Machine-learning-based operation and maintenance method and system

Abstract

The present application relates to the technical field of operation and maintenance. Provided are a machine-learning-based operation and maintenance method and system. The method specifically comprises: collecting historical order data; establishing a customer relationship-machine learning model to predict the potential market value of processing orders for quartz devices; performing high-dimensional clustering analysis on the potential market value of the processing orders for quartz devices; calculating and acquiring the customer stability of relevant customers for different processing orders for quartz devices, and grading enterprise customers on the basis of the customer stability; and executing differentiated operation and maintenance services and personalized marketing plans. The present application solves the problem in the prior art of quartz processing plants suffering from poor enterprise customer stability and lacking personalized services due to the absence of different customized products for different customers.

Description

An Operation and Maintenance Method and System Based on Machine Learning Technical Field

[0001] This application relates to the field of operation and maintenance technology, and is an operation and maintenance method and system based on machine learning. Background Technology

[0002] Machine learning technology has been widely applied in data operations, especially in customer relationship management (CRM). By analyzing and processing customer data through machine learning algorithms, operations teams can better understand customer behavior patterns and demand trends. Furthermore, machine learning can help operations teams identify potential problems and risks, and through real-time monitoring and predictive analytics, proactively prevent possible system failures or service interruptions, thereby improving service stability and reliability.

[0003] Enterprise customer relationship management (CRM) refers to the operational management and maintenance work carried out by enterprises in the process of establishing and maintaining good relationships with customers. This includes understanding and responding to customer needs, communicating and exchanging ideas with customers regularly, providing professional services and support, and continuously monitoring customer satisfaction and changes in needs. Quartz processing companies primarily serve customers who require customized quartz devices or quartz glass products, whose product requirements may be more specific and specialized. Furthermore, unlike one-off transactions or short-term collaborations in other industries, quartz processing companies often require long-term, stable partnerships. Therefore, improving the intelligence of CRM and the continuity of technical services is a significant challenge currently facing quartz processing companies.

[0004] Among the existing publicly disclosed inventions, such as Chinese Patent Publication No. CN117829466A, a data-driven operation and maintenance service scheduling method is disclosed. This method includes: constructing a GIS electronic map corresponding to the service area based on the basic information of the enterprise being maintained and the maintenance personnel; in response to the maintenance request from the enterprise being maintained, creating an operation and maintenance service scheduling work order on the GIS electronic map, assigning maintenance personnel to the enterprise being maintained, and the status of the operation and maintenance service scheduling work order includes pending dispatch, pending acceptance, accepted, pending acceptance, completed, and terminated; in response to the maintenance progress feedback from the maintenance personnel, updating the status of the operation and maintenance service scheduling work order; and in response to the status of the operation and maintenance service scheduling work order being completed, performing operation and maintenance service analysis and evaluation on the maintenance personnel, and generating evaluation results.

[0005] The aforementioned patents primarily rely on basic information and maintenance requests for service scheduling, failing to fully consider customers' personalized needs and customized services. They are not suitable for the maintenance of enterprise customer relationships in quartz processing companies, resulting in a poor customer service experience. Summary of the Invention

[0006] The technical problem this application aims to solve is the lack of stable enterprise customers and personalized services for different customers in the existing quartz processing plants. It proposes an operation and maintenance method and system based on machine learning.

[0007] To achieve the above objectives, this application discloses a machine learning-based operation and maintenance method, including:

[0008] S1: Collect historical order data and perform data preprocessing on the collected historical order data to obtain processing data sequence and order data sequence;

[0009] S2: Establish a customer relationship-machine learning model. Input the processing data sequence and order data sequence into the customer relationship-machine learning model to predict the number of processing orders and the total order amount for quartz devices.

[0010] S3: Based on the number and total amount of quartz device processing orders, establish a cluster analysis model to perform high-dimensional cluster analysis on the number and total amount of quartz device processing orders;

[0011] S4: Based on the cluster analysis results and the stability calculation strategy of relevant customers, calculate the stability of relevant customers for different quartz device processing orders, and classify enterprise customers according to their stability to obtain enterprise customer classification data.

[0012] S5: Generate operation and maintenance decision data based on the enterprise customer's tiered data.

[0013] Specifically, the historical order data includes factory processing data and customer order information data; the factory processing data includes quartz device processing parameter data. and factory processing capacity data The quartz device processing parameter data includes the dimensions of the quartz device and the complexity of the processing steps; the factory processing capacity data includes: the average number of normally operating equipment, the average number of on-duty workers, and the factory's processing defect rate during the quartz device processing process; the customer order information data includes customer attention data. and customer spending power data The customer attention data includes the average time interval between the signing dates of adjacent customer orders and the number of customer communications and reminders during the quartz device processing process; and the collection of historical order data and the data preprocessing of the collected historical order data to obtain processing data sequences and order data sequences include: deduplicating historical order data and processing missing values; obtaining processing data sequences and order data sequences related to the signing month of historical order data.

[0014] Specifically, S2 includes:

[0015] S21: Obtain the time series of factory processing data and customer order information data, and perform wavelet decomposition on the time series of factory processing data and customer order information data.

[0016] When the month of signing the quartz device processing order is t, the periodic subsequence T for obtaining the quartz device processing parameter data is obtained. cs Trend subsequence R of quartz device processing parameter data cs C, random residual subsequences of quartz device processing parameter data cs ;

[0017] Obtain the periodic subsequence T of the factory's processing capacity data nl Trend subsequence R of factory processing capacity data nl C, a random residual subsequence of factory processing capacity data nl ;

[0018] The periodic subsequence T for obtaining customer attention data xw Trend subsequence R of customer attention data xw C, a random residual subsequence of customer attention data xw ;

[0019] S22: Establish a customer relationship-machine learning model, and import the subsequences obtained by decomposing the factory processing data time series and customer order information data time series into a sequence correlation fusion strategy to obtain a correlation sequence;

[0020] The sequence correlation fusion strategy is as follows:

[0021] The total number of historical order data collected is K, which includes K processing orders for quartz devices, where k is a superscript number representing the processing order for the kth quartz device, and k ≤ K;

[0022] These are time series forecasts for quartz device processing parameters, factory processing capacity, and customer attention.

[0023] μ 1,cs ,μ 2,cs ,μ 3,cs These are the fusion correlations of periodic subsequences, trend subsequences, and random residual subsequences of quartz device processing parameter data, respectively.

[0024] μ 1,nl ,μ 2,nl ,μ 3,nlThese are the fusion correlations of periodic subsequences, trend subsequences, and random residual subsequences of factory processing capacity data, respectively.

[0025] μ 1,xw ,μ 2,xw ,μ 3,xw These are the fusion correlations of periodic subsequences of customer attention data, the fusion correlations of trend subsequences of customer attention data, and the fusion correlations of random residual subsequences of customer attention data.

[0026] S23: After performing feature selection and feature scaling on the correlation sequence obtained in S22, input it into the customer relationship-machine learning model, and perform feature compression on the features obtained by feature selection to predict the potential market value of quartz device processing orders;

[0027] The potential market value of the quartz device processing orders Φ k for:

[0028] Among them, α1, α2, and α3 are the value factors of quartz device processing parameter data, factory processing capacity data, and customer attention data, respectively.

[0029] Specifically, in S3, the high-dimensional cluster analysis of the potential market value of the quartz device processing orders includes the following specific steps:

[0030] S31: The predicted potential market value of orders is organized into an N-dimensional data feature set L, denoted as L = {a k,n}, where a k,n Let the nth feature value be the value of the kth quartz device processing order; the data feature set is fuzzy processed using a membership function to obtain the fuzzy feature dataset H, denoted as H={b k,n}, where b k,n Corresponding to a k,n The feature values ​​after fuzzing; n represents the number of feature value categories;

[0031] S32: Based on S31, extract the fuzzy feature dataset H, and map the feature dataset H to a high-dimensional plane. Divide the predicted potential market value of orders into Q classes, and set M cluster center data points, where the cluster center data point set for each potential market value class is G = {g1, g2, ..., g...} m}, where the density of feature data points on the high-dimensional plane is ρ; g m Let M be the data point of the m-th cluster center; Q = M;

[0032] S33: Construct a membership degree of dk,m The classification matrix is ​​used to partition and cluster the feature dataset H on the high-dimensional plane, where d k,m Let be the membership degree of the k-th quartz device processing order to the m-th cluster center data point in the feature dataset H, where m∈(0,M).

[0033] Specifically, in S4, the stability W of the relevant customer is calculated using the following strategy:

[0034] Where Z represents the predicted number of quartz device processing orders;

[0035] Z is the value ranking coefficient of the m-th cluster center data point; W is the stability of the relevant customer; z is the order number; z≤Z;

[0036] w 0,z The stability of processing order data for the customer's zth processing order of the year;

[0037] The calculation strategy for the stability of the processing order data is as follows:

[0038] in, These are the predicted time series data for the quartz device processing parameters, factory processing capacity, and customer attention data for the z-th processing order, respectively.

[0039] These are the actual processing parameters, actual factory processing capacity, and actual customer attention data for the quartz components in the z-th processing order.

[0040] Specifically, in S4, the step of classifying enterprise customers based on their stability and obtaining the classification data of enterprise customers is as follows:

[0041] When the stability of a related customer is W≤θ0, the related customer is classified as an unstable relationship customer.

[0042] When the stability of the relevant customer is θ0<W≤θ1, the relevant customer is classified as a general stable relationship customer level;

[0043] When the stability of the relevant customer is θ1<W≤θ2, the relevant customer is classified as a customer with a particularly stable relationship.

[0044] Where θ0, θ1, and θ2 are the threshold values ​​for classifying customer stability levels, and θ0 < θ1 < θ2.

[0045] In addition, this application also discloses a machine learning-based operation and maintenance system. The system is used to implement the aforementioned machine learning-based operation and maintenance method, comprising: a historical data acquisition module for collecting historical order data and performing data preprocessing on the collected historical order data to obtain processing data sequences and order data sequences; a sequence prediction module for establishing a customer relationship-machine learning model, inputting the processing data sequences and order data sequences into the customer relationship-machine learning model to predict the number of quartz device processing orders and the total order amount; a cluster analysis module for establishing a cluster analysis model based on the number of quartz device processing orders and the total order amount, performing high-dimensional cluster analysis on the number of quartz device processing orders and the total order amount; a customer classification module for calculating the stability of relevant customers for different quartz device processing orders based on the cluster analysis results and the stability calculation strategy of relevant customers, classifying enterprise customers according to their stability, and obtaining enterprise customer classification data; and an operation and maintenance decision generation module for generating operation and maintenance decision data based on the enterprise customer classification data.

[0046] This application also discloses a storage medium storing instructions that, when read by a computer, cause the computer to execute the aforementioned machine learning-based operation and maintenance method.

[0047] This application also discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described machine learning-based operation and maintenance method. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0049] Figure 1 is a flowchart illustrating an operation and maintenance method based on machine learning according to this application;

[0050] Figure 2 is a schematic diagram of the structure of an operation and maintenance system based on machine learning according to this application. Detailed Implementation

[0051] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the specific embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0052] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0053] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of this application. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that excludes other embodiments.

[0054] Example 1

[0055] As shown in Figure 1, an operation and maintenance method based on machine learning disclosed in this application includes the following specific steps:

[0056] S1: Collect historical order data and perform data preprocessing on the collected historical order data to obtain processing data sequence and order data sequence;

[0057] In S1, the historical order data includes: factory processing data and customer order information data;

[0058] The factory processing data includes: quartz device processing parameter data. and factory processing capacity data

[0059] The quartz device processing parameters include: the dimensions of the quartz device and the complexity of the processing steps;

[0060] For example, in this embodiment, the processing complexity is the number of processes in the quartz device processing.

[0061] The factory processing capacity data includes: the average number of normally operating equipment in the factory during the quartz device processing process, the average number of on-duty workers in the factory, and the factory's processing defect rate;

[0062] The customer order information data includes: customer attention data. and customer spending power data

[0063] The customer attention data includes: the average time interval between the signing dates of adjacent customer orders and the number of times the customer communicates and urges customers during the quartz device processing process;

[0064] The process of collecting historical order data and performing data preprocessing on the collected historical order data to obtain processing data sequences and order data sequences includes: deduplicating historical order data and processing missing values; and obtaining processing data sequences and order data sequences related to the signing month of historical order data.

[0065] S2: Establish a customer relationship-machine learning model. Input the processing data sequence and order data sequence into the customer relationship-machine learning model to predict the number of processing orders and the total order amount for quartz devices.

[0066] S2 includes the following specific steps:

[0067] S21: Obtain the time series of factory processing data and customer order information data, and perform wavelet decomposition on the time series of factory processing data and customer order information data.

[0068] When the signing month of the historical order data is t, the periodic subsequence T of the quartz device processing parameter data is obtained. cs Trend subsequence R of quartz device processing parameter data cs C, random residual subsequences of quartz device processing parameter data cs ;

[0069] Obtain the periodic subsequence T of the factory's processing capacity data nl Trend subsequence R of factory processing capacity data nl C, a random residual subsequence of factory processing capacity data nl ;

[0070] The periodic subsequence T for obtaining customer attention data xw Trend subsequence R of customer attention data xw C, a random residual subsequence of customer attention data xw ;

[0071] S22: Establish a customer relationship-machine learning model, and import the subsequences obtained by decomposing the factory processing data time series and customer order information data time series into a sequence correlation fusion strategy to obtain a correlation sequence;

[0072] The sequence correlation fusion strategy is as follows:

[0073] The total number of historical order data collected is K, which includes K quartz device processing orders, where k is a superscript number representing the k-th quartz device processing order, and k ≤ K;

[0074] These are time series forecasts for quartz device processing parameters, factory processing capacity, and customer attention.

[0075] μ 1,cs ,μ 2,cs ,μ 3,cs These are the fusion correlations of periodic subsequences, trend subsequences, and random residual subsequences of quartz device processing parameter data, respectively.

[0076] μ 1,nl ,μ 2,nl ,μ 3,nl These are the fusion correlations of periodic subsequences, trend subsequences, and random residual subsequences of factory processing capacity data, respectively.

[0077] μ 1,xw ,μ 2,xw ,μ 3,xw These are the fusion correlations of periodic subsequences of customer attention data, the fusion correlations of trend subsequences of customer attention data, and the fusion correlations of random residual subsequences of customer attention data.

[0078] S23: After performing feature selection and feature scaling on the correlation sequence obtained in S22, input it into the customer relationship-machine learning model, and perform feature compression on the features obtained by feature selection to predict the potential market value of quartz device processing orders;

[0079] The potential market value of the quartz device processing orders Φ k for:

[0080] Among them, α1, α2, and α3 are the value factors of quartz device processing parameter data, factory processing capacity data, and customer attention data, respectively.

[0081] S3: Based on the number and total amount of quartz device processing orders, establish a cluster analysis model to perform high-dimensional cluster analysis on the number and total amount of quartz device processing orders;

[0082] In S3, the high-dimensional cluster analysis of the potential market value of the quartz device processing orders includes the following specific steps:

[0083] S31: The predicted potential market value of orders is organized into an N-dimensional data feature set L, denoted as L = {a k,n}, where a k,nLet the nth feature value be the value of the kth quartz device processing order; the data feature set is fuzzy processed using a membership function to obtain the fuzzy feature dataset H, denoted as H={b k,n}, where b k,n Corresponding to a k,n The feature values ​​after fuzzing; n represents the number of feature value categories;

[0084] S32: Based on S31, extract the fuzzy feature dataset H, and map the feature dataset H to a high-dimensional plane. Divide the predicted potential market value of orders into Q classes, and set M cluster center data points, where the cluster center data point set for each potential market value class is G = {g1, g2, ..., g...} m}, where the density of feature data points on the high-dimensional plane is ρ; g m Let M be the data point of the m-th cluster center; Q = M;

[0085] S33: Construct a membership degree of d k,m The classification matrix is ​​used to partition and cluster the feature dataset H on the high-dimensional plane, where d k,m Let be the membership degree of the k-th quartz device processing order to the m-th cluster center data point in the feature dataset H, where m∈(0,M).

[0086] S4: Based on the cluster analysis results and the stability calculation strategy of relevant customers, calculate the stability of relevant customers for different quartz device processing orders, and classify enterprise customers according to their stability to obtain enterprise customer classification data.

[0087] In S4, the specific strategy for calculating the stability W of the relevant customer is as follows:

[0088] Where Z represents the predicted number of quartz device processing orders;

[0089] Z is the value ranking coefficient of the m-th cluster center data point; W is the stability of the relevant customer; z is the order number; z≤Z;

[0090] w 0,z The stability of processing order data for the customer's zth processing order of the year;

[0091] The calculation strategy for the stability of the processing order data is as follows:

[0092] in, These are the predicted time series data for the quartz device processing parameters, factory processing capacity, and customer attention data for the z-th processing order, respectively.

[0093] These are the actual processing parameters, actual factory processing capacity, and actual customer attention data for the quartz components in the z-th processing order.

[0094] In S4, the specific steps for classifying enterprise customers based on their stability and obtaining their classification data are as follows:

[0095] When the stability of a related customer is W≤θ0, the related customer is classified as an unstable relationship customer.

[0096] When the stability of the relevant customer is θ0<W≤θ1, the relevant customer is classified as a general stable relationship customer level;

[0097] When the stability of the relevant customer is θ1<W≤θ2, the relevant customer is classified as a customer with a particularly stable relationship.

[0098] Where θ0, θ1, and θ2 are the threshold values ​​for classifying customer stability levels, and θ0 < θ1 < θ2.

[0099] S5: Generate operation and maintenance decision data based on the enterprise customer's tiered data.

[0100] For example, in this embodiment, the operation and maintenance decision data in S5 includes: differentiated operation and maintenance services and personalized marketing plans, specifically as follows: based on product characteristics and market positioning, personalized product recommendations, promotional activities and other marketing content are accurately pushed, and customized services and marketing plans are designed according to the characteristics of quartz devices.

[0101] Example 2

[0102] As shown in Figure 2, an operation and maintenance system based on machine learning disclosed in this application includes the following modules:

[0103] The historical data acquisition module is used to collect historical order data and perform data preprocessing on the collected historical order data to obtain processing data sequences and order data sequences;

[0104] The sequence prediction module is used to build a customer relationship-machine learning model. The processing data sequence and order data sequence are input into the customer relationship-machine learning model to predict the number of processing orders and the total order amount for quartz devices.

[0105] The clustering analysis module is used to establish a clustering analysis model based on the number and total amount of quartz device processing orders, and to perform high-dimensional clustering analysis on the number and total amount of quartz device processing orders.

[0106] The customer classification module is used to calculate the stability of relevant customers for different quartz device processing orders based on cluster analysis results and the stability calculation strategy of relevant customers, and to classify enterprise customers according to their stability to obtain enterprise customer classification data.

[0107] The operation and maintenance decision generation module is used to generate operation and maintenance decision data based on the hierarchical data of enterprise customers.

[0108] Example 3

[0109] This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0110] The processor executes the aforementioned machine learning-based operation and maintenance method by calling computer programs stored in memory.

[0111] The electronic device can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the machine learning-based operation and maintenance method provided in the above-described embodiment. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Further details are omitted in this embodiment.

[0112] Example 4

[0113] This embodiment proposes a computer-readable storage medium on which an erasable and rewritable computer program is stored.

[0114] When a computer program runs on a computer device, it causes the computer device to perform one of the aforementioned machine learning-based operation and maintenance methods.

[0115] For example, computer-readable storage media can be read-only memory (ROM), random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage devices.

[0116] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0117] It should be understood that determining B based on A does not mean determining B solely based on A; it also means determining B based on A and / or other information.

[0118] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired network and / or wireless network. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0119] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0121] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0122] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0123] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0124] In the description of this specification, the references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0125] In summary, compared with the prior art, the technical effects of this application are as follows:

[0126] 1. This application takes into account that the customers of quartz processing enterprises are different from those of other industries who engage in one-off transactions or short-term cooperation. They need to establish long-term and stable cooperative relationships. Through continuous monitoring and feedback mechanisms, differentiated operation and maintenance services and personalized marketing plans are implemented based on the graded data of enterprise customers to improve service quality and customer experience.

[0127] 2. This application, by collecting and preprocessing quartz device processing order data and conducting customer relationship analysis based on machine learning models, can more scientifically understand customer needs and market trends, thereby improving the accuracy and efficiency of decision-making.

[0128] 3. This application, through a customer relationship-machine learning model and an order value clustering analysis model, can accurately predict the potential market value of quartz device processing orders, which helps companies formulate more effective sales strategies and service plans.

[0129] 4. This application utilizes a customer stability calculation strategy to classify enterprise customers according to the stability of customer relationships, which is conducive to maintaining high-quality customer relationships and improving customer loyalty and customer satisfaction.

[0130] The foregoing has shown and described the basic principles, main features, and advantages of this application. Those skilled in the art should understand that this application is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this application. Various changes and modifications can be made to this application without departing from the spirit and scope thereof, and all such changes and modifications fall within the scope of this application as claimed. The scope of protection of this application is defined by the appended claims and their equivalents.

Claims

A machine learning-based operation and maintenance method, characterized in that, The method comprises: S1: collecting historical order data, and performing data preprocessing on the collected historical order data to obtain processing data sequence and order data sequence; S2: establishing a customer relationship-machine learning model, inputting the processing data sequence and the order data sequence into the customer relationship-machine learning model, and predicting the processing order quantity and the order total amount of the quartz device; S3: establishing a clustering analysis model according to the processing order quantity and the order total amount of the quartz device, and performing high-dimensional clustering analysis on the processing order quantity and the order total amount of the quartz device; S4: calculating the stability of the related customers of different quartz device processing orders according to the clustering analysis result and a related customer stability calculation strategy, and classifying the enterprise customers according to the stability of the customers to obtain classified data of the enterprise customers; S5: generating operation and maintenance decision data according to the classified data of the enterprise customers. The machine learning-based operation and maintenance method according to claim 1, characterized in that, The historical order data comprises factory processing data and customer order information data; The factory processing data includes quartz device processing parameter data and factory processing capability data The quartz device processing parameter data comprises the size of the quartz device and the processing procedure complexity; The factory processing capacity data comprises the average number of normal operation equipment of the factory, the average number of workers on duty of the factory, and the factory processing rejection rate during the processing of the quartz device; The customer order information data includes: customer attention data and customer spending power data The customer attention data comprises the average time interval of the signing dates of adjacent orders of the customer and the number of customer communication and urging during the processing of the quartz device; The collecting historical order data and performing data preprocessing on the collected historical order data to obtain processing data sequence and order data sequence comprises: removing duplicates and processing missing values of the historical order data; and obtaining processing data sequence and order data sequence related to the signing month of the historical order data. The machine learning-based operation and maintenance method according to claim 2, wherein S2 Comprise: S21: obtaining a factory processing data time sequence and a customer order information data time sequence, and performing wavelet decomposition on the factory processing data time sequence and the customer order information data time sequence; When the signing month of the historical order data is t, a period subsequence T of the quartz device processing parameter data is obtained cs , a trend subsequence R of the quartz device processing parameter data cs , a random residual subsequence C of the quartz device processing parameter data cs ; Periodic sub-sequence T of the acquisition of the factory processing capacity data nl Trend sub-sequence R of the factory processing capacity data nl Random residual sub-sequence C of the factory processing capacity data nl ; Periodic sub-sequence T of customer attention data acquisition xw Trend sub-sequence R of customer attention data xw Random residual sub-sequence C of customer attention data xw ; S22: establishing a customer relationship-machine learning model, inputting the sub-sequences obtained by decomposing the factory processing data time sequence and the customer order information data time sequence into a sequence correlation fusion strategy, and obtaining a correlation sequence; The sequence-dependent fusion strategy is as follows: Wherein, the total number of the collected historical order data is K, the historical order data comprises K processing orders of quartz devices, k is a superscript number, indicating the kth processing order of the quartz device, k≤K; Respectively, the quartz device processing parameter data prediction time sequence, the factory processing capacity data prediction time sequence and the customer attention data prediction time sequence; μ 1,cs ,μ 2,cs ,μ 3,cs are respectively a periodic subsequence fusion correlation degree of the quartz device processing parameter data, a trend subsequence fusion correlation degree of the quartz device processing parameter data, and a random residual subsequence fusion correlation degree of the quartz device processing parameter data; μ 1,nl ,μ 2,nl ,μ 3,nl are respectively the period subsequence fusion correlation degree of the factory processing capacity data, the trend subsequence fusion correlation degree of the factory processing capacity data, and the random residual subsequence fusion correlation degree of the factory processing capacity data. μ 1,xw ,μ 2,xw ,μ 3,xw are respectively a period subsequence fusion correlation of customer attention data, a trend subsequence fusion correlation of customer attention data, and a random residual subsequence fusion correlation of customer attention data. S23: inputting the correlation sequence obtained in S22 into the customer relationship-machine learning model after feature selection and feature scaling processing, and performing feature compression processing on the features obtained by feature selection and predicting the potential market value of the processing order of the quartz device; Potential market value of a processing order for the quartz device k is: Wherein, α1, α2, α3 are value factors of the quartz device processing parameter data, the factory processing capacity data and the customer attention data respectively. The machine learning-based operation and maintenance method according to claim 3, characterized in that, In S3, the high-dimensional clustering analysis on the potential market value of the processing order of the quartz device comprises the following specific steps: S31: the predicted potential market value of the order is composed into an N-dimensional data feature set L, denoted as L={a k,n}, wherein a k,n is the nth feature value of the kth quartz device processing order; the data feature set is processed by a membership function to obtain a feature data set H after fuzzy processing, denoted as H={b k,n}, wherein b k,n corresponds to a k,n feature value after fuzzy processing; n represents the number of feature values. S32: According to S31, the feature data set H after the blurring processing is extracted, and the feature data set H is mapped to a high-dimensional plane, the potential market value of the predicted order is divided into Q categories, and M cluster center data points are set, wherein the cluster center data point set of each potential market value category is G={g1, g2,..., gQ}, and the number of cluster center data points of each potential market value category is Gm. m}wherein the density of the feature data points on the high-dimensional plane is p; g m is the data point of the mth cluster center; Q=M; S33: Construct a classification matrix with membership d k,m , divide and cluster the feature data set H on a high-dimensional plane, where d k,m is the membership of the kth quartz device processing order in the feature data set H to the mth cluster center data point, where m∈(0,M]. The machine learning-based operation and maintenance method according to claim 4, characterized in that, In S4, the calculation strategy of the stability W of the relevant customer is specifically as follows: Wherein, Z is the predicted processing order quantity of the quartz device; is the value level coefficient of the mth cluster center data point; W is the stability of the relevant customer; z is the order number; z≤Z; w 0,z a stability of the processing order data for the zth processing order of the customer for the current year; The calculation strategy of the processing order data stability is as follows: wherein respectively, are the predicted time series of the quartz device processing parameter data, factory processing capacity data, and customer attention data of the zth processing order; respectively, are the actual processing parameter data, actual factory processing capacity data, and actual customer attention data of the zth processing order. The machine learning-based operation and maintenance method according to claim 5, wherein S4 The method comprises the following steps: when the stability of the relevant customer W≤θ0, the relevant customer is classified as an unstable relationship customer level; when the stability of the relevant customer θ0 when the stability of the relevant customer θ1 wherein θ0, θ1, and θ2 are respectively the classification thresholds of the customer stability level, and θ0<θ1<θ2. A machine learning based operation and maintenance system for implementing the machine learning based operation and maintenance method according to any one of claims 1-6, characterized in that, The system comprises: a historical data collection module, configured to collect historical order data, and perform data preprocessing on the collected historical order data to obtain processing data sequences and order data sequences; a sequence prediction module, configured to establish a customer relationship-machine learning model, input the processing data sequences and the order data sequences into the customer relationship-machine learning model, and predict the number of processing orders and the total order amount of the quartz device; a clustering analysis module, configured to establish a clustering analysis model according to the number of processing orders and the total order amount of the quartz device, and perform high-dimensional clustering analysis on the number of processing orders and the total order amount of the quartz device; a customer classification module, configured to calculate the stability of the relevant customer of different quartz device processing orders according to the clustering analysis result and a relevant customer stability calculation strategy, and perform hierarchical processing on the enterprise customers according to the stability of the customers to obtain hierarchical data of the enterprise customers; the operation and maintenance decision generation module, configured to generate operation and maintenance decision data according to the hierarchical data of the enterprise customers. A computer-readable storage medium having stored thereon a computer program, characterized in that The computer program is executed by a processor to implement the operation and maintenance method based on machine learning according to any one of claims 1-6. An electronic device, characterized by comprising: The computer program is executed by a processor to implement the operation and maintenance method based on machine learning according to any one of claims 1-6. The computer program is executed by a processor to implement the operation and maintenance method based on machine learning according to any one of claims 1-6. ​

Images