Sealing system assembly cost prediction method, device and system and storage medium
By collecting and preprocessing multi-source heterogeneous data and training cost models to achieve real-time dynamic optimization of sealing system component costs, the problem of insufficient cost prediction accuracy in traditional methods is solved, and the prediction accuracy and data coverage are improved.
Patent Information
- Application Number
- CN202510737000.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-04
- Publication Date
- 2025-09-19
AI Technical Summary
Traditional sealing system component cost assessment methods rely on manual calculations and cannot fully reflect the various factors affecting cost, resulting in insufficient cost prediction accuracy.
By collecting multi-source heterogeneous data, pre-processing it and inputting it into the cost model for training, and using the theoretical optimal cost annotation to force the model to learn the predicted value close to the optimal cost, real-time dynamic optimization is achieved.
The accuracy of sealing system component cost prediction and the comprehensiveness of cost influencing factors covered are improved, ensuring the real-time and accuracy of data and reducing production costs.
Smart Images
Figure CN120672367A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent prediction technology, and in particular to a sealing system component cost prediction method, device, system and storage medium. Background Art
[0002] Sealing system components are widely used in the automotive industry, and controlling their cost is crucial to corporate profits. Traditional sealing system component cost assessment methods rely primarily on manual calculations. Due to limited input data, they cannot fully reflect the various factors affecting cost. Moreover, the system cannot dynamically obtain data in real time, resulting in insufficient cost prediction accuracy.
[0003] Therefore, how to provide a sealing system component cost prediction method to improve the accuracy of sealing system component cost prediction has become a technical problem that needs to be solved urgently. Summary of the Invention
[0004] The present application provides a sealing system component cost prediction method, device, system and storage medium to improve the accuracy of sealing system component cost prediction.
[0005] This application provides a sealing system component cost prediction method, comprising:
[0006] Collect multi-source heterogeneous data that impacts sealing system component costs;
[0007] Preprocess the collected multi-source heterogeneous data;
[0008] The preprocessed data is input into the cost model to train the cost model. During the training process, the theoretical optimal cost corresponding to the training data is labeled to force the model to learn and output a cost prediction value close to the theoretical optimal cost;
[0009] The component costs of the target sealing system are predicted using the trained cost model to obtain cost prediction values that meet the optimization conditions.
[0010] The beneficial effects of the present application are as follows: the present application collects multi-source heterogeneous data that affects the cost of sealing system components, pre-processes the collected multi-source heterogeneous data; inputs the pre-processed data into a cost model to train the cost model, wherein, during the training process, the theoretical optimal cost corresponding to the training data is annotated to force the model learning to output a cost prediction value close to the theoretical optimal cost; the component cost of the target sealing system is predicted by the trained cost model to obtain a cost prediction value that meets the optimization conditions. Since multi-source heterogeneous data can be obtained in real time, the cost influencing factors covered are more comprehensive, and are obtained and updated in real time, ensuring the comprehensiveness and real-time nature of the data. The cost model is then trained by the pre-processed multi-source heterogeneous data, and the cost of the sealing system components is predicted by the trained cost model, thereby improving the accuracy of the cost prediction of the sealing system components.
[0011] In one embodiment, collecting multi-source heterogeneous data that affects the cost of sealing system components includes:
[0012] Obtain product price data, raw material price data, production process parameter data, and historical cost data from the company's internal system;
[0013] Obtain supplier price details, market demand data, and industry trend data from external data sources;
[0014] Integrate the acquired multi-source data to form a unified data set.
[0015] In one embodiment, the preprocessing of the collected multi-source heterogeneous data includes:
[0016] A sliding window confidence interval strategy is used to identify abnormal data, and an LSTM model is used to detect time series correlation anomalies.
[0017] Implement graph model repair for missing values and infer the optimal interpolation value through the relationship between adjacent nodes;
[0018] Define a unified metadata model and modify data into a standard format;
[0019] Fuzzy matching algorithm aligns material codes from different data sources;
[0020] Key fusion features are extracted based on feature selection algorithms to eliminate redundant dimensions of multi-source data.
[0021] In one embodiment, inputting the pre-processed data into the cost model to train the cost model includes:
[0022] The optimal cost results corresponding to the training data are marked to force the model to screen the lowest-priced material among the available materials for the same component from the training data that meets the constraints, and to select the supplier with the lowest quotation for the same material, so that the prediction results are close to the theoretical optimal cost.
[0023] In one embodiment, predicting the component costs of the target sealing system using the trained cost model to obtain a cost prediction value that meets the optimization conditions includes:
[0024] Determine whether the optimized sealing system component costs meet the preset conditions;
[0025] When the optimized sealing system component cost meets the preset conditions, the current prediction value is determined to be the cost prediction value that meets the optimization conditions.
[0026] In one embodiment, the method further comprises:
[0027] When the optimized sealing system component cost does not meet the preset conditions, continue to collect more multi-source heterogeneous data that affect the sealing system component cost to train the cost model, and predict and optimize the cost based on the trained cost model.
[0028] In one embodiment, the method further comprises:
[0029] When it is detected that there is a target material with a price increase greater than a preset value, an analysis report is generated simultaneously with the output of the cost forecast value, and the key driving factors causing the price increase of the target material are displayed in the analysis report;
[0030] Generate decision recommendations based on the key drivers.
[0031] The present application also provides a sealing system component cost prediction device, comprising:
[0032] A collection module for collecting multi-source heterogeneous data that affects the cost of sealing system components;
[0033] Preprocessing module, used to preprocess the collected multi-source heterogeneous data;
[0034] A training module is used to input the preprocessed data into the cost model and train the cost model. During the training process, the theoretical optimal cost corresponding to the training data is annotated to force the model to learn and output a cost prediction value close to the theoretical optimal cost;
[0035] The prediction module is used to predict the component costs of the target sealing system through the trained cost model to obtain a cost prediction value that meets the optimization conditions.
[0036] In one embodiment, the collection module includes:
[0037] The first acquisition submodule is used to obtain product price data, raw material price data, production process parameter data, and historical cost data from the enterprise's internal system;
[0038] The second acquisition submodule is used to obtain supplier price details data, market demand data, and industry trend data from external data sources;
[0039] The integration submodule is used to integrate the acquired multi-source data to form a unified data set.
[0040] In one embodiment, the pre-processing module includes:
[0041] The identification submodule is used to identify abnormal data using a sliding window confidence interval strategy and to detect time series correlation anomalies in combination with the LSTM model;
[0042] The repair submodule is used to implement graph model repair for missing values and infer the optimal interpolation value through the relationship between adjacent nodes;
[0043] The modification submodule is used to define a unified metadata model and modify the data into a standard format;
[0044] Alignment submodule, used for fuzzy matching algorithm to align material codes from different data sources;
[0045] The extraction submodule is used to extract key fusion features based on feature selection algorithm and eliminate redundant dimensions of multi-source data.
[0046] In one embodiment, the training module is further configured to:
[0047] The optimal cost results corresponding to the training data are marked to force the model to screen the lowest-priced material among the available materials for the same component from the training data that meets the constraints, and to select the supplier with the lowest quotation for the same material, so that the prediction results are close to the theoretical optimal cost.
[0048] In one embodiment, the prediction module includes:
[0049] A judgment submodule is used to judge whether the optimized sealing system component cost meets the preset conditions;
[0050] The determination submodule is used to determine that the current prediction value is a cost prediction value that meets the optimization condition when the optimized sealing system component cost meets the preset condition.
[0051] In one embodiment, the prediction module is further configured to:
[0052] When the optimized sealing system component cost does not meet the preset conditions, continue to collect more multi-source heterogeneous data that affect the sealing system component cost to train the cost model, and predict and optimize the cost based on the trained cost model.
[0053] In one embodiment, the apparatus further comprises:
[0054] A generation module is configured to, when detecting a target material with a price increase greater than a preset value, simultaneously generate an analysis report when outputting a cost forecast value, and display in the analysis report the key driving factors causing the price increase of the target material;
[0055] The generation module is further used to generate decision recommendations based on the key driving factors.
[0056] The present application provides a sealing system component cost prediction system, comprising:
[0057] at least one processor; and,
[0058] a memory communicatively connected to the at least one processor; wherein,
[0059] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to implement the sealing system component cost prediction method described in any one of the above embodiments.
[0060] The present application provides a computer-readable storage medium. When the instructions in the storage medium are executed by a processor corresponding to a sealing system component cost prediction system, the sealing system component cost prediction system can implement the sealing system component cost prediction method described in any of the above embodiments.
[0061] Other features and advantages of the present application will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present application. The purposes and other advantages of the present application can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.
[0062] The technical solution of the present application is further described in detail below through the accompanying drawings and examples. BRIEF DESCRIPTION OF THE DRAWINGS
[0063] The accompanying drawings are used to provide a further understanding of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings:
[0064] Figure 1 This is a flow chart of a sealing system component cost prediction method in one embodiment of the present application;
[0065] Figure 2 This is a schematic structural diagram of a sealing system component cost prediction device in one embodiment of the present application;
[0066] Figure 3 Schematic diagram of the hardware structure of a sealing system component cost prediction system in one embodiment of the present application. DETAILED DESCRIPTION
[0067] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application and are not used to limit the present application.
[0068] This application provides a cost prediction method for sealing system components based on artificial intelligence technology. By integrating multi-source heterogeneous data, comprehensively considering various factors affecting cost, and improving system accuracy through multi-sample data training, it also introduces intelligent optimization algorithms to achieve dynamic cost optimization, thereby effectively reducing the production cost of sealing system components and improving the economic benefits of enterprises.
[0069] Figure 1 This is a flow chart of a sealing system component cost prediction method in one embodiment of the present application. Figure 1 As shown, the method can be implemented as the following steps S101-S104:
[0070] In step S101 , multi-source heterogeneous data that affects the cost of sealing system components is collected;
[0071] In step S102, the collected multi-source heterogeneous data is preprocessed;
[0072] In step S103, the pre-processed data is input into the cost model to train the cost model. During the training process, the theoretical optimal cost corresponding to the training data is marked to force the model to learn and output a cost prediction value close to the theoretical optimal cost.
[0073] In step S104, the component costs of the target sealing system are predicted using the trained cost model to obtain a cost prediction value that meets the optimization conditions.
[0074] In this application, we first collected multi-source heterogeneous data that affects the cost of sealing system components to ensure comprehensiveness and diversity. By integrating production data (such as bill of materials and process parameters), supply chain data (purchase price, logistics time), and market data (demand forecasts and price fluctuations), we established a cross-system data channel to ensure real-time synchronization of data from ERP, MES, and other systems.
[0075] When collecting heterogeneous, multi-source data that impacts sealing system component costs, we not only source data such as product prices, raw material prices, production process parameters, and historical costs from internal enterprise systems, but also gather supplier price details, market demand, and industry trends from external data sources. By integrating this multi-source data, we create a unified dataset, providing comprehensive data support for subsequent data preprocessing and model training.
[0076] Raw material price data includes the prices and fluctuations of raw materials such as rubber, fillers, and additives. Specific indicators include the prices of different products through various channels in each period, as well as year-on-year and month-on-month data. Production process parameter data includes process parameters such as extrusion temperature, vulcanization time, and vulcanization temperature. Historical cost data includes the products and production costs of sealing system components for each historical vehicle model. Market demand data includes the demand for sealing system components in different industries, demand trends, and the proportion of demand in specific segments. Industry trend data includes industry policies and technological development trends, such as industry standards, high-performance material penetration rates, and alternative material cost curves. Data on market demand and industry trends can be obtained through industry research reports and other sources. Furthermore, market demand can be predicted by pre-building a market demand forecast model based on sales volume and price fluctuations, or by building a market demand forecast model based on historical data.
[0077] For the multi-source heterogeneous data, full-process monitoring and aggregation are performed. Taking the price flow process as an example, the primary task is to connect the price data from all processes. On the one hand, real-time price data is collected to ensure the timeliness and accuracy of the data. For example, from the internal system perspective, the company's ERP (Enterprise Resource Planning), SCM (Supply Chain Management), CRM (Customer Relationship Management) and other systems are connected to obtain internal data such as product prices, raw material prices, production process parameters, and historical costs. From the external data perspective, supplier systems, market data platforms, industry reports, etc. are connected to obtain external data such as supplier price details, market price fluctuations, and industry trends. On the other hand, an anomaly detection mechanism is established to monitor abnormal fluctuations in price data in real time. For example, based on historical data, the moving average and standard deviation are calculated, and 3 times the standard deviation is used as the threshold for anomaly identification. When the price fluctuation exceeds the moving average ± 3 times the standard deviation, the price data is considered abnormal. Of course, a range can also be set based on year-on-year and month-on-month comparisons. When the set range is exceeded, the price data is considered abnormal. This can then issue a timely warning for abnormal fluctuations in the price data itself, or process it according to a preset outlier processing method, such as outlier removal or replacement with the mean.
[0078] Secondly, the collected multi-source heterogeneous data is preprocessed, including data cleaning, data standardization, and data fusion, to form a unified data format. For abnormal data identification and processing, a sliding window confidence interval strategy is used to identify abnormal data, combined with the LSTM model to detect time series correlation anomalies to ensure data accuracy and reliability. For missing values, a graph model is used to repair the optimal interpolation value through the relationship between adjacent nodes to fill the data gaps. For data standardization, a unified metadata model is defined to modify the data into a standard format to facilitate subsequent processing and analysis. For example, numerical data such as raw material prices and production costs are converted to standardized values between 0 and 1. For material code alignment, a fuzzy matching algorithm is used to align material codes from different data sources to ensure data consistency and comparability. Finally, a feature selection algorithm is used to extract key fusion features, eliminate redundant dimensions of multi-source data, and improve data quality and efficiency. For example, principal component analysis (PCA) can be used to reduce the dimensionality of 12 process parameters. The first three principal components retain 85% of the information, and the remaining variables can be eliminated. Canonical correlation analysis (CCA) can also be performed. If a potential correlation dimension between supplier prices and logistics timeliness is discovered, a "supply stability" fusion feature can be formed. Of course, unstructured industry reports can also be extracted through deep learning features.
[0079] Then, the preprocessed data is input into the cost model to train the cost model. During the training process, the theoretical optimal cost corresponding to the training data is labeled to force the model to learn and output a cost prediction value close to the theoretical optimal cost. Specifically, the optimal cost results corresponding to the training data are first labeled to guide the model to learn and output a cost prediction value close to the optimal cost, such as screening the lowest-priced material among the available materials for the same component, and selecting the supplier with the lowest quotation for the same material. At the same time, core optimization indicators are determined, such as minimizing total cost and maximizing unit cost profit margin, to provide a clear goal for model training. Next, constraints are set, including hard constraints (such as the lower limit of material strength) and flexible constraints (such as yield rate fluctuation tolerance) to ensure that model training meets actual production needs and constraints. Furthermore, by introducing algorithm evaluation indicators for optimization, the accuracy and generalization ability of the system are further improved to ensure the effectiveness and reliability of the model in practical applications.
[0080] Finally, the trained cost model is used to predict the component costs of the target sealing system to obtain a cost prediction value that meets the optimization conditions. A determination is made as to whether the optimized sealing system component costs meet preset conditions, such as a cost reduction greater than 10% and a prediction accuracy greater than 90%. When the optimized sealing system component costs meet the preset conditions, the current prediction value is determined to be the cost prediction value that meets the optimization conditions. If the optimized sealing system component costs do not meet the preset conditions, further multi-source heterogeneous data that affects the sealing system component costs is collected to train the cost model, and cost prediction and optimization are performed based on the trained cost model.
[0081] Furthermore, in one embodiment of the present application, when a target material is detected to have a price increase greater than a preset value, an analysis report is generated simultaneously with the output of the cost forecast. The report displays the key drivers of the target material price increase, such as market supply and demand, policy changes, and raw material costs. Based on these key drivers, decision-making recommendations are generated, such as adjusting procurement strategies, finding alternative materials, and optimizing production processes. These recommendations help companies address the challenges posed by rising material prices, reduce production costs, and improve economic efficiency.
[0082] For example, an automotive parts company implemented an intelligent procurement system to predict and analyze the procurement costs of aluminum alloy wheels in Q2 2024.
[0083] Summary of forecast results
[0084] Estimated unit price: ￥186.50 / kg (confidence interval ±3.2%);
[0085] Change from previous quarter: +5.8% (mainly due to the increase in LME aluminum prices).
[0086] Key driving factor analysis
[0087] The London aluminum futures price rose by 42%, which was directly transmitted to the raw material cost, with the shipping cost index rising by 23%; the Southeast Asian route rate rose by 18%; the process improvement benefit was -10%, and the new mold reduced the scrap rate by 2.3%; the exchange rate fluctuated by 8%, and the RMB depreciated by 1.5% against the US dollar.
[0088] Decision-making recommendations
[0089] Short-term strategy: It is recommended to lock in 3-month futures when aluminum price is below $2200 / ton;
[0090] Long-term strategy: Develop second-tier suppliers (the qualification rate of the two currently under cultivation has reached 92%);
[0091] Risk warning: Pay attention to the impact of the closure of the Red Sea shipping channel on freight rates in Europe.
[0092] Through the method provided in this application, not only can the cost system be used to predict the cost of sealing system components of new models; it can also be used to predict new costs based on the sealing system component product parameters provided by the new model, such as the sealing system component length, weight, material and other information, after inputting them into the system; for ECRs issued for engineering changes, the estimated engineering price can also be analyzed; at the same time, the price of existing products can be obtained based on the current market situation.
[0093] The beneficial effects of the present application are as follows: the present application collects multi-source heterogeneous data that affects the cost of sealing system components, pre-processes the collected multi-source heterogeneous data; inputs the pre-processed data into a cost model to train the cost model, wherein, during the training process, the theoretical optimal cost corresponding to the training data is annotated to force the model learning to output a cost prediction value close to the theoretical optimal cost; the component cost of the target sealing system is predicted by the trained cost model to obtain a cost prediction value that meets the optimization conditions. Since multi-source heterogeneous data can be obtained in real time, the cost influencing factors covered are more comprehensive, and are obtained and updated in real time, ensuring the comprehensiveness and real-time nature of the data. The cost model is then trained by the pre-processed multi-source heterogeneous data, and the cost of the sealing system components is predicted by the trained cost model, thereby improving the accuracy of the cost prediction of the sealing system components.
[0094] In one embodiment, the above step S101 may be implemented as the following steps A1-A3:
[0095] In step A1, product price data, raw material price data, production process parameter data, and historical cost data are obtained from the enterprise's internal system;
[0096] In step A2, supplier price details, market demand data, and industry trend data are obtained from external data sources;
[0097] In step A3, the acquired multi-source data are integrated to form a unified data set.
[0098] In one embodiment, the above step S102 may be implemented as the following steps B1-B5:
[0099] In step B1, a sliding window confidence interval strategy is used to identify abnormal data, and an LSTM model is used to detect time series correlation anomalies;
[0100] In step B2, the missing values are repaired using a graphical model, and the optimal interpolation value is inferred through the relationship between adjacent nodes.
[0101] In step B3, a unified metadata model is defined to modify the data into a standard format;
[0102] In step B4, a fuzzy matching algorithm aligns the material codes from different data sources;
[0103] In step B5, key fusion features are extracted based on the feature selection algorithm to eliminate redundant dimensions of multi-source data.
[0104] In one embodiment, the above step S103 may be implemented as follows:
[0105] The optimal cost results corresponding to the training data are marked to force the model to screen the lowest-priced material among the available materials for the same component from the training data that meets the constraints, and to select the supplier with the lowest quotation for the same material, so that the prediction results are close to the theoretical optimal cost.
[0106] In one embodiment, the above step S103 may be implemented as the following steps C1-C2:
[0107] In step C1, it is determined whether the optimized sealing system component cost meets the preset conditions;
[0108] In step C2, when the optimized sealing system component cost meets the preset condition, the current prediction value is determined to be a cost prediction value that meets the optimization condition.
[0109] In one embodiment, the method may also be implemented as follows:
[0110] When the optimized sealing system component cost does not meet the preset conditions, continue to collect more multi-source heterogeneous data that affect the sealing system component cost to train the cost model, and predict and optimize the cost based on the trained cost model.
[0111] In one embodiment, the method can also be implemented as the following steps D1-D2:
[0112] In step D1, when it is detected that there is a target material whose price has increased and the increase is greater than a preset value, an analysis report is generated simultaneously with the output of the cost forecast value, and the key driving factors causing the increase in the target material price are displayed in the analysis report;
[0113] In step D2, decision recommendations are generated based on the key driving factors.
[0114] Figure 2 FIG. 1 is a structural diagram of a sealing system component cost prediction device according to an embodiment of the present application. Figure 2 As shown, the device includes:
[0115] A collection module 201 is used to collect multi-source heterogeneous data that affects the cost of sealing system components;
[0116] A preprocessing module 202 is used to preprocess the collected multi-source heterogeneous data;
[0117] A training module 203 is configured to input the pre-processed data into the cost model and train the cost model. During the training process, the theoretical optimal cost corresponding to the training data is annotated to force the model to learn and output a cost prediction value close to the theoretical optimal cost.
[0118] The prediction module 204 is configured to predict the component costs of the target sealing system using the trained cost model to obtain a cost prediction value that meets the optimization conditions.
[0119] In one embodiment, the collection module includes:
[0120] The first acquisition submodule is used to obtain product price data, raw material price data, production process parameter data, and historical cost data from the enterprise's internal system;
[0121] The second acquisition submodule is used to obtain supplier price details data, market demand data, and industry trend data from external data sources;
[0122] The integration submodule is used to integrate the acquired multi-source data to form a unified data set.
[0123] In one embodiment, the pre-processing module includes:
[0124] The identification submodule is used to identify abnormal data using a sliding window confidence interval strategy and to detect time series correlation anomalies in combination with the LSTM model;
[0125] The repair submodule is used to implement graph model repair for missing values and infer the optimal interpolation value through the relationship between adjacent nodes;
[0126] The modification submodule is used to define a unified metadata model and modify the data into a standard format;
[0127] Alignment submodule, used for fuzzy matching algorithm to align material codes from different data sources;
[0128] The extraction submodule is used to extract key fusion features based on feature selection algorithm and eliminate redundant dimensions of multi-source data.
[0129] In one embodiment, the training module is further configured to:
[0130] The optimal cost results corresponding to the training data are marked to force the model to screen the lowest-priced material among the available materials for the same component from the training data that meets the constraints, and to select the supplier with the lowest quotation for the same material, so that the prediction results are close to the theoretical optimal cost.
[0131] In one embodiment, the prediction module includes:
[0132] A judgment submodule is used to judge whether the optimized sealing system component cost meets the preset conditions;
[0133] The determination submodule is used to determine that the current prediction value is a cost prediction value that meets the optimization condition when the optimized sealing system component cost meets the preset condition.
[0134] In one embodiment, the prediction module is further configured to:
[0135] When the optimized sealing system component cost does not meet the preset conditions, continue to collect more multi-source heterogeneous data that affect the sealing system component cost to train the cost model, and predict and optimize the cost based on the trained cost model.
[0136] In one embodiment, the apparatus further comprises:
[0137] A generation module is configured to, when detecting a target material with a price increase greater than a preset value, simultaneously generate an analysis report when outputting a cost forecast value, and display in the analysis report the key driving factors causing the price increase of the target material;
[0138] The generation module is further used to generate decision recommendations based on the key driving factors.
[0139] Figure 3 FIG. 1 is a schematic diagram of the hardware structure of a sealing system component cost prediction system in one embodiment of the present application. Figure 3 As shown, the sealing system component cost prediction system includes:
[0140] at least one processor 320; and,
[0141] A memory 304 in communication with the at least one processor 320; wherein,
[0142] The memory 304 stores instructions that can be executed by the at least one processor 320 . The instructions are executed by the at least one processor 320 to implement the sealing system component cost prediction method described in any of the above embodiments.
[0143] Reference Figure 3 The sealing system component cost prediction system 300 may include one or more of the following components: a processing component 302 , a memory 304 , a power component 306 , a multimedia component 308 , an audio component 310 , an input / output (I / O) interface 312 , a sensor component 314 , and a communication component 316 .
[0144] The processing component 302 generally controls the overall operation of the sealing system component cost prediction system 300. The processing component 302 may include one or more processors 320 to execute instructions to perform all or part of the steps of the method described above. Furthermore, the processing component 302 may include one or more modules to facilitate interaction between the processing component 302 and other components. For example, the processing component 302 may include a multimedia module to facilitate interaction between the multimedia component 308 and the processing component 302.
[0145] The memory 304 is configured to store various types of data to support the operation of the sealing system component cost prediction system 300. Examples of such data include instructions for any application or method operating on the sealing system component cost prediction system 300, such as text, images, videos, etc. The memory 304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.
[0146] The power supply component 306 provides power to the various components of the sealing system component cost prediction system 300. The power supply component 306 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the sealing system component cost prediction system 300.
[0147] The multimedia component 308 includes a screen that provides an output interface between the sealing system component cost prediction system 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor can not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 308 may also include a front camera and / or a rear camera. When the sealing system component cost prediction system 300 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have a focal length and optical zoom capability.
[0148] The audio component 310 is configured to output and / or input audio signals. For example, the audio component 310 includes a microphone (MIC) that is configured to receive external audio signals when the sealing system component cost prediction system 300 is in an operating mode, such as an alarm mode, a recording mode, a voice recognition mode, or a voice output mode. The received audio signals may be further stored in the memory 304 or transmitted via the communication component 316. In some embodiments, the audio component 310 also includes a speaker for outputting audio signals.
[0149] I / O interface 312 provides an interface between processing component 302 and peripheral interface modules, such as a keyboard, click wheel, buttons, etc. These buttons may include but are not limited to: a home button, volume buttons, a start button, and a lock button.
[0150] Sensor assembly 314 includes one or more sensors for providing various status assessments of sealing system component cost prediction system 300. For example, sensor assembly 314 may include an acoustic sensor. Additionally, sensor assembly 314 may detect the open / closed state of sealing system component cost prediction system 300, the relative positioning of components, such as the display and keypad of sealing system component cost prediction system 300, and the operating state of sealing system component cost prediction system 300 or a component thereof, such as the operating state of an air distribution plate, the structural state, the operating state of a discharge scraper, etc., the orientation or acceleration / deceleration of sealing system component cost prediction system 300, and temperature changes of sealing system component cost prediction system 300. Sensor assembly 314 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. Sensor assembly 314 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 314 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, a material stack thickness sensor, or a temperature sensor.
[0151] The communication component 316 is configured to enable the sealing system component cost prediction system 300 to provide the ability to communicate with other devices and cloud platforms in a wired or wireless manner. The sealing system component cost prediction system 300 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 316 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 316 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.
[0152] In an exemplary embodiment, the sealing system component cost prediction system 300 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to execute the sealing system component cost prediction method described in any of the above embodiments.
[0153] The present application provides a computer-readable storage medium. When the instructions in the storage medium are executed by a processor corresponding to a sealing system component cost prediction system, the sealing system component cost prediction system can implement the sealing system component cost prediction method described in any of the above embodiments.
[0154] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) that contain computer-usable program code.
[0155] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0156] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0157] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.
[0158] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.
Claims
1. A sealing system component cost prediction method, characterized in that: include: Collect multi-source heterogeneous data that impacts sealing system component costs; Preprocess the collected multi-source heterogeneous data; The preprocessed data is input into the cost model to train the cost model. During the training process, the theoretical optimal cost corresponding to the training data is labeled to force the model to learn and output a cost prediction value close to the theoretical optimal cost; The component costs of the target sealing system are predicted using the trained cost model to obtain cost prediction values that meet the optimization conditions.
2. The method according to claim 1, wherein The method collects multi-source heterogeneous data that affects the cost of sealing system components, including: Obtain product price data, raw material price data, production process parameter data, and historical cost data from the company's internal system; Obtain supplier price details, market demand data, and industry trend data from external data sources; Integrate the acquired multi-source data to form a unified data set.
3. The method according to claim 1, wherein The preprocessing of the collected multi-source heterogeneous data includes: A sliding window confidence interval strategy is used to identify abnormal data, and an LSTM model is used to detect time series correlation anomalies. Implement graph model repair for missing values and infer the optimal interpolation value through the relationship between adjacent nodes; Define a unified metadata model and modify data into a standard format; Fuzzy matching algorithm aligns material codes from different data sources; Key fusion features are extracted based on feature selection algorithms to eliminate redundant dimensions of multi-source data.
4. The method according to claim 1, wherein The pre-processed data is input into the cost model to train the cost model, including: The optimal cost results corresponding to the training data are marked to force the model to screen the lowest-priced material among the available materials for the same component from the training data that meets the constraints, and to select the supplier with the lowest quotation for the same material, so that the prediction results are close to the theoretical optimal cost.
5. The method according to claim 1, wherein The component costs of the target sealing system are predicted using the trained cost model to obtain a cost prediction value that meets the optimization conditions, including: Determine whether the optimized sealing system component costs meet the preset conditions; When the optimized sealing system component cost meets the preset conditions, the current prediction value is determined to be the cost prediction value that meets the optimization conditions.
6. The method according to claim 5, wherein The method further comprises: When the optimized sealing system component cost does not meet the preset conditions, continue to collect more multi-source heterogeneous data that affect the sealing system component cost to train the cost model, and predict and optimize the cost based on the trained cost model.
7. The method according to claim 1, wherein The method further comprises: When it is detected that there is a target material with a price increase greater than a preset value, an analysis report is generated simultaneously with the output of the cost forecast value, and the key driving factors causing the price increase of the target material are displayed in the analysis report; Generate decision recommendations based on the key drivers.
8. A sealing system component cost prediction device, characterized in that: include: A collection module for collecting multi-source heterogeneous data that affects the cost of sealing system components; Preprocessing module, used to preprocess the collected multi-source heterogeneous data; A training module is used to input the preprocessed data into the cost model and train the cost model. During the training process, the theoretical optimal cost corresponding to the training data is annotated to force the model to learn and output a cost prediction value close to the theoretical optimal cost; The prediction module is used to predict the component costs of the target sealing system through the trained cost model to obtain a cost prediction value that meets the optimization conditions.
9. A sealing system component cost prediction system, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to implement the sealing system component cost prediction method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that When the instructions in the storage medium are executed by a processor corresponding to the sealing system component cost prediction system, the sealing system component cost prediction system can implement the sealing system component cost prediction method according to any one of claims 1 to 7.