An enterprise information management system and method based on artificial intelligence
By collecting and analyzing product failure data, using image recognition algorithms to identify faults and screening the closest maintenance site, the problem of inefficient fault identification and resolution in enterprise product after-sales service is solved, and fast and accurate fault diagnosis and maintenance is achieved. Reduced user costs.
Patent Information
- Application Number
- CN202410005127.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-03
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-01-03
AI Technical Summary
Existing after-sales service for enterprise products cannot accurately identify and quickly resolve user product failures, resulting in users having to go to designated after-sales sites for repair, which increases time and economic costs.
By collecting product failure characteristics image data, user failure location coordinates and personal identity data, using image recognition algorithms to match failure characteristics and analyze solutions, filter the closest maintenance site, and realize online fault diagnosis and maintenance operations.
It realizes rapid feedback and accurate diagnosis of user fault problems, reduces the time and economic cost of fault resolution, and improves after-sales service efficiency and user recognition.
Smart Images

Figure CN117788006B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of enterprise information management, and specifically to an enterprise information management system and method based on artificial intelligence. Background Art
[0002] Enterprise information management refers to the management of the enterprise information implementation process. Enterprise information management mainly includes enterprise change process management, enterprise operation management and the management of information technology, information resources, information equipment and other information implementation processes supported by information technology. The three aspects of enterprise information management are inseparable. They support and complement each other, achieve integration and mutual restriction. Enterprise information management belongs to the category of enterprise strategic management and is of great significance to enterprise development. The essence of enterprise information management is information integration. Its core elements are the construction of data platform and deep mining of data. Through the information management system, the design, procurement, production, manufacturing, finance, marketing, operation, management and other links of the enterprise are integrated to share information and resources. At the same time, modern technical means are used to find potential customers, effectively support the decision-making system of the enterprise, and achieve the purpose of reducing inventory, improving production efficiency and quality, and responding quickly, so as to enhance the market competitiveness of the enterprise; ERP, OA, CRM, BI, PLM, e-commerce, etc. have become indispensable application systems for enterprises in the process of management informationization, among which ERP is moving towards highly integrated full-process management informationization. The purpose of enterprise information management is to improve the quality of products and services, competitiveness and reduce costs and increase efficiency. After-sales service management of products is an important part of improving service and product quality. With the development of informatization, the structure and functions of products are becoming increasingly complex and diversified, which increases the pressure on after-sales service of enterprise products and the economic cost of users using products. The existing after-sales service of enterprise products can only be carried out by users going to designated after-sales sites for product repair and maintenance. Since the fault status, time and location of the product during use are unpredictable, the economic and time cost of users using the product is increased. In the after-sales management of enterprise products, it is impossible to accurately identify the faults that occur when users use the product and provide reliable solutions.
[0003] The Chinese invention patent application with the publication number CN114611861A discloses an enterprise resource information management system, which is provided with a resource management module, a merchant management module, an audit management module, a target management module, and a system management module, wherein the resource management module includes: resource creation, resource viewing, resource copying, and resource change; the merchant management module includes: merchant creation, personnel addition, and configuration workflow; the audit management module includes: resource audit, merchant audit, and resource change audit; the target library management module includes: non-target management and target management; the system management module includes: user management, role management, authority management, password modification, announcement management, and establishment of operation logs. The present invention classifies and creates enterprise resources by providing a resource management module, and at the same time facilitates enterprise personnel to view resources, facilitates employees to understand materials in a timely manner, and is also very convenient for resource changes. However, the resource management of the above enterprise information management scheme lacks after-sales management of enterprise products, and cannot provide users with accurate, convenient, and efficient product after-sales service. Summary of the invention
[0004] 1. Technical issues to be resolved
[0005] In order to solve the problem that the after-sales service of the above-mentioned existing enterprise products can only be carried out by users going to designated after-sales sites for product repair and maintenance, and the economic and time costs of users using the products are increased due to the unpredictable failure status, time and location of the product during use, and the enterprise product after-sales management cannot accurately identify and provide reliable solutions to the failures that occur when users use the products, the purpose of receiving product failures reported by users in real time, accurately identifying the failures, providing reliable product failure solutions and performing product maintenance operations is achieved.
[0006] (II) Technical solution
[0007] The present invention is implemented by the following technical solution: an enterprise information management method based on artificial intelligence, the method comprising the following steps:
[0008] S1. Collect product failure feature image data, coordinates of the location where the user's product failure occurred, and the user's personal identity data. The collection of the user's personal identity data is only carried out under the condition that the user himself allows it;
[0009] S2, matching the fault feature image with the product fault solution data through an image recognition algorithm, analyzing and generating the user product fault corresponding solution data and determining the generation result of the user product fault corresponding solution data;
[0010] S3. When the generated result of the solution data corresponding to the user product failure is empty data, re-execute step S2 and calculate the number of product failure solution analysis times and compare the number with the analysis times threshold;
[0011] S4. When the number of times the product failure solution is analyzed is greater than the set threshold of the number of analyses or the generated result of the solution data corresponding to the user product failure is not empty data, the final product failure solution analysis result data is output; when the final product failure solution analysis result data is output, the distances between the product failure location and different product failure maintenance sites are respectively measured based on the coordinates of the user product failure location and the product failure maintenance site coordinates of different product failure maintenance sites, and the distance data from the product failure maintenance site to the product failure location is generated;
[0012] S5. Perform numerical analysis using a data search algorithm based on the distance data from the product failure maintenance site to the product failure location, search for the distance data from the product failure maintenance site to the product failure location with the smallest distance value, and identify and generate the minimum distance data from the product failure maintenance site to the faulty product;
[0013] S6. Filter out the coordinates of the product failure maintenance site corresponding to the maintenance site according to the minimum distance data from the product failure maintenance site to the faulty product and mark and generate the final product failure maintenance site coordinates;
[0014] S7. Feedback the user personal identity data and the final product failure solution analysis result data to the maintenance site control terminal according to the final product failure maintenance site coordinates and perform product failure maintenance operations.
[0015] Preferably, the steps of collecting product failure feature image data, user product failure location coordinates, and user personal identity data are as follows:
[0016] S11, collect product failure feature image data through mobile terminal shooting, and establish product failure feature image data set A = (a1, ..., a m ,…,a α ), m=1,2,3,…,α; where a m represents the mth product fault feature image data collected at the fault location of the product, and α represents the maximum number of product fault feature image data;
[0017] The mobile terminal locates and collects the spatial position of the faulty product and generates the coordinates of the user product failure location b = (x b ,y b ,z b ), where x b Indicates the longitude of the space where the faulty product is located, yb Indicates the latitude of the space where the faulty product is located, z b Indicates the altitude of the space where the faulty product is located;
[0018] The user enters personal identity information through the mobile terminal to complete the user identity information collection and generate user personal identity data I, where the personal identity data includes one or more of the user's name, contact information, and ID number.
[0019] Preferably, the steps of matching the fault feature image with the product fault solution data through an image recognition algorithm, analyzing and generating the user product fault corresponding solution data, and judging the generation result of the user product fault corresponding solution data are as follows:
[0020] S21. Establish product failure solution data set C = (c1,…,c n ,…,c β ), n=1,2,3,…,β; where c n Indicates the product failure solution data corresponding to the failure feature image of the nth product failure collected, β indicates the maximum value of the product failure solution data type, and the product failure solution includes one or more of a transmission device damage solution, a pump wear solution, a brake device failure solution, a device power loss solution, a motor failure solution, a transformer failure solution, a cable failure solution, a software startup failure solution, and a software operation abnormality solution;
[0021] S22, using a data recognition algorithm to identify the product failure feature image data a in the product failure feature image data set A. m and product failure solution data c in product failure solution data set C n According to the product failure feature image matching, the specific steps of the data recognition algorithm for product failure feature image recognition are as follows:
[0022] S221, initialize and update the maximum number of iterations. In the product failure solution data set C constituting a β-dimensional optimization problem, the solution search sand cat is a 1×β array representing the problem solution, each variable value is a floating point number, and each variable value is between a lower bound and an upper bound. The fitness of each solution search sand cat is obtained by the fitness function of the problem to be solved;
[0023] S222, search for prey, the final parameter and main parameter controlling the transition between the exploration and development phases is R. When |R|>1, the solution search cat searches for the product failure solution data set C with the product failure feature image data a m Matching product failure solution datacn The solution search sand cat search process depends on the release of low-frequency noise. The sensitivity range r of the solution search sand cat is set from 0 to 2kHz. S is inspired by the auditory characteristics of the solution search sand cat and is set to 2. t is the current iteration number, T is the maximum iteration number, and rand (0,1) represents a random number from 0 to 1.
[0024] Each solution searches for the sand cat according to the best candidate position P best Update its own position and current position P and its sensitivity range r in the product failure solution data set C to find the image data a with product failure feature m Matching product failure solution datac n location;
[0025] P(t+1)=r×(P best (t)-rand(0,1)×P best (t)), where
[0026] P(t+1) represents the position of the solution search sand cat in the product failure solution data set C after t+1 iterations, P best (t) represents the optimal position of the sand cat in the product failure solution data set C after t iterations of solution search;
[0027] S223, attack prey, when |R|≤1, solution search sand cat for product failure solution data c n Find and use the best position P best Generate a random position with the current position P, set the sensitivity range of the solution search sand cat to be a circle, use the roulette method to randomly select a random angle θ for each solution search sand cat and perform product failure solution data c n Search, the search formula is in represents the sensitivity vector, cos(θ) represents the cosine value of the angle θ, and P rand represents the random position of the solution-searching sand cat in the product failure solution data set C. The random position can ensure that the solution-searching sand cat is close to its prey, and the random angle can prevent the algorithm from falling into a local optimum;
[0028] S224, when the search algorithm reaches the maximum number of iterations, output a m With c n If the matching result is not reached, then continue to iterate and execute steps S222 to S223 until the maximum number of iterations is met;
[0029] when a m With c nIf the match is successful, the output data of the solution corresponding to the user product failure is
[0030] when a m With c n If the match is not successful, the output data of the solution corresponding to the user product failure is None;
[0031] S23, judging the generation result of the solution data corresponding to the user product failure, when the solution data corresponding to the user product failure is Then the generated result of outputting the solution data corresponding to the user product failure is not empty data; when the solution data corresponding to the user product failure is none, the generated result of outputting the solution data corresponding to the user product failure is empty data.
[0032] Preferably, when the result of generating the solution data corresponding to the user product failure is empty data, the operation steps of re-executing step S2 and counting the number of product failure solution analysis times and comparing the number with the analysis number setting threshold are as follows:
[0033] S31. When the result of generating the solution data corresponding to the user product failure is empty data, re-execute steps S22 to S23 and simultaneously count the total number of times steps S22 to S23 are executed and generate the product failure solution analysis number σ;
[0034] S32, setting the analysis number and setting the threshold σ yuzhi , where σ yuzhi The value of is a positive integer; yuzhi The larger the value of , the more times steps S22 to S23 are repeated when the result of generating the solution data corresponding to the user product failure is empty data; σ yuzhi The smaller the value of indicates that when the result of generating the solution data corresponding to the user product failure is empty data, the number of times steps S22 to S23 are repeatedly executed is less;
[0035] S33, after each repetition of steps S22 to S23, the number of product failure solution analysis σ and the number of analysis are set to a threshold σ yuzhi For numerical comparison, when σ≤σ yuzhi When σ>σ, continue to repeat steps S22 to S23; yuzhi When the process stops, the steps S22 to S23 are repeated, and the solution data corresponding to the user product failure is output.
[0036] Preferably, when the number of times the product fault solution is analyzed is greater than the analysis number setting threshold or the generation result of the user product fault corresponding solution data is not empty data, the final product fault solution analysis result data is output; when the final product fault solution analysis result data is output, the distances between the product fault location and different product fault maintenance sites are respectively measured based on the coordinates of the user product fault location and the product fault maintenance site coordinates of different product fault maintenance sites, and the distance data from the product fault maintenance site to the product fault location are generated as follows:
[0037] S41, when σ > σ yuzhi When the generated result of the user product failure corresponding solution data is not empty data, the output user product failure corresponding solution data is identified as the final product failure solution analysis result data for output;
[0038] S42. Establish a product failure maintenance site coordinate set D = (d1, ..., d u ,…,d χ ), u=1,2,3,…,χ; where d u represents the corresponding product failure maintenance site coordinates of the u-th product failure maintenance site, χ represents the maximum number of product failure maintenance sites; d u =(x u ,y u ,z u ), where x u Indicates the longitude of the space where the product failure maintenance site is located, y u Indicates the latitude of the space where the product failure maintenance site is located, z u Indicates the altitude of the space where the product failure maintenance site is located;
[0039] S43, when outputting the final product failure solution analysis result data, according to the user product failure location coordinate b and the product failure maintenance site coordinate d in the product failure maintenance site coordinate set D, u The distance formula is used to calculate the distance between the product failure location and different product failure maintenance sites and generate a distance data set from the product failure maintenance site to the product failure location E = (e 1,b ,…,e u,b ,…,e χ,b ), u=1,2,3,…,χ; where e u,b Indicates the distance data from the u-th product failure maintenance site to the location where the product failure occurred.
[0040] Preferably, the operation steps of using a data search algorithm to perform numerical analysis based on the distance data from the product fault maintenance site to the product fault occurrence site, searching for the distance data from the product fault maintenance site to the product fault occurrence site with the smallest distance value and identifying the minimum distance data from the product fault maintenance site to the faulty product are as follows:
[0041] S51, obtaining a data set E of the distance from the product failure maintenance site to the location where the product failure occurred;
[0042] S52, using a depth-first search algorithm, searching for the product failure maintenance site to product failure location distance data set E with the smallest value according to the value size. u,b And identify and generate the minimum distance data from the product failure maintenance site to the faulty product min .
[0043] Preferably, the steps of selecting the product failure maintenance site coordinates of the corresponding maintenance site based on the minimum distance data from the product failure maintenance site to the faulty product and marking and generating the final product failure maintenance site coordinates are as follows:
[0044] S61, maintaining the minimum distance data from the product failure site to the failed product min Obtain the product fault maintenance site number u according to the product fault maintenance site number u min The corresponding product failure maintenance site coordinates d u And identify the coordinates of the final product failure maintenance site d queding .
[0045] Preferably, the steps of feeding back the user personal identity data and the final product failure solution analysis result data to the maintenance site control terminal based on the final product failure maintenance site coordinates and performing the product failure maintenance operation are as follows:
[0046] S71, maintaining the site coordinates d according to the final product failure queding The numbered location of the product fault maintenance site is obtained, and the user personal identity data I and the final product fault solution analysis result data are pushed back to the maintenance site control end through the mobile communication network. The maintenance site control end contacts the user based on the user personal identity data I and arranges staff to perform online or offline product fault maintenance operations using the received final product fault solution analysis result data.
[0047] A system for implementing the enterprise information management method based on artificial intelligence, the system comprising a product failure related information acquisition management module, a product failure solution management module, and a product failure processing execution management module;
[0048] The product failure related information acquisition management module includes a product failure acquisition unit, a user product failure location coordinate acquisition unit, and a user identity information acquisition unit;
[0049] The product failure collection unit collects product failure feature image data through the mobile terminal; the user product failure location coordinate collection unit collects the user product failure location coordinates through the mobile terminal; the user identity information collection unit collects user personal identity data through the mobile terminal;
[0050] The product failure solution management module includes a product failure solution storage unit, a product failure solution analysis and judgment unit, a product failure solution analysis times statistics and judgment unit, and a product failure solution result output unit;
[0051] The product failure solution storage unit is used to store product failure solution data; the product failure solution analysis and judgment unit is used to match the product failure feature image with the product failure solution data through an image recognition algorithm, analyze and generate the user product failure corresponding solution data and judge the generation result of the user product failure corresponding solution data; the product failure solution analysis times counting and judgment unit is used to re-execute the product failure solution analysis step and count the product failure solution analysis times and compare the analysis times set threshold value when the generation result of the user product failure corresponding solution data is empty data; the product failure solution result output unit outputs the final product failure solution analysis result data when the product failure solution analysis times is greater than the analysis times set threshold value or the generation result of the user product failure corresponding solution data is not empty data;
[0052] The product failure processing execution management module includes a product failure maintenance site coordinate storage unit, a product failure maintenance site and faulty product distance measurement unit, a product failure maintenance site screening unit, and a product failure after-sales maintenance processing execution unit;
[0053] The product fault maintenance site coordinate storage unit is used to store the product fault maintenance site coordinates; the product fault maintenance site and fault product distance measurement unit, when outputting the final product fault solution analysis result data, measures the distance between the product fault occurrence site and different product fault maintenance sites according to the coordinates of the user product fault occurrence site and the product fault maintenance site coordinates of different product fault maintenance sites, and generates the distance data from the product fault maintenance site to the product fault occurrence site; the product fault maintenance site screening unit is used to perform numerical analysis based on the product fault maintenance site to product fault occurrence site distance data using a data search algorithm, search for the product fault maintenance site to product fault occurrence site distance data with the smallest distance value and identify and generate the product fault maintenance site to fault product minimum distance data, and screen the product fault maintenance site coordinates of the corresponding maintenance site according to the product fault maintenance site to fault product minimum distance data and identify and generate the final product fault maintenance site coordinates; the product fault after-sales maintenance processing execution unit feeds back the user personal identity data and the final product fault solution analysis result data to the maintenance site control end according to the final product fault maintenance site coordinates and executes the product fault maintenance operation.
[0054] (III) Beneficial effects
[0055] The present invention provides an enterprise information management system and method based on artificial intelligence. It has the following beneficial effects:
[0056] 1. The product fault collection unit uses the mobile terminal to accurately collect the product's fault feature images online, so that users can quickly feedback on product failure issues; the user product failure location coordinate collection unit uses the mobile terminal to accurately locate the location of the product failure, providing data support for product after-sales information management; the user identity information collection unit uses the mobile terminal to collect the user's personal identity information, ensuring accurate and efficient communication after product sales and rapid execution of product maintenance operations.
[0057] 2. The product fault solution analysis and judgment unit uses an intelligent recognition algorithm to accurately identify the image features of the collected product fault point images and product fault solution data, intelligently analyze the solution to the product fault, realize online diagnosis of product faults and provide processing solutions, and improve the service efficiency and quality of the company's after-sales information management; the product fault solution analysis frequency statistics and judgment unit and the product fault solution result output unit cooperate with each other to self-check and confirm the product fault solution analysis process, thereby improving the reliability of the product fault solution analysis results, simplifying the company's product after-sales maintenance process, improving user recognition of the product, and improving the quality of the company's information management.
[0058] 3. Through the cooperation between the product fault maintenance site and the faulty product distance measurement unit and the product fault maintenance site screening unit, the fault maintenance site closest to the product fault site is accurately measured and screened out, and the location information of the corresponding maintenance site is obtained, so as to realize the intelligent screening of product fault maintenance sites, reduce the time and economic cost of users to solve product faults, and enhance the competitiveness of enterprise product after-sales service and the enterprise information management capabilities; the product fault after-sales maintenance processing execution unit initiates the nearest product fault maintenance site control terminal to contact the user, and arranges maintenance personnel to perform online or offline product fault maintenance operations on the product fault problem, so as to realize the informatization and standardization of product after-sales maintenance operations, and enhance the intelligence of enterprise product after-sales information management. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] Figure 1 A module schematic diagram of an enterprise information management system based on artificial intelligence provided by the present invention;
[0060] Figure 2 The present invention provides a flowchart of an enterprise information management method based on artificial intelligence. DETAILED DESCRIPTION
[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0062] The embodiments of the enterprise information management system and method based on artificial intelligence are as follows:
[0063] See also Figure 1 - Figure 2 , an enterprise information management method based on artificial intelligence, the method comprises the following steps:
[0064] S1. Collect product failure feature image data, coordinates of the user's product failure location, and user personal identity data;
[0065] S2, matching the fault feature images with the product fault solution data through an image recognition algorithm, analyzing and generating the user product fault corresponding solution data and determining the generation result of the user product fault corresponding solution data;
[0066] S3. When the result of generating the solution data corresponding to the user's product failure is empty data, re-execute step S2 and count the number of product failure solution analyses and compare the number with the analysis number threshold;
[0067] S4. When the number of product failure solution analysis times is greater than the set analysis times threshold or the generated result of the user product failure corresponding solution data is not empty data, the final product failure solution analysis result data is output; when the final product failure solution analysis result data is output, the distances between the product failure location and the different product failure maintenance sites are measured based on the coordinates of the user product failure location and the coordinates of the product failure maintenance sites of the different product failure maintenance sites, and the distance data from the product failure maintenance site to the product failure location is generated;
[0068] S5. Perform numerical analysis using a data search algorithm based on the distance data from the product failure maintenance site to the product failure location, search for the distance data from the product failure maintenance site to the product failure location with the smallest distance value, and identify and generate the minimum distance data from the product failure maintenance site to the faulty product;
[0069] S6. Filter out the coordinates of the corresponding maintenance site based on the minimum distance data between the product failure maintenance site and the faulty product, and mark and generate the final coordinates of the product failure maintenance site;
[0070] S7. Feedback the user's personal identity data and the final product failure solution analysis result data to the maintenance site control end based on the final product failure maintenance site coordinates and perform product failure maintenance operations.
[0071] For further information, see Figure 1 - Figure 2 The steps for collecting product failure feature image data, user product failure location coordinates, and user personal identity data are as follows:
[0072] S11, collect product failure feature image data through mobile terminal shooting, and establish product failure feature image data set A = (a1, ..., a m ,…,a α ), m=1,2,3,…,α; where a m represents the mth product fault feature image data collected at the fault location of the product, and α represents the maximum number of product fault feature image data;
[0073] The mobile terminal locates and collects the spatial position of the faulty product and generates the coordinates of the user product failure location b = (x b ,y b ,z b ), where x b Indicates the longitude of the space where the faulty product is located, y b Indicates the latitude of the space where the faulty product is located, z b Indicates the altitude of the space where the faulty product is located;
[0074] The user enters personal identity information through the mobile terminal to complete the user identity information collection and generate user personal identity data I, where the personal identity data includes one or more of the user's name, contact information, and ID number.
[0075] The product fault collection unit uses the mobile terminal to accurately collect the product's fault feature images online, enabling users to quickly feedback on product failure issues; the user product failure location coordinate collection unit uses the mobile terminal to accurately locate the location of the product failure, providing data support for product after-sales information management; the user identity information collection unit uses the mobile terminal to collect the user's personal identity information, ensuring accurate and efficient communication after product sales and rapid execution of product maintenance operations.
[0076] For further information, see Figure 1 - Figure 2 The steps of matching the fault feature image data with the product fault solution data through the image recognition algorithm, analyzing and generating the user product fault corresponding solution data and judging the generation result of the user product fault corresponding solution data are as follows:
[0077] S21. Establish product failure solution data set C = (c1,…,c n ,…,c β ), n=1,2,3,…,β; where c n Indicates the product failure solution data corresponding to the failure feature image of the nth product failure collected, β indicates the maximum value of the product failure solution data type, and the product failure solution includes one or more of the following: transmission device damage solution, pump wear solution, brake device failure solution, equipment power loss solution, motor failure solution, transformer failure solution, cable failure solution, software startup failure solution, and software abnormal operation solution;
[0078] S22, using a data recognition algorithm to identify the product failure feature image data set A m and product failure solution data c in product failure solution data set C n According to the product fault feature image matching, the specific steps of the data recognition algorithm for product fault feature image recognition are as follows:
[0079] S221, initialize and update the maximum number of iterations. In the product failure solution data set C constituting a β-dimensional optimization problem, the solution search sand cat is a 1×β array representing the problem solution, each variable value is a floating point number, and each variable value is between a lower bound and an upper bound. The fitness of each solution search sand cat is obtained by the fitness function of the problem to be solved;
[0080] S222, search for prey, the final parameter and main parameter controlling the transition between the exploration and development phases is R. When |R|>1, the solution search cat searches for the product failure solution data set C with the product failure feature image data a m Matching product failure solution datac n The solution search sand cat search process depends on the release of low-frequency noise. The sensitivity range r of the solution search sand cat is set from 0 to 2kHz. S is inspired by the auditory characteristics of the solution search sand cat and is set to 2. t is the current iteration number, T is the maximum iteration number, and rand (0,1) represents a random number from 0 to 1.
[0081] Each solution searches for the sand cat according to the best candidate position P best Update its own position and current position P and its sensitivity range r in the product failure solution data set C to find the image data a with product failure feature m Matching product failure solution datac n location;
[0082] P(t+1)=r×(P best (t)-rand(0,1)×P best (t)), where
[0083] P(t+1) represents the position of the solution search sand cat in the product failure solution data set C after t+1 iterations, P best (t) represents the optimal position of the sand cat in the product failure solution data set C after t iterations of solution search;
[0084] S223, attack prey, when |R|≤1, solution search sand cat for product failure solution data c n Find and use the best position P best Generate a random position with the current position P, set the sensitivity range of the solution search sand cat to be a circle, use the roulette method to randomly select a random angle θ for each solution search sand cat and perform product failure solution data c n Search, the search formula is in represents the sensitivity vector, cos(θ) represents the cosine value of the angle θ, and P rand represents the random position of the solution-searching sand cat in the product failure solution data set C. The random position can ensure that the solution-searching sand cat is close to its prey, and the random angle can prevent the algorithm from falling into a local optimum;
[0085] S224, when the search algorithm reaches the maximum number of iterations, output am With c n If the matching result is not reached, then continue to iterate and execute steps S222 to S223 until the maximum number of iterations is met;
[0086] when a m With c n If the match is successful, the output data of the solution corresponding to the user product failure is
[0087] when a m With c n If the match is not successful, the output data of the solution corresponding to the user product failure is None;
[0088] S23, judging the generation result of the solution data corresponding to the user product failure, when the solution data corresponding to the user product failure is Then the generated result of outputting the solution data corresponding to the user product failure is not empty data; when the solution data corresponding to the user product failure is none, the generated result of outputting the solution data corresponding to the user product failure is empty data.
[0089] When the result of generating the solution data corresponding to the user's product failure is empty data, the steps of re-executing step S2 and counting the number of product failure solution analyses and comparing the number with the analysis number setting threshold are as follows:
[0090] S31. When the result of generating the solution data corresponding to the user product failure is empty data, re-execute steps S22 to S23 and simultaneously count the total number of times steps S22 to S23 are executed and generate the product failure solution analysis number σ;
[0091] S32, setting the analysis number and setting the threshold σ yuzhi , where σ yuzhi The value of is a positive integer; yuzhi The larger the value of , the more times steps S22 to S23 are repeated when the result of generating the solution data corresponding to the user product failure is empty data; σ yuzhi The smaller the value of indicates that when the result of generating the solution data corresponding to the user product failure is empty data, the number of times steps S22 to S23 are repeatedly executed is less;
[0092] S33, after each repetition of steps S22 to S23, the number of product failure solution analysis σ and the number of analysis are set to a threshold σ yuzhi For numerical comparison, when σ≤σ yuzhi When σ>σ, continue to repeat steps S22 to S23; yuzhiWhen the process stops, the execution of steps S22 to S23 is repeated, and the corresponding solution data of the user product failure is output.
[0093] The product failure solution analysis and judgment unit uses an intelligent recognition algorithm to accurately identify the image features of the collected product failure point images and product failure solution data, intelligently analyze the solution to the product failure, realize online diagnosis of product failures and provide processing solutions, and improve the service efficiency and quality of the company's after-sales information management; the product failure solution analysis frequency statistics and judgment unit and the product failure solution result output unit cooperate with each other to self-check and confirm the product failure solution analysis process, thereby improving the reliability of the product failure solution analysis results, simplifying the company's product after-sales maintenance process, improving user recognition of the product, and improving the quality of the company's information management.
[0094] For further information, see Figure 1 - Figure 2 , when the number of product failure solution analysis times is greater than the analysis times setting threshold or the generated result of the user product failure corresponding solution data is not empty data, the final product failure solution analysis result data is output; when the final product failure solution analysis result data is output, the distance between the product failure location and different product failure maintenance sites is measured according to the coordinates of the user product failure location and the product failure maintenance site coordinates of different product failure maintenance sites, and the operation steps of generating the distance data from the product failure maintenance site to the product failure location are as follows:
[0095] S41, when σ > σ yuzhi When the generated result of the user product failure corresponding solution data is not empty data, the output user product failure corresponding solution data is identified as the final product failure solution analysis result data for output;
[0096] S42. Establish a product failure maintenance site coordinate set D = (d1, ..., d u ,…,d χ ), u=1,2,3,…,χ; where d u represents the corresponding product failure maintenance site coordinates of the u-th product failure maintenance site, χ represents the maximum number of product failure maintenance sites; d u =(x u ,y u ,z u ), where x u Indicates the longitude of the space where the product failure maintenance site is located, y u Indicates the latitude of the space where the product failure maintenance site is located, z u Indicates the altitude of the space where the product failure maintenance site is located;
[0097] S43, when outputting the final product failure solution analysis result data, according to the user product failure location coordinates b and the product failure maintenance site coordinates d in the product failure maintenance site coordinate set D u The distance formula is used to calculate the distance between the product failure location and different product failure maintenance sites and generate a distance data set from the product failure maintenance site to the product failure location E = (e 1,b ,…,e u,b ,…,e χ,b ), u=1,2,3,…,χ; where e u,b Indicates the distance data from the u-th product failure maintenance site to the location where the product failure occurred.
[0098] According to the distance data from the product failure maintenance site to the product failure location, a data search algorithm is used to perform numerical analysis, and the distance data from the product failure maintenance site to the product failure location with the smallest distance value is searched and marked. The operation steps for generating the minimum distance data from the product failure maintenance site to the faulty product are as follows:
[0099] S51, obtaining a data set E of the distance from the product failure maintenance site to the location where the product failure occurred;
[0100] S52, using a depth-first search algorithm, searching for the product failure maintenance site to product failure location distance data set E with the smallest value according to the value size. u,b And identify and generate the minimum distance data from the product failure maintenance site to the faulty product min .
[0101] The steps for selecting the coordinates of the corresponding maintenance site based on the minimum distance data between the product failure maintenance site and the faulty product and marking and generating the final coordinates of the product failure maintenance site are as follows:
[0102] S61. Based on the product failure, maintain the minimum distance data from the site to the faulty product. min Obtain the product fault maintenance site number u according to the product fault maintenance site number u min The corresponding product failure maintenance site coordinates d u And identify the coordinates of the final product failure maintenance site d queding .
[0103] The steps for feeding back the user's personal identity data and the final product failure solution analysis result data to the maintenance site control terminal and performing the product failure maintenance operation according to the final product failure maintenance site coordinates are as follows:
[0104] S71. Maintain site coordinates based on final product failure d quedingThe numbered location of the product fault maintenance site is obtained, and the user's personal identity data I and the final product fault solution analysis result data are pushed back to the maintenance site control end through the mobile communication network. The maintenance site control end contacts the user based on the user's personal identity data I and arranges staff to use the received final product fault solution analysis result data to perform online or offline product fault maintenance operations.
[0105] Through the cooperation between the product fault maintenance site and the faulty product distance measurement unit and the product fault maintenance site screening unit, the fault maintenance site closest to the product fault site is accurately measured and screened out, and the location information of the corresponding maintenance site is obtained, so as to realize the intelligent screening of product fault maintenance sites, reduce the time and economic cost of users to solve product faults, and enhance the competitiveness of enterprise product after-sales service and the enterprise information management capabilities; the product fault after-sales maintenance processing execution unit initiates the nearest product fault maintenance site control terminal to contact the user, and arranges maintenance personnel to perform online or offline product fault maintenance operations on the product fault problem, so as to realize the informatization and standardization of product after-sales maintenance operations, and enhance the intelligence of enterprise product after-sales information management.
[0106] A system for implementing an enterprise information management method based on artificial intelligence, the system comprising a product failure related information acquisition management module, a product failure solution management module, and a product failure processing execution management module;
[0107] The product failure related information acquisition management module includes a product failure acquisition unit, a user product failure location coordinate acquisition unit, and a user identity information acquisition unit;
[0108] The product failure collection unit collects product failure feature image data through the mobile terminal; the user product failure location coordinate collection unit collects the user product failure location coordinates through the mobile terminal; the user identity information collection unit collects the user's personal identity data through the mobile terminal;
[0109] The product failure solution management module includes a product failure solution storage unit, a product failure solution analysis and judgment unit, a product failure solution analysis times statistics and judgment unit, and a product failure solution result output unit;
[0110] A product failure solution storage unit is used to store product failure solution data; a product failure solution analysis and judgment unit is used to match the failure feature image with the product failure solution data through an image recognition algorithm, analyze and generate the user product failure corresponding solution data and judge the generation result of the user product failure corresponding solution data; a product failure solution analysis times counting and judgment unit is used to re-execute the product failure solution analysis step and count the product failure solution analysis times and compare the analysis times with the analysis times setting threshold value when the generation result of the user product failure corresponding solution data is empty data; a product failure solution result output unit is used to output the final product failure solution analysis result data when the product failure solution analysis times is greater than the analysis times setting threshold value or the generation result of the user product failure corresponding solution data is not empty data;
[0111] The product failure processing execution management module includes a product failure maintenance site coordinate storage unit, a product failure maintenance site and faulty product distance measurement unit, a product failure maintenance site screening unit, and a product failure after-sales maintenance processing execution unit;
[0112] A product failure maintenance site coordinate storage unit is used to store the coordinates of the product failure maintenance site; a product failure maintenance site and faulty product distance measurement unit, when outputting the final product failure solution analysis result data, measures the distance between the product failure location and different product failure maintenance sites based on the coordinates of the user's product failure location and the product failure maintenance site coordinates of different product failure maintenance sites, and generates distance data from the product failure maintenance site to the product failure location; a product failure maintenance site screening unit is used to perform numerical analysis based on the distance data from the product failure maintenance site to the product failure location using a data search algorithm, search for the product failure maintenance site with the smallest distance value to the product failure location and identify the minimum distance data from the product failure maintenance site to the faulty product, and screen out the product failure maintenance site coordinates of the corresponding maintenance site based on the minimum distance data from the product failure maintenance site to the faulty product, and identify the final product failure maintenance site coordinates; a product failure after-sales maintenance processing execution unit feeds back the user's personal identity data and the final product failure solution analysis result data to the maintenance site control end based on the final product failure maintenance site coordinates and executes the product failure maintenance operation.
[0113] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. An enterprise information management method based on artificial intelligence, characterized in that: The method comprises the following steps: S1. Collect product failure feature image data, coordinates of the user's product failure location, and user personal identity data; S2, matching the fault feature image with the product fault solution data through an image recognition algorithm, analyzing and generating the user product fault corresponding solution data and determining the generation result of the user product fault corresponding solution data; S3. When the generated result of the solution data corresponding to the user product failure is empty data, re-execute step S2 and calculate the number of product failure solution analysis times and compare the number with the analysis times threshold; S4. When the number of product failure solution analysis is greater than the analysis number setting threshold or the generated result of the user product failure corresponding solution data is not empty data, output the final product failure solution analysis result data; When the final product failure solution analysis result data is output, the distances between the product failure location and the different product failure maintenance sites are respectively measured based on the coordinates of the user product failure location and the product failure maintenance site coordinates of the different product failure maintenance sites, and the distance data from the product failure maintenance site to the product failure location is generated; S5. Perform numerical analysis using a data search algorithm based on the distance data from the product failure maintenance site to the product failure location, search for the distance data from the product failure maintenance site to the product failure location with the smallest distance value, and identify and generate the minimum distance data from the product failure maintenance site to the faulty product; S6. Filter out the coordinates of the product failure maintenance site corresponding to the maintenance site according to the minimum distance data from the product failure maintenance site to the faulty product and mark and generate the final product failure maintenance site coordinates; S7, feeding back the user personal identity data and the final product failure solution analysis result data to the maintenance site control terminal according to the final product failure maintenance site coordinates and performing product failure maintenance operations; The S2 comprises the following steps: S21. Establish product failure solution data set C = (c1,…,c n ,…,c β ), n=1,2,3,…,β; where c n Indicates the product failure solution data corresponding to the failure feature image of the nth product failure collected, β indicates the maximum value of the product failure solution data type, and the product failure solution includes one or more of a transmission device damage solution, a pump wear solution, a brake device failure solution, a device power loss solution, a motor failure solution, a transformer failure solution, a cable failure solution, a software startup failure solution, and a software operation abnormality solution; S22, using a data recognition algorithm to identify the product failure feature image data set A m and product failure solution data c in product failure solution data set C n According to the product failure feature image matching, the specific steps of the data recognition algorithm for product failure feature image recognition are as follows: S221, initialize and update the maximum number of iterations. In the product failure solution data set C constituting a β-dimensional optimization problem, the solution search sand cat is a 1×β array representing the problem solution, each variable value is a floating point number, and each variable value is between a lower bound and an upper bound. The fitness of each solution search sand cat is obtained by the fitness function of the problem to be solved; S222, search for prey, the final parameter and main parameter controlling the transition between the exploration and development phases is R. When |R|>1, the solution search cat searches for the product failure solution data set C with the product failure feature image data a m Matching product failure solution datac n ,The solution search sand cat search process relies on the release of low frequency noise; Each solution searches for the sand cat according to the best candidate position P best Update its own position and current position P and its sensitivity range r in the product failure solution data set C to find the image data a with product failure feature m Matching product failure solution datac n location; S223, attack prey, when |R|≤1, solution search sand cat for product failure solution data c n Find and use the best position P best Generate a random position with the current position P, set the sensitivity range of the solution search sand cat to be a circle, use the roulette method to randomly select a random angle θ for each solution search sand cat and perform product failure solution data c n search; S224, when the search algorithm reaches the maximum number of iterations, output a m With c n If the matching result is not reached, then continue to iterate and execute steps S222 to S223 until the maximum number of iterations is met; when a m With c n If the match is successful, the output data of the solution corresponding to the user product failure is when a m With c n If the match is not successful, the output data of the solution corresponding to the user product failure is None; S23, judging the generation result of the solution data corresponding to the user product failure, when the solution data corresponding to the user product failure is Then the generated result of outputting the solution data corresponding to the user product failure is not empty data; when the solution data corresponding to the user product failure is none, the generated result of outputting the solution data corresponding to the user product failure is empty data.
2. The enterprise information management method based on artificial intelligence according to claim 1, characterized in that: The S1 comprises the following steps: S11, collect product failure feature image data through mobile terminal shooting, and establish product failure feature image data set A = (a1, ..., a m ,…,a α ), m=1,2,3,…,α; where a m represents the mth product fault feature image data collected at the fault location of the product, and α represents the maximum number of product fault feature image data; The mobile terminal locates and collects the spatial position of the faulty product and generates the coordinates of the user product failure location b = (x b ,y b ,z b ), where x b Indicates the longitude of the space where the faulty product is located, y b Indicates the latitude of the space where the faulty product is located, z b Indicates the altitude of the space where the faulty product is located; The user enters personal identity information through the mobile terminal to complete the user identity information collection and generate user personal identity data I, where the personal identity data includes one or more of the user's name, contact information, and ID number.
3. The enterprise information management method based on artificial intelligence according to claim 2 is characterized by: The S3 comprises the following steps: S31. When the result of generating the solution data corresponding to the user product failure is empty data, re-execute steps S22 to S23 and simultaneously count the total number of times steps S22 to S23 are executed and generate the product failure solution analysis number σ; S32, setting the analysis number and setting the threshold σ yuzhi , where σ yuzhi The value of is a positive integer; yuzhi The larger the value of , the more times steps S22 to S23 are repeated when the result of generating the solution data corresponding to the user product failure is empty data; σ yuzhi The smaller the value of indicates that when the result of generating the solution data corresponding to the user product failure is empty data, the number of times steps S22 to S23 are repeatedly executed is less; S33, after each repetition of steps S22 to S23, the number of product failure solution analysis σ and the number of analysis are set to a threshold σ yuzhi For numerical comparison, when σ≤σ yuzhi When σ>σ, continue to repeat steps S22 to S23; yuzhi When the process stops, the steps S22 to S23 are repeated, and the solution data corresponding to the user product failure is output.
4. The enterprise information management method based on artificial intelligence according to claim 3 is characterized by: The S4 comprises the following steps: S41, when σ > σ yuzhi When the generated result of the user product failure corresponding solution data is not empty data, the output user product failure corresponding solution data is identified as the final product failure solution analysis result data for output; S42. Establish a product failure maintenance site coordinate set D = (d1, ..., d u ,…,d χ ), u=1,2,3,…,χ; where d u represents the corresponding product failure maintenance site coordinates of the u-th product failure maintenance site, χ represents the maximum number of product failure maintenance sites; d u =(x u ,y u ,z u ), where x u Indicates the longitude of the space where the product failure maintenance site is located, y u Indicates the latitude of the space where the product failure maintenance site is located, z u Indicates the altitude of the space where the product failure maintenance site is located; S43, when outputting the final product failure solution analysis result data, according to the user product failure location coordinate b and the product failure maintenance site coordinate d in the product failure maintenance site coordinate set D, u The distance formula is used to calculate the distance between the product failure location and different product failure maintenance sites and generate a distance data set from the product failure maintenance site to the product failure location E = (e 1,b ,…,e u,b ,…,e χ,b ), u=1,2,3,…,χ; where e u,b Indicates the distance data from the u-th product failure maintenance site to the location where the product failure occurred.
5. The enterprise information management method based on artificial intelligence according to claim 4 is characterized by: The S5 comprises the following steps: S51, obtaining a data set E of the distance from the product failure maintenance site to the location where the product failure occurred; S52, using a depth-first search algorithm, search for the product failure maintenance site to product failure location distance data set E with the smallest value according to the value size. u,b And identify and generate the minimum distance data from the product failure maintenance site to the faulty product min .
6. The enterprise information management method based on artificial intelligence according to claim 5 is characterized by: The S6 comprises the following steps: S61, maintaining the minimum distance data from the product failure site to the failed product min Obtain the product fault maintenance site number u according to the product fault maintenance site number u min The corresponding product failure maintenance site coordinates d u And identify the coordinates of the final product failure maintenance site d queding .
7. The enterprise information management method based on artificial intelligence according to claim 6 is characterized by: The S7 comprises the following steps: S71, maintaining the site coordinates d according to the final product failure queding The numbered location of the product fault maintenance site is obtained, and the user personal identity data I and the final product fault solution analysis result data are pushed back to the maintenance site control end through the mobile communication network. The maintenance site control end contacts the user based on the user personal identity data I and arranges staff to perform online or offline product fault maintenance operations using the received final product fault solution analysis result data.
8. A system for implementing the enterprise information management method based on artificial intelligence according to any one of claims 1 to 7.
