beneficial to improve the accuracy of the repair data, and further improve the accuracy of subsequent repair cost accounting.

[0139] In some embodiments, the present application also provides an artificial intelligence analysis system based on digital factory operation data. The artificial intelligence analysis system based on digital factory operation data includes:

[0140] Data acquisition module, data storage module and data processing module. The data acquisition module is used to collect the operation data of the digital factory from the equipment, production system and quality inspection system of the digital factory, and is used to collect the abnormal information of the current abnormal product, the abnormal information includes the abnormal product label, abnormal characteristic parameters and abnormal process node information, wherein the abnormal characteristic parameters correspond to the abnormal process node information. Exemplarily, the data acquisition module may include sensors, image acquisition devices and detection devices arranged in each production equipment or detection equipment.

[0141] The data storage module is used to store the operation data of the digital factory and the abnormal information of the abnormal products. The data processing module is used to determine the rework cost and scrap cost of the current abnormal product based on the operation data and the abnormal information. The control module is used to control the production equipment to return the current abnormal product to the process node corresponding to the abnormal process node information for rework when the rework cost is less than the scrap cost, and to control the data storage module to update the historical rework data corresponding to the production equipment parameters.

[0142] In some embodiments, the artificial intelligence analysis system based on the operation data of the digital factory also includes a terminal system connected to the data storage module, and the terminal system is used to digitally display the operation data of the smart factory. Exemplarily, the terminal system may include a display terminal and a host. The display terminal is connected to the host to digitally display the operation data of the digital factory through the display terminal.

[0143] In short, the above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0144] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0145] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

Claims

1. An artificial intelligence analysis method based on digital factory operation data, characterized in that: Analytical methods include: Acquire operation data of the digital factory, wherein the operation data includes production equipment parameters, quality inspection system parameters, and historical repair data corresponding to the production equipment parameters and the quality inspection system parameters respectively; Acquire abnormal information of the current abnormal product, wherein the abnormal information includes an abnormal product label, abnormal characteristic parameters, and abnormal process node information, wherein the abnormal characteristic parameters correspond to the abnormal process node information; Determine the rework cost and scrap cost of the current abnormal product according to the operation data and the abnormal information; At least when the repair cost is less than the scrap cost, according to the abnormal process node information, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair; Updating the historical repair data corresponding to the production equipment parameters; The historical repair data includes: abnormal information of historical repair products and cost information of historical repair products; Determining the repair cost of the current abnormal product according to the operation data and the abnormal information includes: Fitting the abnormal information of historical repaired products and the cost information of historical repaired products to obtain a repair cost fitting function; The rework cost of the current abnormal product is determined according to the abnormal information and the rework cost fitting function.

2. The analysis method according to claim 1, characterized in that The abnormal information of the historical repaired products includes abnormal characteristic parameters of the historical repaired products and historical repair equipment occupation costs, repair result parameters and repaired product scrapping costs respectively corresponding to the abnormal characteristic parameters of the historical repaired products; The abnormal information of historical repaired products and the cost information of historical repaired products are fitted to obtain the repair cost fitting function, including: Fitting the abnormal characteristic parameters of historical repaired products and the historical repair equipment occupancy costs corresponding to the abnormal characteristic parameters of historical repaired products to obtain a repair product equipment occupancy cost fitting function; Fitting the abnormal characteristic parameters of the historical repaired products and the repair result parameters corresponding to the abnormal characteristic parameters of the historical repaired products to obtain a fitting function of the scrap rate of the historical repaired products corresponding to the abnormal characteristic parameters of the historical repaired products; Obtaining a repair product scrap cost fitting function based on a scrap rate fitting function of historical repair products corresponding to the abnormal characteristic parameters of historical repair products and a repair product scrap cost corresponding to the abnormal characteristic parameters of historical repair products; A repair cost fitting function is obtained according to the repair product equipment occupancy cost fitting function and the repair product scrap cost fitting function.

3. The analysis method according to claim 2, characterized in that The operation data also includes historical data of unrepaired products corresponding to the production equipment parameters, and before the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair, the analysis method also includes: Determine the yield rate of the current abnormal product after repair based on the abnormal information and the scrap rate fitting function of the historical repaired products; Determine the yield rate of the unrepaired products of the production equipment corresponding to the abnormal information according to the historical data of the unrepaired products corresponding to the production equipment parameters; Determine the change in the production efficiency of the digital factory after the repair of the current abnormal product according to the yield rate of the current abnormal product after repair and the yield rate of the unrepaired product of the production equipment corresponding to the abnormal information; At least when the change in the production efficiency is less than zero and the rework cost is less than the scrap cost, the current abnormal product is temporarily stored.

4. The analysis method according to claim 3, characterized in that The operation data also includes: order management system parameters and production system parameters, wherein the order management system includes sales quantity information within multiple historical time periods, and the production system parameters include current product production rate; Before returning the current abnormal product to the process node corresponding to the abnormal process node information for repair, the analysis method further includes: Determine multiple sales quantity information sequences based on sales quantity information within multiple historical time periods; According to the sales quantity information sequence, predict the future sales rate within a preset time period; At least when the change in the production efficiency is less than zero and the rework cost is less than the scrap cost, temporarily storing the current abnormal product includes: At least when the future sales rate is greater than or equal to the current product production rate, the change in production efficiency is less than zero, and the rework cost is less than the scrap cost, the current abnormal product is temporarily stored.

5. The analysis method according to claim 4, characterized in that At least when the rework cost is less than the scrap cost, returning the current abnormal product to the process node corresponding to the abnormal process node information for rework according to the abnormal process node information, includes: At least when the change in the production efficiency is greater than or equal to zero, and the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for rework; or, at least when the future sales rate is greater than or equal to the current product production rate, the change in the production efficiency is greater than or equal to zero, and the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for rework.

6. The analysis method according to claim 4, characterized in that The operation data also includes product inventory parameters, the product inventory parameters include a current inventory number and an optimal inventory number, and the analysis method includes: at least when the previous inventory number is greater than or equal to the optimal inventory number, the change in production efficiency is greater than zero, and the rework cost is less than the scrap cost, temporarily storing the current abnormal product; And / or, at least when the rework cost is less than the scrap cost, based on the abnormal process node information, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair, including: at least when the previous inventory number is less than the optimal inventory number, the change in production efficiency is greater than or equal to zero, and the rework cost is less than the scrap cost, the current abnormal product is returned to the process node corresponding to the abnormal process node information for repair.

7. The analysis method according to any one of claims 1 to 6, characterized in that The production equipment parameters include: the working state of the production equipment and the rated circuit information of the production equipment corresponding to the working state of the production equipment; Before updating the historical repair data corresponding to the production equipment parameters, the analysis method further includes: Acquire the working status of the production equipment and the actual circuit information of the production equipment during the process of repairing the current abnormal product; In the case where the actual circuit information is outside the rated circuit information, removing the repair data of the current abnormal product; Updating the historical repair data corresponding to the production equipment parameters includes: In a case where the actual circuit information is within the rated circuit information, the historical repair data corresponding to the production equipment parameters is updated according to the repair data of the current abnormal product.

8. An artificial intelligence analysis system based on digital factory operation data, characterized in that: include: A data acquisition module, the data acquisition module is used to collect the operation data of the digital factory from the equipment, production system, and quality inspection system of the digital factory, and is used to collect abnormal information of the current abnormal product, the abnormal information includes abnormal product labels, abnormal characteristic parameters, and abnormal process node information, wherein the abnormal characteristic parameters correspond to the abnormal process node information; A data storage module, the data storage module is used to store the operation data of the digital factory and the abnormal information of abnormal products; A data processing module, the data processing module is used to determine the rework cost and scrap cost of the current abnormal product according to the operation data and the abnormal information; A control module, wherein the control module is used to control the production equipment to return the current abnormal product to the process node corresponding to the abnormal process node information for repair when the rework cost is less than the scrap cost, and to control the data storage module to update the historical rework data corresponding to the production equipment parameters.

9. The analysis system according to claim 8, characterized in that It also includes a terminal system, which is connected to the data storage module and is used to digitally display the operating data of the digital factory.

Images