Product life cycle management method and system based on artificial intelligence assistance

Abstract

The invention relates to the field of data processing, in particular to a product life cycle management method and system based on artificial intelligence assistance, and the method comprises the steps: starting a life cycle management system through a product optimization instruction, and recognizing a to-be-optimized product, extracting multiple pieces of historical archiving data which belong to the same product line as the to-be-optimized product from the historical data archiving unit, obtaining product line optimization characteristics by utilizing the multiple pieces of historical archiving data, obtaining market guiding characteristics, obtaining an optimization simulation strategy by utilizing the intelligent decision-making unit, the market guiding characteristics and the product line optimization characteristics, and obtaining a product line optimization result; and performing optimization simulation on the optimization simulation strategy to obtain an evaluation result, and obtaining an optimized product by using the optimization simulation strategy received by the Internet of Things monitoring unit to complete product life cycle management based on artificial intelligence assistance. According to the invention, product line historical data and market oriented features are integrated, an optimization simulation strategy is generated by using the intelligent decision-making unit, product optimization is carried out after simulation verification, and product full-period management is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a product life cycle management method and system based on artificial intelligence assistance. BACKGROUND

[0002] Product life cycle management is a core management system for realizing resource optimization and decision-making collaboration by integrating product data and processes from concept design, production and manufacturing, market launch to withdrawal and recycling.

[0003] Traditional life cycle management stores various design parameters, production indicators, and user feedback, and then professional personnel analyzes market reports and formulates optimization direction strategies.

[0004] Although the above method can realize the whole life cycle management of the product, artificial intelligence technology is not applied to the whole process of management, so a product life cycle management method based on artificial intelligence assistance is needed to assist R&D personnel in integrating product line historical data and market-oriented features, generate optimization simulation strategies using an intelligent decision unit, and optimize products after simulation verification, thereby realizing product life cycle management. SUMMARY

[0005] The present application provides a product life cycle management method based on artificial intelligence assistance and a computer readable storage medium, which mainly integrates product line historical data and market-oriented features, generates optimization simulation strategies using an intelligent decision unit, optimizes products after simulation verification, and realizes product life cycle management.

[0006] To achieve the above purpose, the present application provides a product life cycle management method based on artificial intelligence assistance, which comprises:

[0007] Receiving a product optimization instruction, and starting a life cycle management system using the product optimization instruction, wherein the life cycle management system comprises an intelligent decision unit, a historical data archiving unit, an Internet of Things monitoring unit, a simulation unit and a market monitoring unit;

[0008] Identifying a product to be optimized based on the product optimization instruction, and extracting a plurality of historical archived data belonging to the same product line as the product to be optimized from the historical data archiving unit based on the product to be optimized;

[0009] Evaluating a plurality of historical archived data using a pre-constructed evaluation method to obtain a plurality of product evaluation value groups, and obtaining product line optimization features based on the plurality of product evaluation value groups;

[0010] Obtaining a plurality of effective competitors of the product to be optimized using a market monitoring unit, extracting market optimization features of the plurality of effective competitors, and obtaining market-oriented features;

[0011] The market-oriented features and product line optimization features are analyzed by using an intelligent decision unit to obtain a plurality of optimization demand data, and according to the plurality of optimization demand data, associated node data retrieval is performed in a pre-constructed product configuration framework to obtain a plurality of associated nodes;

[0012] An optimization simulation strategy for the product to be optimized is obtained based on the plurality of associated nodes, and optimization simulation is performed on the product to be optimized based on a simulation unit and the optimization simulation strategy to obtain simulation data;

[0013] The simulation data is evaluated to obtain an evaluation result, and if the evaluation result meets the preset production optimization demand, the optimization simulation strategy is sent to the Internet of Things monitoring unit, and the optimized product is obtained by using the optimization simulation strategy received by the Internet of Things monitoring unit, thereby completing the product life cycle management assisted by artificial intelligence.

[0014] Optionally, the plurality of historical archived data of the product to be optimized belonging to the same product line are extracted from the historical data archiving unit based on the product to be optimized, and the method further comprises the following steps:

[0015] The target product line of the product to be optimized is confirmed, a plurality of historical products are obtained based on the target product line, and the following operations are performed on each historical product in the plurality of historical products:

[0016] The production data set of the historical product is received, and the following operations are performed on each production data in the production data set:

[0017] The modal type of the production data is identified, wherein the modal type includes structured data and unstructured data, the structured data is a numerical data type, and the unstructured data includes image data type, audio and video data type, and text data type;

[0018] According to the modal type, a pre-processing method tag corresponding to the modal type is called from a pre-constructed pre-processing algorithm library, and a pre-processing operation tag is used to identify the production data, thereby obtaining identified production data;

[0019] The identified production data is summarized based on the modal type to obtain a plurality of same type production data groups corresponding to the production data set, wherein the same type production data group includes a plurality of same type production data, and the modal type of the plurality of same type production data is the same;

[0020] Vectorization and parallel pre-processing are performed on the plurality of same type production data groups to obtain a plurality of pre-processed same type production data groups, and the plurality of pre-processed same type production data groups are summarized to obtain historical production data;

[0021] The market data set is received, and historical market data is obtained based on the market data set;

[0022] The initial historical archiving data is obtained by aggregating historical production data and historical market data, and the historical archiving data is obtained by performing archiving identification based on historical products on the initial historical archiving data, and the historical archiving data is archived into the historical data archiving unit based on a product configuration framework.

[0023] Optionally, the vectorized parallel preprocessing is performed on the plurality of homogeneous production data sets to obtain a plurality of preprocessed homogeneous production data sets, including:

[0024] The following operations are performed on each of the plurality of homogeneous production data sets:

[0025] A parallel processing data vector is constructed based on the homogeneous production data set, and a preprocessing method instruction sequence corresponding to a preprocessing method label is received, wherein the parallel processing data vector includes a plurality of parallel processing data, and the parallel processing data correspond one-to-one to homogeneous production data in the homogeneous production data set, and the preprocessing method instruction sequence includes a plurality of preprocessing instructions, and the plurality of preprocessing instructions are sorted according to the order of actual preprocessing operations;

[0026] The preprocessing instructions are sequentially extracted from the plurality of preprocessing instructions, and the parallel processing data vector is processed using the extracted preprocessing instructions to obtain an intermediate processing data vector, and the intermediate processing data vector is used as the parallel processing data vector, and the step of sequentially extracting preprocessing instructions from the plurality of preprocessing instructions is returned until the preprocessing instructions in the preprocessing method instruction sequence are extracted, to obtain a preprocessed homogeneous production data set, wherein the preprocessed homogeneous production data in the preprocessed homogeneous production data set corresponds one-to-one to the homogeneous production data;

[0027] The preprocessed homogeneous production data sets are aggregated to obtain a plurality of preprocessed homogeneous production data sets.

[0028] Optionally, the plurality of historical archiving data is evaluated using a pre-constructed evaluation method to obtain a plurality of product evaluation value sets, and product line optimization features are obtained based on the plurality of product evaluation value sets, and the foregoing further includes:

[0029] The following operations are performed on the historical archiving data sequentially extracted from the plurality of historical archiving data:

[0030] The following operations are performed on the historical archiving data sequentially extracted from the plurality of historical archiving data:

[0031] If the to-be-evaluated data is not structured data, a non-structured data numericalization operation based on multi-feature analysis is performed on the to-be-evaluated data to obtain a numerical evaluation numerical value group;

[0032] The numerical evaluation numerical value and the numerical evaluation numerical value group are summarized to obtain an initial evaluation numerical value set of the historical archive data;

[0033] Based on the plurality of historical archive data, a full data category is obtained, a missing category set in the initial evaluation numerical value set is identified based on the full data category, and a zero filling operation is performed on the initial evaluation numerical value set based on the missing category set to obtain a numerical evaluation set;

[0034] The numerical evaluation set is summarized to obtain an original evaluation numerical value group set of the plurality of historical archive data, a normalization operation based on the same feature is performed on the original evaluation numerical value group set to obtain a normalized evaluation group set, and an objective value evaluation is performed on the normalized evaluation group set by using a pre-constructed entropy weight method to obtain a plurality of product evaluation values;

[0035] A clustering analysis based on three clustering clusters is performed on the plurality of product evaluation values by using a pre-constructed clustering method to obtain a plurality of evaluation clustering clusters, wherein the plurality of evaluation clustering clusters include a high evaluation value cluster, a medium evaluation value cluster, and a low evaluation value cluster;

[0036] The high evaluation value cluster is extracted from the plurality of evaluation clustering clusters, and a difference feature extraction based on the medium evaluation value cluster and the low evaluation value cluster is performed on the high evaluation value cluster by using an intelligent decision unit to obtain a product line optimization feature.

[0037] Optionally, the non-structured data numericalization operation based on multi-feature analysis on the to-be-evaluated data to obtain the numerical evaluation numerical value group includes:

[0038] Identifying a same type data set of the to-be-evaluated data, wherein the same type data set does not include the to-be-evaluated data, and the same type data in the same type data set and the to-be-evaluated data come from different historical products;

[0039] If the same type data set is an empty set, the to-be-evaluated data is set to 1 to obtain the numerical evaluation numerical value group;

[0040] If the same type data set is not an empty set, a plurality of unit features that can be numerically evaluated are parsed from the to-be-evaluated data to obtain a unit feature group, and a numerical evaluation numerical value group of the unit feature group is obtained.

[0041] Optionally, the parsing of the market-oriented feature and the product line optimization feature by using the intelligent decision unit to obtain the plurality of optimization demand data further includes:

[0042] The historical products are sequentially extracted from the target product line to obtain a target product, and the following operations are performed on the target product:

[0043] identify a plurality of product components of the target product, sequentially extract product components from the plurality of product components, obtain unit component configurations of the extracted product components, and obtain a plurality of unit component configurations corresponding to the plurality of product components, wherein the product components and the unit component configurations are one-to-one corresponding, the unit component configurations include a plurality of batch product components with different component batch codes, and the batch product components include one or more unit part configurations, and the unit part configurations include a plurality of batch product parts with different part batch codes;

[0044] construct a part-level structure tree according to the unit part configuration, aggregate the part-level structure trees corresponding to the plurality of batch product components according to the unit component configuration, and obtain a part-level structure tree set, wherein the part-level structure tree set includes one or more part-level structure trees, and construct a component-level structure tree according to the target product, the part-level structure tree set, and the plurality of product components;

[0045] aggregate the component-level structure tree to obtain an initial configuration framework corresponding to the target product line;

[0046] perform a node framework expansion operation on the initial configuration framework to obtain a suboptimal configuration framework, and perform a point position label identification operation on the suboptimal configuration framework to obtain a product configuration framework.

[0047] Optionally, the node framework expansion operation on the initial configuration framework to obtain a suboptimal configuration framework, and the point position label identification operation on the suboptimal configuration framework to obtain a product configuration framework, include:

[0048] confirm a plurality of configuration nodes in the initial configuration framework, and perform the following operations on each configuration node in the plurality of configuration nodes:

[0049] construct a data category list based on the configuration node, wherein the data category list includes a plurality of data categories, and the data categories include demand data, manufacturing data, experimental data, and maintenance data;

[0050] sequentially extract data categories from the plurality of data categories, and construct a plurality of unit data trees based on the extracted data categories, wherein the unit data trees sequentially include structured data nodes, extension path links, and non-structured extension nodes, and the extension path links connect the structured data nodes and the non-structured extension nodes;

[0051] perform a node linking operation between the plurality of unit data trees and the configuration nodes to obtain a data category tree corresponding to the configuration node;

[0052] aggregate the data category trees to obtain a suboptimal configuration framework, perform a partition encryption operation on the suboptimal configuration framework based on access permissions to obtain an encrypted configuration framework;

[0053] identify a plurality of input value sites in the encryption configuration framework, and identify a label path of each of the plurality of input value sites in the encryption configuration framework to obtain a plurality of label paths, and perform input value site pairing on the plurality of label paths to obtain a product configuration framework.

[0054] Optionally, the market monitoring unit obtains a plurality of effective competitors of the product to be optimized, comprising:

[0055] Obtain a keyword set of the product to be optimized, and perform same-type product retrieval based on the keyword set to obtain a plurality of initial same-type products, wherein the initial same-type products are identified by uniform resource locators;

[0056] Perform ambiguity entity cleaning on the plurality of initial same-type products to obtain a plurality of cleaned same-type products;

[0057] Perform product screening operation based on compliance on the plurality of cleaned same-type products to obtain a plurality of initial competitors;

[0058] Obtain competitor evaluation indexes of the plurality of initial competitors according to the full data category, and obtain a plurality of high evaluation competitor values according to the competitor evaluation indexes, the evaluation method and the clustering method, and confirm the plurality of initial competitors corresponding to the plurality of high evaluation competitor values as the plurality of effective competitors.

[0059] Optionally, the market optimization feature extraction is performed on the plurality of effective competitors to obtain market-oriented features, comprising:

[0060] Sequentially extract effective competitors from the plurality of effective competitors, and perform the following operations on the extracted effective competitors:

[0061] Extract a plurality of component structure feature groups according to the product configuration framework, wherein the component structure feature groups correspond one-to-one to historical products;

[0062] Identify competitor structure feature groups of the effective competitors by using an intelligent decision unit, and obtain structure differences between the competitor structure feature groups and each of the plurality of component structure feature groups to obtain a plurality of structure difference data;

[0063] Identify a plurality of function difference data of the effective competitors by using the intelligent decision unit;

[0064] Perform matching operation based on historical products on the plurality of structure difference data and the plurality of function difference data to obtain a plurality of initial market optimization feature groups, wherein the initial market optimization feature groups include structure difference data, function difference data and corresponding historical products;

[0065] Perform zero feature data deletion operation on the plurality of initial market optimization feature groups to obtain unit competitor optimization orientation data;

[0066] The competitive product optimization orientation data of the units is summarized to obtain market orientation characteristics of multiple effective competitive products.

[0067] To achieve the above-mentioned purpose, the application further provides a life cycle management system based on artificial intelligence assistance, comprising:

[0068] A historical data acquisition module is configured to receive a product optimization instruction and start the life cycle management system by using the product optimization instruction, wherein the life cycle management system comprises an intelligent decision unit, a historical data archiving unit, an Internet of Things monitoring unit, a simulation unit and a market monitoring unit;

[0069] Based on the product optimization instruction, a plurality of historical archiving data of the same product line as the to-be-optimized product are extracted from the historical data archiving unit;

[0070] A feature acquisition module is configured to evaluate the plurality of historical archiving data by using a pre-constructed evaluation method to obtain a plurality of product evaluation value groups, and obtain product line optimization features based on the plurality of product evaluation value groups;

[0071] A plurality of effective competitive products of the to-be-optimized product are obtained by using the market monitoring unit, market optimization features of the plurality of effective competitive products are extracted to obtain market orientation characteristics;

[0072] An associated node module is configured to analyze the market orientation characteristics and the product line optimization features by using the intelligent decision unit to obtain a plurality of optimization demand data, and perform associated node data retrieval in a pre-constructed product configuration framework according to the plurality of optimization demand data to obtain a plurality of associated nodes;

[0073] A post-processing module is configured to analyze the market orientation characteristics and the product line optimization features by using the intelligent decision unit to obtain a plurality of optimization demand data, and perform associated node data retrieval in a pre-constructed product configuration framework according to the plurality of optimization demand data to obtain a plurality of associated nodes;

[0074] An optimization simulation strategy of the to-be-optimized product is obtained based on the plurality of associated nodes, and the to-be-optimized product is optimized and simulated based on the simulation unit and the optimization simulation strategy to obtain simulation data;

[0075] The simulation data is evaluated to obtain an evaluation result, if the evaluation result meets a pre-set production optimization demand, the optimization simulation strategy is sent to the Internet of Things monitoring unit, and an optimized product is obtained by using the optimization simulation strategy received by the Internet of Things monitoring unit to complete the product life cycle management based on artificial intelligence assistance.

[0076] To solve the above-mentioned problems, the application further provides an electronic device, comprising:

[0077] a memory storing at least one instruction; and a processor executing the instruction stored in the memory to implement the above-mentioned artificial intelligence assisted product lifecycle management method.

[0078] To solve the above-mentioned problems, the present application further provides a computer readable storage medium, wherein at least one instruction is stored in the computer readable storage medium, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned artificial intelligence assisted product lifecycle management method.

[0079] To solve the problems described in the background art, the present application realizes intelligent optimization management of the whole process of a product by embedding artificial intelligence technology in a product lifecycle management system. The product lifecycle management system can monitor and collect real-time data of the whole process of product design, production and manufacturing, and market feedback, and analyze and process data of multiple modalities by using artificial intelligence algorithms. In addition, the product lifecycle management system can generate a direction for product optimization according to an evaluation method based on historical archival data and multiple effective competitive products. Finally, the product lifecycle management system can extract associated nodes in the product configuration framework according to the optimization direction, intelligently provide prompts for professional personnel to generate optimization simulation strategies, and finally simulate according to the optimization simulation strategies to obtain an optimized product. The present application integrates product line historical data and market-oriented characteristics, generates optimization simulation strategies by using an intelligent decision unit, optimizes the product after simulation verification, and realizes product lifecycle management. BRIEF DESCRIPTION OF DRAWINGS

[0080] Figure 1 A flowchart of the artificial intelligence assisted product lifecycle management method provided by an embodiment of the present application is shown.

[0081] Figure 2 A functional module diagram of the artificial intelligence assisted product lifecycle management system provided by an embodiment of the present application is shown.

[0082] Figure 3 A structural diagram of an electronic device for implementing the artificial intelligence assisted product lifecycle management method provided by an embodiment of the present application is shown.

[0083] Figure 4 A schematic diagram of a unit part configuration for implementing the artificial intelligence assisted product lifecycle management method provided by an embodiment of the present application is shown.

[0084] Figure 5 A schematic diagram of a component level structure tree for implementing the artificial intelligence assisted product lifecycle management method provided by an embodiment of the present application is shown.

[0085] REFERENCE SIGNS

[0086] 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0087] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0088] It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

[0089] An embodiment of the present application provides a product life cycle management method based on artificial intelligence assistance. An execution subject of the product life cycle management method based on artificial intelligence assistance includes but is not limited to at least one of electronic devices such as a server and a terminal, which can be configured to execute the method provided by the embodiment of the present application. In other words, the product life cycle management method based on artificial intelligence assistance can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to a single server, a server cluster, a cloud server or a cloud server cluster.

[0090] Referring to Figure 1 FIG. 1 shows a flowchart of a product life cycle management method based on artificial intelligence assistance provided by an embodiment of the present application. In the embodiment, the product life cycle management method based on artificial intelligence assistance includes the following steps.

[0091] S1, receiving a product optimization instruction, and starting a life cycle management system by using the product optimization instruction, wherein the life cycle management system includes an intelligent decision unit, a historical data archiving unit, an Internet of Things monitoring unit, a simulation unit and a market monitoring unit.

[0092] It can be understood that the present application realizes intelligent optimization management of the whole process of a product by embedding artificial intelligence technology in a product life cycle management system. The product life cycle management system can perform real-time data monitoring and collection on all links from product design, production and manufacturing to market feedback, and analyze and process data of multiple modalities by using an artificial intelligence algorithm. In addition, the product life cycle management system can generate a product optimization direction according to historical archiving data and multiple effective competitive products. Finally, the product life cycle management system can generate an optimization simulation strategy according to the optimization direction, and perform simulation according to the optimization simulation strategy, so as to obtain an optimized product.

[0093] It should be noted that the product optimization instruction refers to an instruction for optimizing a product to be optimized. For example, an existing product designer expects to find some optimization directions for optimizing the product to be optimized with the aid of artificial intelligence based on the data stored in the product lifecycle management system, and therefore initiates the product optimization instruction. The lifecycle management system is a system for managing products, and the lifecycle management system includes an intelligent decision unit, a historical data archiving unit, an Internet of Things monitoring unit, a simulation unit, and a market monitoring unit.

[0094] Further, the intelligent decision unit is a unit capable of providing artificial intelligence assistance for product optimization, and therefore includes various artificial intelligence assistance units, and different existing technologies are applied, including but not limited to natural language processing, convolutional neural networks, image processing, etc. The image processing includes but is not limited to edge extraction, noise reduction, etc. The edge extraction applies technologies including but not limited to Canny operator, etc. The noise reduction unit includes but is not limited to Gaussian filtering, etc. Each artificial intelligence assistance unit in the intelligent decision unit can be added or deleted according to actual needs.

[0095] Specifically, the historical data archiving unit is a unit for storing historical data of products based on a product configuration framework. The Internet of Things monitoring unit is a unit for tracking and monitoring components, parts, etc. of a product during production, for obtaining data of the product during the production process. The simulation unit is a unit integrating various simulation software, including but not limited to ansys, comsol, SolidWorks, etc. The specific simulation software can be added or deleted according to actual needs. After the optimization direction of the product to be optimized is determined, the simulation unit is used for simulation according to the optimization simulation strategy, so as to assist the research and development personnel in judging the pros and cons of the optimization direction. The market monitoring unit is a unit capable of reading product data from the network, including but not limited to reading data from the network by using Python, etc. The application of specific units is described in subsequent embodiments.

[0096] S2, identifying a product to be optimized based on the product optimization instruction, and extracting a plurality of historical archiving data of the same product line as the product to be optimized from the historical data archiving unit based on the product to be optimized.

[0097] It should be noted that the product to be optimized is a product that needs to be optimized. The process of identifying the product to be optimized based on the product optimization instruction includes: identifying the target product in the lifecycle management system based on the unique identifier embedded in the product optimization instruction by analyzing the product optimization instruction, and then confirming the optimization product to be optimized. The same product line refers to a product series having the same functional positioning, target user, etc.

[0098] For example, a manufacturer of a certain car releases an initial 1.0 version, then releases a 2.0 version by replacing the interior, and then releases a 3.0 version by replacing the engine. It is confirmed by the manufacturer that the 1.0 version, the 2.0 version and the 3.0 version constitute different iterative models of the same product line, and the plurality of historical archive data are historical data of the 1.0 version of the car, the 2.0 version of the car and the 3.0 version of the car.

[0099] Further, the extracting, from the historical data archiving unit, the plurality of historical archive data belonging to the same product line as the product to be optimized based on the product to be optimized further comprises:

[0100] Confirming a target product line of the product to be optimized, obtaining a plurality of historical products based on the target product line, and performing the following operations on each of the plurality of historical products:

[0101] Receiving a production data set of the historical product, and performing the following operations on each of the production data in the production data set:

[0102] Identifying the modal type of the production data, wherein the modal type includes structured data and unstructured data, the structured data is a numerical data type, and the unstructured data includes image data type, audio and video data type and text data type;

[0103] According to the modal type, a pre-processing method tag corresponding to the modal type is called from a pre-constructed pre-processing algorithm library, and a pre-processing operation tag is used to pre-process the production data to obtain identified production data;

[0104] Based on the modal type, the identified production data is summarized respectively to obtain a plurality of same type production data groups corresponding to the production data set, wherein the same type production data group includes a plurality of same type production data, and the modal type of the plurality of same type production data is the same;

[0105] Performing vectorization and parallel pre-processing on the plurality of same type production data groups to obtain a plurality of pre-processed same type production data groups, and summarizing the plurality of pre-processed same type production data groups to obtain historical production data;

[0106] Receiving a market data set, and obtaining historical market data based on the market data set;

[0107] Summarizing the historical production data and the historical market data to obtain initial historical archive data, performing archiving identification operation based on the historical product on the initial historical archive data to obtain historical archive data, and archiving the historical archive data into the historical data archiving unit based on the product configuration framework.

[0108] It can be understood that the target product line is the product line of the product to be optimized, and the plurality of historical products are a plurality of products included in the target product line. The production data set refers to a collection of data generated in the production process of the historical product. For example, when painting a car door, the painting time, operator, color, and the like, as well as the door image before painting, can all be production data. Therefore, the specific content of the production data can be determined according to the actual production process.

[0109] Further, the numerical data type refers to the type of production data that is a numerical modal type. The image data type, audio and video data type, and text data type refer to modal types of production data that are images, audio and video, and text, respectively. For example, in the production process of a car door, the numerical data type can represent the weight of the car door and the like, which can be quantified. The image data type, audio and video data type, and text data type correspond to visual records generated in the production process, such as door photos, audio and video files, such as quality inspection audio and video recordings, and text descriptions, such as process parameter records, and different modal production data.

[0110] It should be noted that the preprocessing algorithm library is a database containing a plurality of algorithms for preprocessing data, and the plurality of algorithms can be implemented by combining existing technologies. The preprocessing method label is the label name when executing one of the plurality of algorithms, and is used to identify and name different processing methods. In the preprocessing algorithm library, each modal type corresponds to a preprocessing method label.

[0111] For example, the present application sets the preprocessing method label according to the modal type of the production data: when the production data is of the numerical data type, the preprocessing method label is set to

no processing

image processing method

audio and video processing method

text processing method

[0112] Further, the production data is preprocessed and identified by using the preprocessing method label, that is, the preprocessing method label is added to the production data for identification, which is marked in text form and used to specify the preprocessing method for the production data. The identified production data is summarized based on the modal type to obtain a plurality of same type production data groups corresponding to the production data set, that is, the production data of the same modal type is stored in the same data group, and the stored same data group is the same type production data group, so the same type production data group includes a plurality of same type production data, and the modal types of the plurality of same type production data are the same, and the modal types of different same type production data groups are different. The number of same type production data groups is equal to the number of modal types.

[0113] Further, the market data set refers to the historical product operation data set collected through the market channel after listing, including but not limited to sales volume, click rate, return rate, buyer comments and other data reflecting market performance. The process of obtaining historical market data based on the market data set and obtaining historical production data based on the production data is similar and can achieve the same effect, which will not be repeated here.

[0114] It should be noted that in the data preprocessing process of the present application, the modal type of the production data is first identified, and the corresponding preprocessing method label is determined according to the modal type, and the data is classified and identified by using the label, and the production data of the same modal type is classified into a group, so that batch unified processing of the same type data is realized, and resource waste of processing one by one is avoided, and the data management efficiency is improved.

[0115] Further, the initial historical archive data is identified based on the historical product to obtain the historical archive data, that is, the data is extracted from the initial historical archive data to obtain the target archive data, at this time, the data extracted from the initial historical archive data is any data existing in the initial historical archive data, in the product configuration framework, the input value site of the target archive data is found, that is, the label path of the input value site, and the label path of the input value site is used as the identification text to identify the target archive data (the identification can be performed by the text identification method), and when all the data in the initial historical archive data are extracted and identified, the historical archive data is obtained. The historical archive data is archived in the historical data archiving unit based on the product configuration framework, that is, each identified data in the historical archive data is stored in the corresponding input value site according to the label path.

[0116] It can be understood that the vectorization parallel preprocessing is performed on the plurality of same type production data groups to obtain a plurality of preprocessed same type production data groups, including:

[0117] The following operations are performed on each of the plurality of homogeneous production data sets:

[0118] A parallel processing data vector is constructed based on the homogeneous production data set, and a pre-processing method instruction sequence corresponding to the pre-processing method tag is received, wherein the parallel processing data vector includes a plurality of parallel processing data, and the parallel processing data corresponds one-to-one to the homogeneous production data in the homogeneous production data set, and the pre-processing method instruction sequence includes a plurality of pre-processing instructions, and the plurality of pre-processing instructions are sorted according to the order of actual pre-processing operations;

[0119] The pre-processing instructions are sequentially extracted from the plurality of pre-processing instructions, and the extracted pre-processing instructions are used to perform vectorized parallel data processing on the parallel processing data vector to obtain an intermediate processing data vector, wherein the intermediate processing data vector is the parallel processing data vector, and the step of sequentially extracting pre-processing instructions from the plurality of pre-processing instructions is returned until the pre-processing instructions in the pre-processing method instruction sequence are extracted, and a pre-processed homogeneous production data set is obtained, wherein the pre-processed homogeneous production data in the pre-processed homogeneous production data set corresponds one-to-one to the homogeneous production data;

[0120] The pre-processed homogeneous production data sets are summarized to obtain a plurality of pre-processed homogeneous production data sets.

[0121] It should be noted that the parallel processing data vector is a vector composed of a plurality of parallel processing data, and the elements in the parallel processing data vector are the parallel processing data. The pre-processing method instruction sequence is a plurality of operation instructions of the algorithm represented by the pre-processing method tag.

[0122] For example, when the pre-processing method tag is

audio and video processing method

[0123] Further, the vectorized parallel data processing performed on the parallel processing data vector by using the extracted preprocessing instruction refers to that each parallel processing data in the parallel processing data vector is executed by using the extracted preprocessing instruction, so as to realize the parallel processing of each parallel processing data in the parallel processing data vector, thereby improving the processing speed and management efficiency. Further, the intermediate processing data vector is the result data obtained by executing the extracted preprocessing instruction on the parallel processing data vector. The intermediate processing data vector maintains the same structure as the parallel processing data vector and the elements are one-to-one corresponding. The only difference is that each element has completed the processing of the corresponding preprocessing instruction. When the preprocessing instruction extraction in the preprocessing method instruction sequence is completed, that is, each preprocessing instruction in the preprocessing method instruction sequence is executed on each vector in the parallel processing data vector, the data obtained is the preprocessing homogeneous production data.

[0124] S3, using a pre-constructed evaluation method, evaluating a plurality of historical archive data to obtain a plurality of product evaluation value groups, and obtaining product line optimization features based on the plurality of product evaluation value groups.

[0125] Further, the use of a pre-constructed evaluation method to evaluate a plurality of historical archive data to obtain a plurality of product evaluation value groups, and obtaining product line optimization features based on the plurality of product evaluation value groups, further comprises:

[0126] Sequentially extracting historical archive data from a plurality of historical archive data, and performing the following operations on the historical archive data:

[0127] Extracting an evaluation data set from the historical archive data, and sequentially extracting evaluation data from the evaluation data set. If the evaluation data is structured data, the evaluation data is confirmed as a numerical evaluation value.

[0128] If the evaluation data is not structured data, performing a non-structured data numerical operation based on multi-feature analysis on the evaluation data to obtain a numerical evaluation value group;

[0129] Summarizing the numerical evaluation value and the numerical evaluation value group to obtain an initial evaluation value set of the historical archive data;

[0130] Based on a plurality of historical archive data, obtaining a full data category, identifying a missing category set in the initial evaluation value set based on the full data category, and performing a zero filling operation on the initial evaluation value set based on the missing category set to obtain a numerical evaluation set;

[0131] aggregate the numerical evaluation set to obtain a plurality of original evaluation numerical group sets of historical archive data, perform normalization operation on the original evaluation numerical group sets based on the same characteristics to obtain a normalized evaluation group set, perform objective value evaluation on the normalized evaluation group set by using a pre-constructed entropy weight method to obtain a plurality of product evaluation values;

[0132] perform three clustering cluster-based clustering analysis on the plurality of product evaluation values by using a pre-constructed clustering method to obtain a plurality of evaluation clustering clusters, wherein the plurality of evaluation clustering clusters include a high evaluation value cluster, a medium evaluation value cluster and a low evaluation value cluster;

[0133] extract the high evaluation value cluster from the plurality of evaluation clustering clusters, and perform difference feature extraction on the high evaluation value cluster based on the medium evaluation value cluster and the low evaluation value cluster by using an intelligent decision unit to obtain product line optimization features.

[0134] It should be noted that the to-be-evaluated data set is artificially specified, and is a set composed of data to be evaluated. The to-be-evaluated data set includes a plurality of to-be-evaluated data, and is part of data that is artificially specified and has evaluation value and is filtered out from historical archive data. For example, if the product is a car, although the historical archive data contains production data such as door paint spraying time, the production data such as door paint spraying time does not belong to the to-be-evaluated data set because such data cannot reflect the product characteristics. Optionally, the to-be-evaluated data set includes but is not limited to production cycle, production energy consumption, etc.

[0135] It should be noted that the full data category is the union of the categories of the to-be-evaluated data set corresponding to each historical archive data in the plurality of historical archive data. The purpose of obtaining the full data category is to ensure that all historical archive data can be compared under a unified evaluation system. At the same time, because different historical products have different structural features and functional features, etc., the present application performs zero filling on the missing categories, so that the number of numerical evaluation values in the numerical evaluation set corresponding to each historical archive data is equal and one-to-one. The missing category set includes zero, one or more missing categories, and the missing category is a data category that exists in the full data category but does not exist in the historical archive data. The zero filling operation on the initial evaluation numerical set based on the missing category set means setting the numerical value corresponding to the position of the missing category set in the initial evaluation numerical set to 0.

[0136] For example, the to-be-evaluated data corresponding to the historical archive data A includes production cycle and production energy consumption, the to-be-evaluated data corresponding to the historical archive data B includes production cycle, production energy consumption and intelligent driving system, and the to-be-evaluated data corresponding to the historical archive data C includes repurchase rate, production energy consumption and intelligent driving system. The full data category is repurchase rate, production cycle, production energy consumption and intelligent driving system. The repurchase rate and intelligent driving system do not exist in the historical archive data A, so the missing category set corresponding to the historical archive data A is the intelligent driving system, and the initial evaluation value set can be 【0, production cycle, production energy consumption, 0】. Similarly, the present application will not be exemplified and described here.

[0137] Specifically, the original evaluation value group set includes a plurality of original evaluation value groups, and each original evaluation value group corresponds to historical archive data. The original evaluation value group includes a plurality of original evaluation values, and the plurality of original evaluation values are a plurality of numerical evaluation values in the numerical evaluation set. The normalized evaluation value corresponds to the original evaluation value one by one.

[0138] Further, in the process of performing the normalization operation based on the same feature on the original evaluation value group set to obtain the normalized evaluation value, the original evaluation value (for example, the production cycle of historical product A) is extracted from the plurality of original evaluation values, and all the same features of the data corresponding to the original evaluation value in the original evaluation value group set (for example, the production cycle set of each historical product in the product line) are found. After normalizing the original evaluation value and all the same features in the original evaluation value group set (for example, combining the production cycle of historical product A with the plurality of production cycles, and normalizing the combined production cycle), the normalized value of the original evaluation value is recorded as the normalized evaluation value.

[0139] Further, the method for using the pre-constructed entropy weight method to evaluate the objective value of the normalized evaluation group set to obtain a plurality of product evaluation values is a prior art, that is, the product evaluation value of each normalized evaluation value group in the normalized evaluation group set is calculated by using the entropy weight method, and the present application will not be described here.

[0140] It can be understood that the clustering method is a k-means clustering method, therefore, the three-cluster-based clustering analysis is: setting the final clustering cluster to 3, and using the k-means clustering method to divide the plurality of product evaluation values into three clustering clusters, and the three clustering clusters divided are the high evaluation value cluster, the medium evaluation value cluster and the low evaluation value cluster, which is the prior art and will not be repeated here. The mean of the product evaluation value in the high evaluation value cluster is greater than the mean in the medium evaluation value cluster, and the mean of the medium evaluation value cluster is greater than the mean in the low evaluation value cluster, so in the present application, it is considered that the product evaluation value in the high evaluation value cluster has a higher score, and when the product to be optimized needs to be optimized, the structural features or functional features of the product corresponding to the high evaluation value cluster can be selected as a reference for the optimization direction.

[0141] Further, in the process of using the intelligent decision unit to extract the distinguishing features of the high evaluation value cluster based on the medium evaluation value cluster and the low evaluation value cluster, it means that all products corresponding to the high evaluation value cluster are extracted first to obtain a high-value product set, and all products in the medium evaluation value cluster and the low evaluation value cluster are extracted to obtain a reference product set, and the intelligent decision unit is used to find a plurality of differences between the high-value product set and the reference product set, and the plurality of differences are product line optimization features.

[0142] It should be noted that the non-structured data numerical operation based on multi-feature analysis of the to-be-evaluated data to obtain a numerical evaluation value group comprises:

[0143] Identifying a same type data set of the to-be-evaluated data, wherein the same type data set does not include the to-be-evaluated data, and the same type data in the same type data set and the to-be-evaluated data come from different historical products;

[0144] If the same type data set is an empty set, the to-be-evaluated data is set to 1 to obtain a numerical evaluation value group;

[0145] If the same type data set is not an empty set, a plurality of unit features of the to-be-evaluated data that can be numerically evaluated are analyzed to obtain a unit feature group, and a numerical evaluation value group of the unit feature group is obtained.

[0146] It can be understood that in the present application, the numerically evaluated data is retained and the data corresponding to the missing category is zero-filled when the original evaluation value group set is obtained, resulting in completely different units and dimensions of different original evaluation values, which cannot be directly evaluated by the original evaluation value group for historical products, therefore, the present application performs a same-feature-based normalization operation on the original evaluation value group set to obtain a normalized evaluation value, so as to eliminate the difference in dimensions and provide a basis for the subsequent application of the entropy weight method.

[0147] Further, the same kind data set is the same kind of data as the extracted to-be-evaluated data index type, for example, the to-be-evaluated index data is an intelligent driving system, and each same kind data in the same kind data set is an intelligent driving system, and the same kind data and the to-be-evaluated data are intelligent driving systems of different historical products. Further, when the other historical products do not contain the intelligent driving system except the historical product corresponding to the to-be-evaluated data, the to-be-evaluated index is set to 1, indicating that the historical product corresponding to the historical archive data contains the intelligent driving system. It can be understood that at this time, since the other historical archive data does not contain the intelligent driving system, the to-be-evaluated index of containing the intelligent driving system is missing for the other historical archive data.

[0148] Further, if the same kind data set is not empty, it means that there are other similar data in the plurality of historical archive data, for example, the to-be-evaluated data is whether to contain intelligent driving, if the same kind data set is not empty, it means that at least one historical product of the other historical product corresponding to the other historical archive data contains the intelligent driving system, therefore, a plurality of unit features of the to-be-evaluated index are obtained, wherein the unit feature is an index artificially selected and can be numerically evaluated. For example, when the to-be-evaluated data contains the intelligent driving system, the network response speed, data transmission speed, etc. of the intelligent driving system can be artificially selected as a plurality of unit features, thereby realizing the conversion of unstructured production data into quantifiable numerical evaluation value groups. The numerical evaluation value group includes a plurality of numerical evaluation values, and the numerical evaluation value corresponds to the unit feature one by one.

[0149] S4, obtaining a plurality of effective competitors of the to-be-optimized product by using a market monitoring unit, performing market optimization feature extraction on the plurality of effective competitors to obtain market-oriented features.

[0150] Further, the plurality of effective competitors of the to-be-optimized product obtained by using the market monitoring unit comprises:

[0151] Obtaining a keyword set of the to-be-optimized product, performing same kind product retrieval based on the keyword set to obtain a plurality of initial same kind products, wherein the initial same kind products are identified by using uniform resource locators;

[0152] Performing ambiguity entity cleaning on the plurality of initial same kind products to obtain a plurality of cleaned same kind products;

[0153] Performing product screening operation based on compliance on the plurality of cleaned same kind products to obtain a plurality of initial competitors;

[0154] According to the full data category, a plurality of initial competitors are obtained, and competitor evaluation indexes of the plurality of initial competitors are obtained; according to the competitor evaluation indexes, the evaluation method and the clustering method, a plurality of high evaluation competitor values are obtained, and a plurality of initial competitors corresponding to the plurality of high evaluation competitor values are confirmed as a plurality of effective competitors.

[0155] It should be noted that the keyword set refers to a set of searchable keywords actively set by the product to be optimized on each transaction platform. The same product retrieval based on the keyword set to obtain a plurality of initial same products refers to obtaining the retrieval results (such as 500 products) as initial same products by inputting keywords on each platform. Further, the ambiguity entity cleaning of the plurality of initial same products to obtain a plurality of cleaned same products refers to screening the plurality of initial same products, so that the cleaned same products are consistent with the product category of the product to be optimized. For example, when the product to be optimized is an electric mosquito repellent, since the keyword “mosquito repellent” may misidentify non-same products such as mosquito nets, the cleaning process will eliminate such ambiguous non-same products.

[0156] It should be noted that the ambiguity entity cleaning of the plurality of initial same products to obtain a plurality of cleaned same products can be achieved by a convolutional neural network unit of an intelligent decision unit. This technology is prior art and will not be described here. The purpose of identifying the initial same products using a uniform resource locator is to distinguish different initial same products.

[0157] Further, the product screening operation for compliance refers to obtaining a plurality of screening indexes, sequentially extracting screening indexes from the plurality of screening indexes, and confirming a screening numerical value interval according to the extracted screening indexes. The numerical value of the cleaned same product under the screening index is obtained, and a to-be-confirmed numerical value is obtained. If the to-be-confirmed numerical value is not in the screening numerical value interval, the cleaned same product is deleted, and the remaining cleaned same products are summarized to obtain a plurality of initial competitors. For example, the screening index is sales volume, and the screening numerical value interval is sales volume greater than or equal to one hundred. Therefore, in the plurality of initial competitors, there is no initial competitor with sales volume less than one hundred.

[0158] Further, the competitor evaluation index is an evaluation index for evaluating the plurality of initial competitors, and the competitor evaluation index is contained in the full data category. The method of obtaining a plurality of high evaluation competitor values according to the competitor evaluation index, the evaluation method and the clustering method, and the method of obtaining a high evaluation value cluster according to the to-be-evaluated data set are similar, except that there is no product line in the plurality of initial competitors. Therefore, the plurality of initial competitors are taken as product lines to perform the operation of obtaining a plurality of high evaluation competitor values according to the competitor evaluation index, the evaluation method and the clustering method.

[0159] Further, the market optimization feature extraction of the plurality of effective competitors to obtain market-oriented features includes:

[0160] extracting the effective competitors in turn from the plurality of effective competitors, and performing the following operation on the extracted effective competitors:

[0161] extracting a plurality of component structure feature groups from the product configuration framework, wherein the component structure feature groups correspond to historical products one by one;

[0162] identifying, by using an intelligent decision unit, a competitor structure feature group of the effective competitor, and obtaining structure differences between the competitor structure feature group and each of the plurality of component structure feature groups, to obtain a plurality of structure difference data;

[0163] identifying, by using the intelligent decision unit, a plurality of function difference data of the effective competitor;

[0164] performing a matching operation based on historical products on the plurality of structure difference data and the plurality of function difference data, to obtain a plurality of initial market optimization feature groups, wherein the initial market optimization feature groups include the structure difference data, the function difference data and corresponding historical products;

[0165] performing a zero feature data deletion operation on the plurality of initial market optimization feature groups, to obtain unit competitor optimization guide data;

[0166] summarizing the unit competitor optimization guide data, to obtain market guide features of the plurality of effective competitors.

[0167] It should be noted that the component structure feature group is a combination of a plurality of components recorded in the product configuration framework of the historical product. It should be noted that in actual application, the component structure feature can also be a part in the mechanical structure. When the structure represented by the component structure feature group is a part, the structure in the competitor structure feature group is also a part, that is, the component structure feature and the competitor structure feature should have the same degree of structure refinement. Whether to refine to a part or a component can be selected by a professional. In the present application, the degree of refinement is represented by a component. The competitor structure feature group is a combination of a plurality of components similar to the component structure feature group in the effective competitor. The obtaining of the structure difference between the competitor structure feature group and each of the plurality of component structure feature groups to obtain a plurality of structure difference data means that the component structure feature group is extracted from the plurality of component structure feature groups in turn, and the intelligent decision system is used to find the difference between the component structure feature group and the competitor structure feature group, and the difference between the component structure feature group and the competitor structure feature group is taken as the structure difference data.

[0168] For example, the component structure feature set of the ergonomic chair includes a seat, a backrest, a waist support, and a headrest. The competitive product structure feature set of the effective competitor includes a seat, a backrest, a waist support, a headrest, and a footrest. Therefore, the structural difference data can be

the same structure of the seat, backrest, waist support, and headrest, and the effective competitor additionally has a footrest

[0169] Further, the multiple function difference data of the effective competitor identified by the intelligent decision unit are the same as the multiple structure difference data of the effective competitor identified by the intelligent decision unit, and the same effect can be achieved. Here, the example function difference data is

the same ergonomic chair can adjust the inclination angle, sitting depth, and armrest height, and the effective competitor additionally has a footrest that can be extended and retracted

[0170] As can be understood, the matching operation based on historical products on the multiple structure difference data and the multiple function difference data to obtain the multiple initial market optimization feature sets means that the structure difference data and the function difference data of the same historical product for the effective competitor are combined together, and the obtained data is an initial market optimization feature set.

[0171] Further, the zero feature data deletion operation on the multiple initial market optimization feature sets to obtain the multiple effective market optimization feature sets means that the initial market feature optimization set is sequentially extracted from the multiple initial market feature optimization sets. If the structure difference data in the initial market feature optimization set indicates that there is no difference (for example, the competitive product structure features of the effective competitor and the component structure feature set of a certain historical product are the same, and the structure difference data indicates that there is no difference), the structure difference data without difference is deleted. If the function difference data in the initial market feature optimization set indicates that there is no difference, the function difference data without difference is deleted. The purpose is to retain the different features between the historical product and the effective competitor as unit competitor optimization guide data to provide a reference for subsequent designers to optimize the product to be optimized.

[0172] S5, using an intelligent decision unit to analyze market guide features and product line optimization features to obtain multiple optimization demand data, and performing associated node data retrieval in a pre-constructed product configuration framework according to the multiple optimization demand data to obtain multiple associated nodes.

[0173] It should be noted that the use of an intelligent decision unit to analyze market guide features and product line optimization features to obtain multiple optimization demand data also includes the following:

[0174] The historical product is sequentially extracted from the target product line to obtain a target product, and the following operations are performed on the target product:

[0175] Identify multiple product components of the target product, and extract the product components sequentially from the multiple product components to obtain the unit component configuration of the extracted product components, thereby obtaining multiple unit component configurations corresponding to multiple product components. Here, there is a one-to-one correspondence between product components and unit component configurations. The unit component configuration includes multiple batches of product components with different component batch codes, and each batch of product components includes one or more unit part configurations. The unit part configuration includes multiple batches of product components with different part batch codes.

[0176] A part-level structure tree is constructed based on the configuration of a single part. The part-level structure trees corresponding to multiple batches of product components are summarized based on the configuration of a single component to obtain a set of part-level structure trees. The set of part-level structure trees includes one or more part-level structure trees. A component-level structure tree is constructed based on the target product, the set of part-level structure trees, and the multiple product components.

[0177] By summarizing the component-level structure tree, the initial configuration framework corresponding to the target product line is obtained;

[0178] Perform node framework expansion operations on the initial configuration framework to obtain a suboptimal configuration framework, and perform point labeling operations on the suboptimal configuration framework to obtain a product configuration framework.

[0179] Understandably, the product component is a component of the target product, for example, an accessory. Figure 5 The interior, doors, and other components have a certain degree of integration. Each product part corresponds one-to-one with a unit part configuration. Each unit part configuration includes multiple batches of product parts with different batch codes. The batch code represents the code for a different batch of product parts, and each batch of product parts represents a product part with a different batch code. For example, [attached...] Figure 5 The [Interior 1, Part Batch Code N] and [Interior 1, Part Batch Code M] are two product parts from two different batches.

[0180] Furthermore, when a batch of product components includes multiple parts, the product component corresponds to multiple unit part configurations. Each unit part configuration consists of multiple batch product parts with different batch codes. (See attached image.) Figure 4 As shown,

Interior 1, Part Batch Code N

Seat Cushion 1, Part Batch Code 1234

Seat Cushion 1, Part Batch Code 1235

Seat Cushion 1, Part Batch Code 1236

glass 1, part batch code 34

glass 1, part batch code 35

door frame 1, part batch code 45

door frame 1, part batch code 46

[0181] Further, the unit part configuration corresponds to the part-level structure tree one by one, and the part-level structure tree is a structure tree constructed by taking multiple batch product parts with different part batch codes as child nodes and taking batch product components as parent nodes. As shown in Figure 4 As shown, the part-level structure tree is a structure tree constructed by taking the interior as the parent node and multiple seat cushions 1 with different part batch codes as the child node.

[0182] It should be noted that when the component batch codes of the seat cushion 1 are different, it means that the production time and process parameters of different seat cushions 1 may be different, but the product corresponding to each component batch code can still be applied to the original historical product (such as historical product A) as the seat cushion 1.

[0183] When the structure of the batch product component changes, the changed batch product component should belong to the batch product component in another historical product, and the batch product part is the same. For example, the seat cushion 1 belongs to a batch product part in historical product A, and a handrail is added in the seat cushion 1. Therefore, the seat cushion 1 with the added handrail should belong to a batch product part in historical product B. Therefore, the evolution between historical products in the target product line is the change of structure and function, and in actual production, multiple structures or functions are often improved to form a new generation of historical products.

[0184] Further, the component-level structure tree is as shown in Figure 5 As shown, the component-level structure tree is a structure tree constructed by taking the product component as the parent node, connecting the part-level structure tree corresponding to the product component, and then taking the connected different product components as the child node and taking the corresponding historical product as the parent node.

[0185] Further, the node framework expansion operation is performed on the initial configuration framework to obtain a suboptimal configuration framework, and a point label identification operation is performed on the suboptimal configuration framework to obtain a product configuration framework, comprising:

[0186] Confirming multiple configuration nodes in the initial configuration framework, and performing the following operation on each configuration node in the multiple configuration nodes:

[0187] Based on the configuration node, a data category list is constructed, wherein the data category list includes a plurality of data categories, and the data category includes demand data, manufacturing data, experimental data and maintenance data.

[0188] From the plurality of data categories, a data category is sequentially extracted, and based on the extracted data category, a plurality of unit data trees are constructed, wherein the unit data tree sequentially includes a structured data node, an extension path link and a non-structured extension node, and the extension path link connects the structured data node and the non-structured extension node.

[0189] A node linking operation is performed between the plurality of unit data trees and the configuration node, and a data category tree corresponding to the configuration node is obtained.

[0190] The data category tree is summarized to obtain a suboptimal configuration framework, and a partition encryption operation based on access rights is performed on the suboptimal configuration framework to obtain an encrypted configuration framework.

[0191] A plurality of input value sites in the encrypted configuration framework are identified, and a label path of each of the plurality of input value sites in the encrypted configuration framework is identified to obtain a plurality of label paths, and the plurality of label paths are paired with input value sites to obtain a product configuration framework.

[0192] It should be noted that each node in the initial configuration framework can be a configuration node, for example, Figure 4

Interior 1

Cushion 1, Part Batch Code 1234

Adjustable cushion height

[0193] Further, in a data category, a plurality of data are included, for example, in the manufacturing data, size data, weight data and the like are included, and each data corresponds to a first unit data tree.

[0194] It can be understood that in the data category list, a plurality of unstructured data are obviously included, the unstructured data have better data volume than the numerical data, and directly storing the unstructured data in the historical data archiving unit needs to consume a large amount of memory, therefore, the application sets an unstructured expansion node for each configuration node, for storing the unstructured data, therefore the unstructured expansion node can be stored in the cloud, an external memory or the like, and the expansion path link is a link connecting the configuration node and the unstructured expansion node. The structured data node is a node in the configuration node for storing numerical data. For example, there is a configuration node for the weight of a door in manufacturing data, and the structured data node stores akg, through the expansion path link, the experimental report of the door weight when weighing can also be queried in the unstructured expansion node.

[0195] Further, the execution node linking operation refers to establishing a link with each unit data tree in the plurality of unit data trees as a child node and the configuration node as a parent node, so as to form a data category tree. It needs to be explained that after each configuration node in the initial configuration framework corresponds to a data category tree, a suboptimal configuration framework can be obtained.

[0196] It needs to be explained that the application mentioned in the foregoing can set an unstructured expansion node in the cloud and link through the expansion path link, at the same time, in the life cycle management of a product, various different people will access the historical data, therefore, the application only displays the name of data in plaintext in each configuration node, and encrypts the specific data, and the specific data can be displayed only when the access personnel has corresponding access authority. For example, there is a configuration node for the weight of a door in manufacturing data, and the structured data node stores akg, at this time, the weight of the door is in plaintext, akg is in ciphertext, at the same time, the experimental report of the door weight when weighing is in plaintext in the unstructured expansion node, and the data in the experimental report is in ciphertext, so as to realize the partition encryption operation of the suboptimal configuration framework based on access authority, and obtain an encrypted configuration framework. There are various existing technologies for encryption, which will not be described here.

[0197] Further, the input value site is a node for inputting a specific numerical value, for example, there is a configuration node for inputting the weight of the door in the manufacturing data, and the structured data node stores akg, at this time, the site where akg is located is the input value site, therefore, there are multiple input value sites in the encrypted configuration framework. The label path is a label for identifying the path of the input value site, when the numerical value corresponding to the label path is obtained next time, and it is expected to input the numerical value corresponding to the label path to the input value site, the label path can be searched according to the label path, so as to facilitate archiving the data. Therefore, the input value site pairing of the multiple label paths to obtain the product configuration framework refers to the operation of corresponding and associating the input value site and the corresponding label path, and then the encrypted configuration framework is recorded as the product configuration framework.

[0198] It can be understood that the intelligent decision unit analyzing the market orientation feature and the product line optimization feature refers to comparing the market orientation feature and the product line optimization feature, so as to provide a suggestion of confirming parts or components that need to be optimized according to the product configuration framework for professionals, at this time, the suggestion of confirming parts or components that need to be optimized is the multiple optimization demand data, for example, the market orientation feature and the product line optimization feature confirm that a seat armrest needs to be added, therefore, in the product configuration framework, the optimization demand data of optimizing the seat and adding the armrest component can be extracted.

[0199] Further, when the optimization demand data is confirmed, it is also necessary to confirm which other components or parts actually need to be optimized, therefore, the present application performs associated node data retrieval in the pre-constructed product configuration framework to obtain multiple associated nodes. For example, when the optimization demand data of optimizing the seat and adding the armrest component is confirmed, the size change after adding the armrest needs to be considered, therefore, the demand data of the seat is changed, therefore, the configuration node corresponding to the seat is the associated node, further, a unit part configuration (for example, the unit part configuration corresponding to adding the armrest) also needs to be added, therefore, all the configuration nodes in the added unit part configuration are associated nodes, and for example, whether the armrest is added in the historical product vehicle where the seat is located, the engine does not need to be added, therefore, the engine is not an associated node. By analogy, details are not described here.

[0200] S6, based on the multiple associated nodes, an optimization simulation strategy of a to-be-optimized product is obtained, and the to-be-optimized product is optimized and simulated based on the simulation unit and the optimization simulation strategy to obtain simulation data.

[0201] It can be understood that the optimization simulation strategy refers to that professionals analyze a plurality of optimization demand data and associated nodes, and then determine the optimization direction through artificial decision-making and formulate the specific implementation scheme according to the optimization direction, which is used to guide the specific implementation of the optimization of the product to be optimized. Specifically, it can be exemplified as selecting the node to be optimized from a plurality of associated nodes, and formulating the specific content of the optimization, for example, there is an associated node of a cushion, and the optimization direction of the cushion in the optimization simulation strategy is to reduce the thickness of the cushion.

[0202] It should be noted that before directly applying the optimization simulation strategy to the product to be optimized, simulation needs to be performed to confirm the feasibility of the optimization simulation strategy. Therefore, the present application performs optimization simulation on the product to be optimized based on the simulation unit and the optimization simulation strategy to obtain simulation data. The optimization simulation includes but is not limited to model establishment, stress test, etc., which is set by professionals. The simulation data is the record obtained in the optimization simulation.

[0203] S7, the simulation optimization result of the simulation data is evaluated to obtain an evaluation result. If the evaluation result meets the preset production optimization demand, the optimization simulation strategy is sent to the Internet of Things monitoring unit, and the optimized product is obtained by using the optimization simulation strategy received by the Internet of Things monitoring unit, and the product life cycle management based on artificial intelligence assistance is completed.

[0204] Further, the evaluation result is the numerical representation of a plurality of indexes set by humans, including but not limited to product performance, product design cycle and a plurality of parameters. The production optimization demand is the conclusion obtained after the professionals evaluate the simulation data, including but not limited to whether the optimization purpose is reached, etc. It can be realized by a variety of existing technologies such as scaling method and multi-person scoring, which will not be described here. If the evaluation result meets the preset production optimization demand, it is considered that the optimization simulation strategy can be used, and the product can be produced according to the optimization simulation strategy to realize the optimization of the product to be optimized. The optimization simulation strategy is sent to the Internet of Things monitoring unit, and the optimized product is obtained by using the optimization simulation strategy received by the Internet of Things monitoring unit, which means that the Internet of Things monitoring unit is used to monitor the production process in real time according to the optimization simulation strategy, and the production parameters are adjusted to realize the production of the optimized product.

[0205] To solve the problems described in the background art, the present application realizes intelligent optimization management of the whole process of the product by embedding artificial intelligence technology in the product lifecycle management system. The product lifecycle management system can monitor and collect real-time data of the whole link from product design, production and manufacturing to market feedback, and use artificial intelligence algorithm to analyze and process data of multiple modalities. It can also generate a direction for product optimization according to the evaluation method based on historical archival data and multiple effective competitors. Finally, it can extract associated nodes in the product configuration framework according to the optimization direction, intelligently provide prompts for professional personnel to generate optimization simulation strategies, and finally simulate according to the optimization simulation strategy to obtain the optimized product. The present application integrates product line historical data and market-oriented characteristics, uses an intelligent decision unit to generate an optimization simulation strategy, and performs product optimization after simulation verification, realizing product lifecycle management.

[0206] As Figure 2 shown, it is a functional module diagram of the lifecycle management system based on artificial intelligence assistance provided by an embodiment of the present application.

[0207] The lifecycle management system based on artificial intelligence assistance 100 can be installed in an electronic device. According to the realized functions, the lifecycle management system based on artificial intelligence assistance 100 can include a historical data acquisition module 101, a feature acquisition module 102, an associated node module 103, and a post-processing module 104. The modules of the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, stored in the memory of the electronic device.

[0208] The historical data acquisition module 101 is used to construct a sensor network and extract a sensor cluster from the sensor network, wherein the sensor cluster includes a plurality of sensors, mine subsidence information is collected using the sensor cluster, and the mine subsidence information is preprocessed to obtain a digital signal set, wherein the digital signal set includes a discrete signal set and a continuous signal set;

[0209] The feature acquisition module 102 is used to perform timestamp calibration on the digital signal set to obtain a time-calibrated signal set;

[0210] The associated node module 103 is used to upload the time-calibrated signal set to a pre-constructed edge processing device to obtain a temporary data set, identify an on-site data set in the temporary data set, transmit the on-site data set to a pre-constructed on-site device to obtain the on-site data set, and upload the temporary data set to a pre-constructed centralized processing center to obtain a to-be-processed data set. The edge processing device is deleted based on the edge processing device to obtain a cleaned edge processing device;

[0211] The post-processing module 104 is configured to perform spatio-temporal fusion on the to-be-processed data set to obtain an initial fusion data set, perform data cleaning on the initial fusion data set by using the pre-constructed abnormal data identification model to obtain an optimized fusion data set, update the pre-constructed visualization model set according to the optimized fusion data set, and complete the product life cycle management based on artificial intelligence assistance based on the updated visualization model set, the edge cleaning device, and the field data set.

[0212] In detail, the modules in the life cycle management system 100 based on artificial intelligence assistance in the embodiments of the present application use the same technical means as the product life cycle management method based on artificial intelligence assistance in the above Figure 1 , and can produce the same technical effects, which will not be described here.

[0213] As shown in Figure 3 , it is a structural schematic diagram of an electronic device for implementing the product life cycle management method based on artificial intelligence assistance according to an embodiment of the present application.

[0214] The electronic device 1 can include a processor 10, a memory 11, and a bus 12, and can further include a computer program stored in the memory 11 and executable on the processor 10, such as a product life cycle management method based on artificial intelligence assistance program.

[0215] The memory 11 includes at least one type of readable storage medium, including a flash memory, a mobile hard disk, a multimedia card, a card-type memory (such as an SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 can also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 11 includes an internal storage unit of the electronic device 1 and also includes an external storage device. The memory 11 can be used not only to store application software and various data installed in the electronic device 1, such as the code of the product life cycle management method based on artificial intelligence assistance program, but also to temporarily store data that has been output or will be output.

[0216] The processor 10 can be composed of integrated circuits in some embodiments, for example, can be composed of a single packaged integrated circuit, or can be composed of multiple packaged integrated circuits with the same function or different functions, including one or more central processing units (CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, etc. The processor 10 is the control core of the electronic device, which connects various components of the entire electronic device through various interfaces and lines, executes programs or modules stored in the memory 11 (such as an artificial intelligence assisted product lifecycle management method program, etc.), and calls data stored in the memory 11 to perform various functions and process data of the electronic device 1.

[0217] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11, the at least one processor 10, etc.

[0218] Figure 3 Only the electronic device with components is shown, and those skilled in the art can understand that, Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and can include fewer or more components than shown, or combine certain components, or different component arrangements.

[0219] For example, although not shown, the electronic device 1 can also include a power supply (such as a battery) for powering various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management system, so as to realize functions such as charge management, discharge management, and power consumption management through the power management system. The power supply can also include one or more direct current or alternating current power supplies, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, etc. The electronic device 1 can also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which are not described here.

[0220] Further, the electronic device 1 can also include a network interface, which can optionally include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), and is usually used to establish a communication connection between the electronic device 1 and other electronic devices.

[0221] Optionally, the electronic device 1 can also include a user interface, which can be a display, an input unit such as a keyboard, and optionally a standard wired interface, a wireless interface. Optionally, in some embodiments, the display can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, etc. Among them, the display can also be appropriately called a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display the visualized user interface.

[0222] The product lifecycle management method program based on artificial intelligence assistance stored in the memory 11 in the electronic device 1 is a combination of multiple instructions, which can realize the following functions when running in the processor 10:

[0223] Receiving product optimization instructions, and starting a lifecycle management system by using the product optimization instructions, wherein the lifecycle management system includes an intelligent decision unit, a historical data archiving unit, an Internet of Things monitoring unit, a simulation unit, and a market monitoring unit;

[0224] Identifying a product to be optimized based on the product optimization instructions, and extracting multiple historical archiving data of the same product line as the product to be optimized from the historical data archiving unit based on the product to be optimized;

[0225] Using a pre-constructed evaluation method to evaluate the multiple historical archiving data to obtain multiple product evaluation value groups, and obtaining product line optimization features based on the multiple product evaluation value groups;

[0226] Using the market monitoring unit to obtain multiple effective competitors of the product to be optimized, performing market optimization feature extraction on the multiple effective competitors to obtain market-oriented features;

[0227] Using the intelligent decision unit to analyze the market-oriented features and the product line optimization features to obtain multiple optimization demand data, and performing associated node data retrieval in a pre-constructed product configuration framework according to the multiple optimization demand data to obtain multiple associated nodes;

[0228] Obtaining an optimization simulation strategy for the product to be optimized based on the multiple associated nodes, and performing optimization simulation on the product to be optimized based on the simulation unit and the optimization simulation strategy to obtain simulation data;

[0229] The simulation optimization result evaluation is performed on the simulation data to obtain an evaluation result, and if the evaluation result meets the preset production optimization demand, the optimization simulation strategy is sent to the IoT monitoring unit, and the optimized product is obtained by using the optimization simulation strategy received by the IoT monitoring unit, and the product life cycle management based on the artificial intelligence assistance is completed.

[0230] Specifically, the specific implementation method of the processor 10 to the above instructions can refer to Figures 1 to 5 The description of related steps in the corresponding embodiments will not be repeated here.

[0231] Further, the modules / units integrated in the electronic device 1, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. The computer readable storage medium can be volatile or non-volatile. For example, the computer readable medium can include any entity or system capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory).

[0232] The application also provides a computer readable storage medium, the readable storage medium stores a computer program, the computer program can realize the following when being executed by the processor of the electronic device:

[0233] Receiving product optimization instructions, and starting a life cycle management system by using the product optimization instructions, wherein the life cycle management system comprises an intelligent decision unit, a historical data archiving unit, an IoT monitoring unit, a simulation unit, and a market monitoring unit;

[0234] Identifying a product to be optimized based on the product optimization instructions, and extracting a plurality of historical archiving data of a same product line as the product to be optimized from the historical data archiving unit based on the product to be optimized;

[0235] Evaluating the plurality of historical archiving data by using a pre-constructed evaluation method to obtain a plurality of product evaluation value groups, and obtaining product line optimization features based on the plurality of product evaluation value groups;

[0236] Obtaining a plurality of effective competitive products of the product to be optimized by using the market monitoring unit, performing market optimization feature extraction on the plurality of effective competitive products to obtain market-oriented features;

[0237] Analyzing the market-oriented features and the product line optimization features by using the intelligent decision unit to obtain a plurality of optimization demand data, performing associated node data retrieval in a pre-constructed product configuration framework according to the plurality of optimization demand data to obtain a plurality of associated nodes;

[0238] Obtain an optimization simulation strategy of the product to be optimized based on the plurality of associated nodes, and perform optimization simulation on the product to be optimized based on the simulation unit and the optimization simulation strategy to obtain simulation data.

[0239] Evaluate the simulation optimization result of the simulation data to obtain an evaluation result, and if the evaluation result meets the preset production optimization requirement, send the optimization simulation strategy to the Internet of Things monitoring unit, and obtain an optimized product by using the optimization simulation strategy received by the Internet of Things monitoring unit, thereby completing the product life cycle management assisted by artificial intelligence.

[0240] In several embodiments provided in the present application, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the above-described system embodiments are only illustrative, and actual implementation can have another division manner.

[0241] The modules described as separate components can or can not be physically separated, and the components shown as modules can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to actual needs.

[0242] In addition, the functional modules in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of hardware plus software functional modules.

[0243] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0244] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. An artificial intelligence assistance-based product lifecycle management method, characterized by, The method comprises: receiving product optimization instructions, starting a life cycle management system using the product optimization instructions, wherein the life cycle management system comprises: an intelligent decision unit, a historical data archiving unit, an Internet of Things monitoring unit, a simulation unit, and a market monitoring unit; based on the product optimization instructions, identifying a product to be optimized, and extracting a plurality of historical archived data of the same product line as the product to be optimized from the historical data archiving unit based on the product to be optimized; using a pre-constructed evaluation method, evaluating the plurality of historical archived data to obtain a plurality of product evaluation value groups, and obtaining product line optimization features based on the plurality of product evaluation value groups; using the market monitoring unit, obtaining a plurality of effective competitors of the product to be optimized, performing market optimization feature extraction on the plurality of effective competitors, and obtaining market-oriented features; using the intelligent decision unit, analyzing the market-oriented features and the product line optimization features to obtain a plurality of optimization demand data, performing associated node data retrieval in a pre-constructed product configuration framework according to the plurality of optimization demand data, and obtaining a plurality of associated nodes; based on the plurality of associated nodes, obtaining an optimization simulation strategy for the product to be optimized, and performing optimization simulation on the product to be optimized based on the simulation unit and the optimization simulation strategy to obtain simulation data; performing simulation optimization result evaluation on the simulation data to obtain an evaluation result, and if the evaluation result meets a pre-set production optimization demand, sending the optimization simulation strategy to the Internet of Things monitoring unit, and obtaining an optimized product using the optimization simulation strategy received by the Internet of Things monitoring unit to complete artificial intelligence assisted product life cycle management. 2.The artificial intelligence assistant-based product lifecycle management method of claim 1, wherein, The method further comprises: confirming a target product line of the product to be optimized, obtaining a plurality of historical products based on the target product line, and performing the following operations on each of the plurality of historical products: receiving a production data set of the historical product, and performing the following operations on each of the production data in the production data set: identifying a modal type of the production data, wherein the modal type comprises: structured data and unstructured data, the structured data is a numerical data type, and the unstructured data comprises image data type, audio and video data type, and text data type; according to the modal type, calling a pre-processing method label corresponding to the modal type from a pre-constructed pre-processing algorithm library, and performing a pre-processing operation identification on the production data using the pre-processing method label to obtain identified production data; based on the modal type, respectively summarizing the identified production data to obtain a plurality of homogeneous production data groups corresponding to the production data set, wherein each of the plurality of homogeneous production data groups comprises a plurality of homogeneous production data, and the modal type of the plurality of homogeneous production data is the same; performing vectorization parallel pre-processing on the plurality of homogeneous production data groups to obtain a plurality of pre-processed homogeneous production data groups, and summarizing the plurality of pre-processed homogeneous production data groups to obtain historical production data; receiving a market data set, and obtaining historical market data based on the market data set; The initial historical archiving data is obtained by aggregating historical production data and historical market data, the historical archiving data is obtained by performing archiving identification based on historical products, and the historical archiving data is archived into the historical data archiving unit based on a product configuration framework. 3.The artificial intelligence assistant-based product lifecycle management method of claim 2, wherein, The vectorized parallel preprocessing is performed on the multiple same-type production data sets to obtain multiple preprocessed same-type production data sets, including: The following operations are performed on each of the multiple same-type production data sets: A parallel processing data vector is constructed based on the same-type production data set, and a preprocessing method instruction sequence corresponding to a preprocessing method label is received, wherein the parallel processing data vector includes multiple parallel processing data, and the parallel processing data correspond to the same-type production data in the same-type production data set one by one, and the preprocessing method instruction sequence includes multiple preprocessing instructions, and the multiple preprocessing instructions are sorted according to the order of actual preprocessing operations; The preprocessing instructions are extracted from the multiple preprocessing instructions in sequence, and the vectorized parallel data processing is performed on the parallel processing data vector by using the extracted preprocessing instructions to obtain an intermediate processing data vector, and the intermediate processing data vector is used as the parallel processing data vector, and the step of extracting the preprocessing instructions from the multiple preprocessing instructions in sequence is returned until the preprocessing instructions in the preprocessing method instruction sequence are extracted, to obtain the preprocessed same-type production data set, wherein the preprocessed same-type production data in the preprocessed same-type production data set correspond to the same-type production data one by one; The preprocessed same-type production data sets are aggregated to obtain multiple preprocessed same-type production data sets.

4. The artificial intelligence assistance-based product lifecycle management method of claim 3, wherein, The multiple product evaluation values are obtained by using the pre-constructed evaluation method to evaluate the multiple historical archiving data, and the product line optimization features are obtained based on the multiple product evaluation value groups, and the previous method further includes: The historical archiving data is extracted from the multiple historical archiving data in sequence, and the following operations are performed on the historical archiving data: The evaluation data set is extracted from the historical archiving data, and the evaluation data is extracted from the evaluation data set in sequence, if the evaluation data is the structured data, the evaluation data is confirmed as the numerical evaluation value; If the evaluation data is not the structured data, the non-structured data numerical operation based on the multi-feature analysis is performed on the evaluation data to obtain the numerical evaluation value group; The numerical evaluation value and the numerical evaluation value group are aggregated to obtain the initial evaluation value set of the historical archiving data; The full data category is obtained based on the multiple historical archiving data, the missing category set in the initial evaluation value set is identified based on the full data category, and the initial evaluation value set is zero-filled based on the missing category set to obtain the numerical evaluation set; The numerical evaluation set is aggregated to obtain the original evaluation value group set of the multiple historical archiving data, the normalized evaluation group set is obtained by performing the normalization operation based on the same feature on the original evaluation value group set, the objective value evaluation is performed on the normalized evaluation group set by using the pre-constructed entropy weight method to obtain the multiple product evaluation values. Performing a three-cluster-based cluster analysis on the plurality of product evaluation values by using a pre-constructed clustering method, to obtain a plurality of evaluation clusters, wherein the plurality of evaluation clusters include a high evaluation value cluster, a medium evaluation value cluster, and a low evaluation value cluster; Extracting the high evaluation value cluster from the plurality of evaluation clusters, and performing a medium evaluation value cluster and a low evaluation value cluster-based feature extraction on the high evaluation value cluster by using an intelligent decision unit, to obtain product line optimization features. 5.The artificial intelligence assistant-based product lifecycle management method of claim 4, wherein, The non-structured data numerical operation based on multi-feature analysis of the to-be-evaluated data obtains a numerical evaluation value group, including: Identifying a same type data set of the to-be-evaluated data, wherein the same type data set does not include the to-be-evaluated data, and the same type data in the same type data set and the to-be-evaluated data come from different historical products; If the same type data set is empty, set the to-be-evaluated data to 1 to obtain a numerical evaluation value group; If the same type data set is not empty, analyze the plurality of unit features of the to-be-evaluated data that can be numerically evaluated to obtain a unit feature group, and obtain a numerical evaluation value group of the unit feature group. 6.The AI assistant-based product lifecycle management method of claim 5, wherein, The market-oriented features and product line optimization features are analyzed by using an intelligent decision unit to obtain a plurality of optimization demand data, which further includes: Extracting historical products from the target product line in sequence to obtain a target product, and performing the following operations on the target product: Identifying a plurality of product components of the target product, and sequentially extracting product components from the plurality of product components to obtain a plurality of unit component configurations corresponding to the plurality of product components, wherein the product components and the unit component configurations are one-to-one corresponding, the unit component configuration includes a plurality of batch product components with different component batch codes, and the batch product components include one or more unit part configurations, and the unit part configuration includes a plurality of batch product parts with different part batch codes; Constructing a part-level structure tree according to the unit part configuration, and aggregating the part-level structure trees corresponding to the plurality of batch product components according to the unit component configuration to obtain a part-level structure tree set, wherein the part-level structure tree set includes one or more part-level structure trees, and a component-level structure tree is constructed according to the target product, the part-level structure tree set, and the plurality of product components; Aggregating the component-level structure trees to obtain an initial configuration framework corresponding to the target product line; Performing a node framework expansion operation on the initial configuration framework to obtain a suboptimal configuration framework, and performing a point position label identification operation on the suboptimal configuration framework to obtain a product configuration framework. 7.The artificial intelligence assistant-based product lifecycle management method of claim 6, wherein, The node framework expansion operation on the initial configuration framework to obtain a suboptimal configuration framework, and the point position label identification operation on the suboptimal configuration framework to obtain a product configuration framework, including: Confirming a plurality of configuration nodes in the initial configuration framework, and performing the following operations on each configuration node in the plurality of configuration nodes: Based on the configuration node, constructing a data category list, wherein the data category list includes a plurality of data categories, and the data categories include demand data, manufacturing data, experimental data, and maintenance data; The data categories are sequentially extracted from multiple data categories, and multiple unit data trees are constructed based on the extracted data categories, wherein the unit data trees sequentially include a structured data node, an extended path link, and a non-structured expansion node, and the extended path link connects the structured data node and the non-structured expansion node; A node linking operation is performed between the multiple unit data trees and the configuration node to obtain a data category tree corresponding to the configuration node; The data category tree is summarized to obtain a suboptimal configuration framework, and a partition encryption operation based on access permissions is performed on the suboptimal configuration framework to obtain an encrypted configuration framework; Multiple input value sites in the encrypted configuration framework are identified, and a label path of each of the multiple input value sites in the encrypted configuration framework is identified to obtain multiple label paths, and the multiple label paths are paired with input value sites to obtain a product configuration framework. 8.The AI assistant-based product lifecycle management method of claim 7, wherein, The market monitoring unit is used to obtain multiple effective competitive products of the product to be optimized, including: A keyword set of the product to be optimized is obtained, and a same-type product search is performed based on the keyword set to obtain multiple initial same-type products, wherein the initial same-type products are identified by uniform resource locators; The multiple initial same-type products are subjected to ambiguity entity cleaning to obtain multiple cleaned same-type products; The multiple cleaned same-type products are subjected to product screening based on compliance to obtain multiple initial competitive products; Competitive product evaluation indexes of the multiple initial competitive products are obtained based on the full data category, and multiple high-evaluation competitive product values are obtained based on the competitive product evaluation indexes, the evaluation method, and the clustering method, and the multiple initial competitive products corresponding to the multiple high-evaluation competitive product values are confirmed as the multiple effective competitive products. 9.The artificial intelligence assistant-based product lifecycle management method of claim 8, wherein, The market optimization feature extraction is performed on the multiple effective competitive products to obtain market-oriented features, including: The effective competitive products are sequentially extracted from the multiple effective competitive products, and the following operations are performed on the extracted effective competitive products: Multiple component structure feature groups are extracted based on the product configuration framework, wherein the component structure feature groups correspond to historical products one by one; The competitive product structure feature groups of the effective competitive products are identified by using the intelligent decision unit, and structure differences between the competitive product structure feature groups and each of the multiple component structure feature groups are obtained to obtain multiple structure difference data; Multiple function difference data of the effective competitive products are identified by using the intelligent decision unit; The multiple structure difference data and the multiple function difference data are subjected to matching based on historical products to obtain multiple initial market optimization feature groups, wherein the initial market optimization feature groups include structure difference data, function difference data, and corresponding historical products; Zero feature data deletion is performed on the multiple initial market optimization feature groups to obtain unit competitive product optimization guide data; The market-oriented features of the multiple effective competitive products are obtained by summarizing the unit competitive product optimization guide data.

10. An artificial intelligence assistance based life cycle management system characterized in that, The system includes: A historical data acquisition module is configured to receive a product optimization instruction, and start a life cycle management system by using the product optimization instruction, wherein the life cycle management system includes an intelligent decision unit, a historical data archiving unit, an Internet of Things monitoring unit, a simulation unit, and a market monitoring unit; The product optimization instruction is used to identify a product to be optimized, and based on the product to be optimized, a plurality of historical archive data of the same product line as the product to be optimized are extracted from a historical data archiving unit; The feature acquisition module is configured to evaluate the plurality of historical archive data by using a pre-constructed evaluation method to obtain a plurality of product evaluation value groups, and to obtain product line optimization features based on the plurality of product evaluation value groups; The market monitoring unit is used to obtain a plurality of effective competitive products of the product to be optimized, and market optimization features are extracted from the plurality of effective competitive products to obtain market-oriented features; The correlation node module is configured to analyze the market-oriented features and the product line optimization features by using an intelligent decision unit to obtain a plurality of optimization demand data, to perform correlation node data retrieval in a pre-constructed product configuration framework according to the plurality of optimization demand data, and to obtain a plurality of correlation nodes; The post-processing module is configured to analyze the market-oriented features and the product line optimization features by using an intelligent decision unit to obtain a plurality of optimization demand data, to perform correlation node data retrieval in a pre-constructed product configuration framework according to the plurality of optimization demand data, and to obtain a plurality of correlation nodes; Based on the plurality of correlation nodes, an optimization simulation strategy of the product to be optimized is obtained, and the product to be optimized is optimized by simulation based on a simulation unit and the optimization simulation strategy to obtain simulation data; The simulation data is evaluated to obtain an evaluation result, and if the evaluation result meets a pre-set production optimization demand, the optimization simulation strategy is sent to the Internet of Things monitoring unit, and an optimized product is obtained by using the optimization simulation strategy received by the Internet of Things monitoring unit, thereby completing the product life cycle management assisted by artificial intelligence.

Images