CN115630943A

CN115630943A - Intelligent scheduling method and device

Info

Publication number: CN115630943A
Application number: CN202211377392.4A
Authority: CN
Inventors: 徐林嘉; 陈李龙; 袁如怡; 李睿琦
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2022-11-04
Filing date: 2022-11-04
Publication date: 2023-01-20

Abstract

The application provides an intelligent scheduling method and device, which can be used in the financial field or other fields. The method comprises the following steps: acquiring real-time network point data acquired by an operation and maintenance system; analyzing the real-time network point data to obtain the scheduling number through a scheduling prediction model constructed by time sequence decomposition and a graph neural network algorithm; and generating the scheduling data of the website according to the scheduling number. According to the scheduling prediction module based on time sequence decomposition and the graph neural network, intelligent prediction of network point scheduling work is achieved, and the problem that the conventional manual scheduling is too dependent on intuition and is not accurate is solved.

Description

Intelligent scheduling method and device

Technical Field

The application relates to the technical field of website analysis, can be applied to the financial field and other fields, and particularly relates to an intelligent scheduling method and device.

Background

Artificial intelligence is a novel strategic technology leading the future, and is an important force for driving a new technological revolution and product revolution. At present, the continuous evolution of the artificial intelligence related technology and the continuous acceleration of industrialization and commercialization progress are accelerating to be fused with the deep of thousands of industries. In this context, machine learning models play an important role in the fields of computer vision, natural language processing, speech recognition, intelligent wind control, precision marketing, smart cities and the like. In the banking industry, banking outlets are important contacts for interaction between banks and customers.

However, the scheduling of bank outlet workers has been a significant problem. In the past, bank business outlets perform work scheduling work by means of subjective judgment and historical experience, and the randomness of staff vacation arrangement is high, so that the situation that the staff cannot deal with the peak of passenger flow often occurs, the staff delay goes off duty, and the service experience of customers is influenced.

Disclosure of Invention

The embodiment of the application mainly aims to provide an intelligent scheduling method and device, and solves the problems that the scheduling work of the current network point depends too much on manual intuition, so that staff at the network point cannot schedule accurately, staff arrangement quantity is often insufficient, customer service experience is poor, or too many staff are arranged, and resource waste is caused.

In order to achieve the above object, the present application provides an intelligent shift scheduling method, including: acquiring real-time network point data acquired by an operation and maintenance system; analyzing the real-time network point data through a scheduling prediction model constructed by time sequence decomposition and a graph neural network algorithm to obtain the scheduling number; and generating the scheduling data of the website according to the scheduling number.

In the above intelligent shift scheduling method, optionally, the obtaining of the shift scheduling amount by analyzing the real-time mesh data through a shift scheduling prediction model constructed by time sequence decomposition and a graph neural network algorithm includes: inputting the real-time network point data into a time sequence decomposition model in the scheduling prediction model to obtain an influence parameter; and analyzing through a timing diagram neural network model in the scheduling prediction model according to the real-time mesh data and the influence parameters to obtain the scheduling quantity.

In the above intelligent scheduling method, optionally, the real-time website data includes: the number of the users waiting for the network, the number of the persons waiting for the network, special items and geographical positions.

In the above intelligent scheduling method, optionally, the step of inputting the real-time website data into the time sequence decomposition model in the scheduling prediction model to obtain an influence parameter includes: analyzing the real-time network point data through the time sequence decomposition model to obtain influence factors, trend terms, period terms and error terms; and taking the influence factor, the trend term, the period term and the error term as influence parameters.

In the above intelligent shift scheduling method, optionally, the method further includes: obtaining historical mesh point data, and obtaining an influence parameter by using a time sequence decomposition algorithm according to the historical mesh point data; and training an initial scheduling model through the influence parameters and the historical mesh point data by a graph neural network algorithm to obtain the scheduling prediction model.

In the above intelligent scheduling method, optionally, the obtaining the scheduling prediction model by training an initial scheduling model through the influence parameter and the historical mesh point data by using a graph neural network algorithm includes: generating a graph neural network through a graph neural network algorithm according to the influence parameters and the historical dot data; and training an initial scheduling model through a time recursion neural network according to the graph neural network to obtain the scheduling prediction model.

In the above intelligent shift scheduling method, optionally, the generating of the shift scheduling data of the website according to the shift scheduling number includes: extracting prestored website personnel information according to the scheduling number; and generating personnel data to be selected according to the personnel information of the website, and screening according to the scheduling number and the personnel data to be selected to obtain scheduling data.

The application also provides an intelligence device of scheduling, the device includes: the acquisition module is used for acquiring real-time website data acquired by the operation and maintenance system; the analysis module is used for analyzing the real-time network point data to obtain the scheduling number through a scheduling prediction model constructed by time sequence decomposition and a graph neural network algorithm; and the processing module is used for generating the network point scheduling data according to the scheduling number.

The application also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the method.

The present application also provides a computer-readable storage medium storing a computer program for executing the above method.

The present application also provides a computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the above method.

The intelligent scheduling system based on the time sequence decomposition and the graph neural network achieves intelligent prediction of scheduling work of the network points through an innovative scheduling prediction module based on the time sequence decomposition and the graph neural network by acquiring information such as the number of waiting clients, the number of receptionists, special items and geographic positions of the network points at a certain time, and solves the problem that manual scheduling in the past depends too much on intuition and is not accurate.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.

Fig. 1 is a flowchart of an intelligent scheduling method according to an embodiment of the present application;

FIG. 2 is a logic diagram illustrating the generation of a shift prediction model according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a process for obtaining a shift scheduling prediction model in an embodiment of the present application;

FIG. 4 is a schematic diagram of the structure of a knowledge-graph in an embodiment of the present application;

FIG. 5 is a schematic structural diagram of a neural model with timing diagrams according to an embodiment of the present disclosure;

FIG. 6 is a schematic diagram of a shift scheduling prediction model according to another embodiment of the present application;

FIG. 7 is a schematic diagram illustrating a process of generating a shift amount according to an embodiment of the present application;

FIG. 8 is a schematic diagram illustrating a process for generating scheduling data of an outlet in yet another embodiment of the present application;

FIG. 9 is a schematic view of an application flow of an intelligent shift scheduling method according to still another embodiment of the present application;

FIG. 10 is a schematic diagram illustrating an intelligent shift scheduling device according to yet another embodiment of the present application;

FIG. 11 is a schematic diagram illustrating an application structure of an intelligent shift scheduling device according to yet another embodiment of the present application;

fig. 12 is a schematic structural diagram of an electronic device according to an embodiment of the application.

Detailed Description

The embodiment of the application provides an intelligent operation and maintenance method and device, which can be used in the financial field and other fields, and it should be noted that the intelligent operation and maintenance method and device can be used in the financial field and any fields except the financial field.

The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

Fig. 1 is a flowchart illustrating an intelligent shift scheduling method according to an embodiment of the present application, where the method includes:

s101, acquiring real-time network point data acquired by an operation and maintenance system;

s102, analyzing the real-time network point data through a scheduling prediction model constructed by time sequence decomposition and a graph neural network algorithm to obtain the scheduling quantity;

s103, generating the scheduling data of the network points according to the scheduling number.

In the embodiment, the intelligent scheduling prediction of the network node scheduling work is realized by acquiring information construction characteristics such as the number of waiting clients, the number of receptionists, special matters, geographic positions and the like at a certain time through an innovative scheduling prediction module based on time sequence decomposition and a graph neural network, so that the client experience is improved, the staff resources are effectively utilized, and the waste of the staff resources is reduced. Wherein the real-time mesh point data may include: the number of the network point reception users, the number of the reception personnel, special items, geographical positions and the like.

Referring to fig. 2, in an embodiment of the present application, the method further includes:

s201, obtaining historical mesh point data, and obtaining an influence parameter by using a time sequence decomposition algorithm according to the historical mesh point data;

s202, training an initial scheduling model through the influence parameters and the historical dot data through a graph neural network algorithm to obtain the scheduling prediction model.

Further, referring to fig. 3, the obtaining of the shift scheduling prediction model by training the initial shift scheduling model through the graph neural network algorithm according to the influence parameters and the historical dot data includes:

s301, generating a graph neural network through a graph neural network algorithm according to the influence parameters and the historical dot data;

s302, training an initial scheduling model through a time recursion neural network according to the graph neural network to obtain the scheduling prediction model.

Specifically, in actual work, the haba bara hayata prediction model is an outbound call promotion degree prediction model based on a gain model, and comprises a time sequence decomposition model and a time sequence neural network model, and the specific construction logic is as follows:

a time sequence decomposition model: y (t) = s (t) + g (t) + p (t) + e (t);

wherein y (t) is the numerical value of a certain index at t time, and the index can be the number of the clients in the network at a certain day, the number of the clients in the service consultation at a certain day and the total queuing waiting time of the clients at a certain day. In the present application, y (t) is decomposed into the following 4 items:

s (t) is the value of the influence factors of the special items of the t time (such as holiday influence degree, sports events, important items and the like);

g (t) is a trend term value of t time;

p (t) is the value of the periodic term at time t (e.g., quarterly effects);

e (t) is the value of the error term at time t.

The numerical values of the terms are calculated as follows:

and the s (t) special item influence factor value is obtained by calculation according to whether special item factors occur at the moment t of artificial input, and represents the influence factor value brought by the special item at the moment t. Input is x _{s_t} The input numerical value indicates the degree of influence that the matter may have, 0 is no special matter occurs, and the upper limit is 10; s (t) is then calculated by a generalized linear model. I.e., s (t) = g ^-1 (w ^T x _{s_t} + b), where g is the scaling function, w is the weight coefficient matrix, and b is the bias vector, to be fitted in subsequent calculations.

And g (t) the trend term value is determined according to y (t) and s (t) in a period of time before the t moment, and represents the trend value at the t moment. Is calculated in a manner of

Where T is the length of the set time window, and fitting is performed in subsequent calculations.

The value of the p (t) period term is determined according to y (t), s (t) and g (t) in a period of time before the t moment, and represents the value of the period term at the t moment. The calculation method is

Where N is the length of the set time window, a _ N and b _ N are coefficients, and p (t) will be oriented in the direction of [ y (t) -s (t) -g (t)]And (6) fitting.

The e (t) error term value represents the magnitude of the error existing in time series decomposition at the time t, and is calculated in a mode of e (t) = y (t) -s (t) -g (t) -p (t).

Therefore, the number of the first and second electrodes is increased,

y(t)＝s(t)+g(t)+p(t)+e(t)；

s(t)＝g ^-1 (w ^T x _{s_t} +b)；

e(t)＝y(t)-s(t)-g(t)-p(t)。

through the decomposition mode, the numerical value at a certain time t can be decomposed into a special item influence factor numerical value, a trend item numerical value, a period item numerical value and an error item numerical value; the values involved all contribute to the subsequent calculations.

With respect to the timing graph neural network, it is first indicated that the graph is essentially composed of nodes, each Node representing an entity existing in the real world, and edges, each Edge representing a relationship between an entity and a Node. In short, a knowledge graph is a relationship network that links together all of the different kinds of information, and this relationship network can be known over time.

In this scenario, each mesh point is taken as a node. The information of the geographical position, the size, the mechanism grade and the like of each network point is used as the attribute of the node, and meanwhile, the special item influence factor value, the trend item value, the period item value and the error item value of various data (such as the number of queuing people and the number of staff in the network points) obtained by the time sequence decomposition method at the time t are also used as the attribute information of the node of the graph at the time t. Further, the distance between mesh points is taken as an edge weight in each graph node. Thereafter, through a time chart neural network algorithm, the change and trend of the chart structure along with time are analyzed, and the number of employees to be scheduled on a certain day is predicted. Specifically, referring to fig. 4, the differences between the nodes and the edges represent different types, and it can be seen that the structure of the graph changes significantly from time t to time t + 1.

Referring to fig. 5, in order to realize the time sequence dynamic learning in the model of the timing neural network (T-GNN), the present application provides the following two points: 1. learning a graph neural network (GCN) separately for each time slice, each GCN input being different and being represented by a different adjacency matrix of the graph; 2. to account for dynamic map association, sequence learning is performed by concatenating each time slice GCN model parameter using the LSTM algorithm.

The T-GNN uses an LSTM model to perform series learning on the parameters, and the hidden state of the LSTM model uses the parameters at the previous moment

With the input also using the last time parameter

The update formula is as follows:

thus, node Embedding

And

the update formula of (c) is as follows,

1:function

2

3

4:end function

the model structure can be seen in fig. 6, where GCN is a graph convolution neural network, which acts as a feature extractor, but its object is graph data, in fact, as the convolution neural network CNN.

In the graph data, there are N nodes (nodes), each node has its own features, the features of the nodes are configured to form an N × D matrix X, and then the relationship between the nodes also forms an N × N matrix a, which is also called an adjacency matrix (adjacency matrix). X and A are the inputs to the model.

The GCN is also a neural network layer, and the propagation modes among layers are as follows:

in this formula:

a waves = a + I, I is the identity matrix;

the D wave is the degree matrix (degree matrix) of the A wave, and the formula is

H is a characteristic of each layer, and H is X for the input layer;

σ is a nonlinear activation function.

The two models are uniformly trained in an end-to-end mode, original characteristic data are used as input, and a prediction result of a timing graph neural network is used as final output; the model is trained with the aim of reducing Mean Absolute Error (MAE), and all parameters to be fitted are fitted in the training process.

The calculation mode of the MAE is as follows:

wherein y _ predict is the number of employees to be arranged by the website of model prediction, y _ actual is the number of employees to be arranged by the website of manual annotation, and n is the number of training samples.

Referring to fig. 7, in an embodiment of the present application, analyzing the real-time mesh data to obtain the shift scheduling amount through a shift scheduling prediction model constructed by a time sequence decomposition and a graph neural network algorithm includes:

s701, inputting the real-time network point data into a time sequence decomposition model in the scheduling prediction model to obtain an influence parameter;

s702, analyzing through a timing diagram neural network model in the scheduling prediction model according to the real-time network point data and the influence parameters to obtain the scheduling quantity.

Wherein, the time sequence decomposition model for inputting the real-time network point data into the scheduling prediction model to obtain the influence parameters may include: analyzing the real-time mesh data through the time sequence decomposition model to obtain influence factors, trend terms, period terms and error terms; and taking the influence factor, the trend term, the period term and the error term as influence parameters.

Referring to fig. 8, in an embodiment of the present application, the generating of the website shift data according to the shift amount includes:

s801, extracting pre-stored website personnel information according to the scheduling quantity;

s802, generating data of people to be selected according to the personnel information of the website, and screening according to the scheduling number and the data of the people to be selected to obtain scheduling data.

In the embodiment, the personnel information of the website can be prestored in a designated position, and then the corresponding personnel information of the website can be called after the scheduling number is determined, and the personnel with the corresponding number is selected as scheduling data according to the scheduling number; certainly, in the actual use process, a scheduling rule may also be preset, and screening is performed in the information of the website personnel based on the scheduling rule, and relevant technicians in the field may select and use the scheduling rule according to actual needs, which is not further limited in this application.

In order to make the overall process of the intelligent shift scheduling method provided by the present application more clearly understood, please refer to fig. 9, which is a flowchart illustrating the above embodiment as a whole.

The process of characteristic information acquisition, time sequence decomposition and graph neural network prediction and scheduling in the intelligent scheduling system of the network points based on the time sequence decomposition and the graph neural network comprises the following four steps:

the method comprises the following steps: and acquiring characteristic data. Acquiring information construction characteristics such as the number of waiting clients, the number of waiting workers, special items, geographic positions and the like of a certain time network point;

step two: and (5) training a model. And training a time sequence decomposition model and a time sequence neural network model by using historical characteristic data and the manually marked number of staff of the network points to be arranged. (wherein the links of the model training have been described in detail in the foregoing embodiments and are not described in detail here)

Step three: and predicting the number of staff to be arranged at a website at a certain time in the future by using the latest feature data obtained in the first step and the intelligent scheduling model obtained by training in the second step.

Step four: and (4) arranging the number of the staff of the network point according to the predicted value obtained in the third step.

Referring to fig. 10, the present application further provides an intelligent shift scheduling apparatus, including: the acquisition module is used for acquiring real-time website data acquired by the operation and maintenance system; the analysis module is used for analyzing the real-time network point data to obtain the scheduling number through a scheduling prediction model constructed by time sequence decomposition and a graph neural network algorithm; and the processing module is used for generating the network point scheduling data according to the scheduling number.

As shown in fig. 11, in actual work, the acquisition module, i.e., the feature information acquisition portion, mainly completes information acquisition, i.e., acquiring and extracting features such as the number of customers waiting at a node t, the number of workers waiting at a node t, special events (festivals) at a node t, geographical locations of nodes, and other information; then, entering an analysis module constructed by a time sequence decomposition model and a time sequence neural network to complete data analysis to obtain the corresponding number of forecasted people; and finally, the number of workers to be scheduled at a certain time in the future and the scheduling of the network points are predicted, namely the processing module finishes the scheduling. The process comprises the following steps:

s01, a characteristic information acquisition unit: the method comprises the following steps that information construction characteristics such as the number of waiting clients, the number of waiting workers, special matters, geographic positions and the like are received by a certain time network point; this information is then passed to the S02 module.

S02, an outbound time sequence decomposition and graph neural network module: and the S02 module receives the relevant characteristic information transmitted by the S01 module. And (3) transmitting the prediction result to an S03 module through an innovative scheduling prediction model based on time sequence decomposition and a graph neural network.

S03 subsequent processing unit: and the S03 module receives the predicted number of the workers to be arranged at a certain time in the future sent by the S02 unit and carries out task scheduling of the network-site workers according to the prediction result.

According to the network point intelligent scheduling system based on the time sequence decomposition and the graph neural network, information such as the number of waiting clients, the number of receptionists, special items and geographic positions of network points in a certain time is obtained, and an innovative scheduling prediction module based on the time sequence decomposition and the graph neural network is used for achieving intelligent prediction of network point scheduling work, so that the problem that the conventional manual scheduling is too dependent on intuition and is not accurate is solved.

The present application also provides a computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the above-described method.

As shown in fig. 12, the electronic device 600 may further include: communication module 110, input unit 120, audio processor 130, display 160, power supply 170. It is noted that the electronic device 600 does not necessarily include all of the components shown in FIG. 12; furthermore, the electronic device 600 may also comprise components not shown in fig. 12, which may be referred to in the prior art.

As shown in fig. 12, the central processor 100, sometimes referred to as a controller or operational control, may include a microprocessor or other processor device and/or logic device, the central processor 100 receiving input and controlling the operation of the various components of the electronic device 600.

The memory 140 may be, for example, one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The information relating to the failure may be stored, and a program for executing the information may be stored. And the cpu 100 may execute the program stored in the memory 140 to realize information storage or processing, etc.

The input unit 120 provides input to the cpu 100. The input unit 120 is, for example, a key or a touch input device. The power supply 170 is used to provide power to the electronic device 600. The display 160 is used to display an object to be displayed, such as an image or a character. The display may be, for example, but is not limited to, an LCD display.

The memory 140 may be a solid state memory such as Read Only Memory (ROM), random Access Memory (RAM), a SIM card, or the like. There may also be a memory that holds information even when power is off, can be selectively erased, and is provided with more data, an example of which is sometimes called an EPROM or the like. The memory 140 may also be some other type of device. Memory 140 includes buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application/function storage section 142 for storing application programs and function programs or a flow for executing the operation of the electronic device 600 by the central processing unit 100.

The memory 140 may also include a data store 143 for storing data, such as contacts, digital data, pictures, sounds, and/or any other data used by the electronic device. The driver storage portion 144 of the memory 140 may include various drivers of the electronic device for communication functions and/or for performing other functions of the electronic device (e.g., messaging application, address book application, etc.).

The communication module 110 is a transmitter/receiver 110 that transmits and receives signals via an antenna 111. The communication module (transmitter/receiver) 110 is coupled to the central processor 100 to provide an input signal and receive an output signal, which may be the same as in the case of a conventional mobile communication terminal.

Based on different communication technologies, a plurality of communication modules 110, such as a cellular network module, a bluetooth module, and/or a wireless local area network module, may be provided in the same electronic device. The communication module (transmitter/receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132 to implement general telecommunications functions. Audio processor 130 may include any suitable buffers, decoders, amplifiers and so forth. In addition, an audio processor 130 is also coupled to the central processor 100, enabling recording locally through a microphone 132, and enabling locally stored sound to be played through a speaker 131.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The principle and the implementation of the present application are explained by applying specific embodiments in the present application, and the description of the above embodiments is only used to help understand the method and the core idea of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims

1. An intelligent scheduling method, characterized in that the method comprises:

acquiring real-time network point data acquired by an operation and maintenance system;

analyzing the real-time network point data through a scheduling prediction model constructed by time sequence decomposition and a graph neural network algorithm to obtain the scheduling number;

and generating the scheduling data of the network points according to the scheduling number.

2. The intelligent scheduling method of claim 1 wherein analyzing the real-time mesh point data to obtain the scheduling number through a scheduling prediction model constructed by time-series decomposition and a graph neural network algorithm comprises:

inputting the real-time network point data into a time sequence decomposition model in the scheduling prediction model to obtain an influence parameter;

and analyzing through a timing diagram neural network model in the scheduling prediction model according to the real-time mesh data and the influence parameters to obtain the scheduling quantity.

3. The intelligent scheduling method of claim 2 wherein the real-time website data comprises: the number of the network point reception users, the number of the waiting users, the number of the reception personnel, special items and geographic positions.

4. The intelligent scheduling method of claim 2 wherein inputting the real-time mesh point data into a timing decomposition model in the scheduling prediction model to obtain an impact parameter comprises:

analyzing the real-time mesh data through the time sequence decomposition model to obtain influence factors, trend terms, period terms and error terms;

and taking the influence factor, the trend term, the period term and the error term as influence parameters.

5. The intelligent shift scheduling method of claim 1, wherein the method further comprises:

obtaining historical mesh point data, and obtaining an influence parameter by using a time sequence decomposition algorithm according to the historical mesh point data;

and training an initial scheduling model through the influence parameters and the historical dot data by a graph neural network algorithm to obtain the scheduling prediction model.

6. The intelligent scheduling method of claim 5 wherein the training of an initial scheduling model through a neural network algorithm with the impact parameters and the historical dot data to obtain the scheduling prediction model comprises:

generating a graph neural network through a graph neural network algorithm according to the influence parameters and the historical mesh data;

and training an initial scheduling model through a time recursion neural network according to the graph neural network to obtain the scheduling prediction model.

7. The intelligent scheduling method of claim 1, wherein generating the website scheduling data according to the scheduling number comprises:

extracting prestored website personnel information according to the scheduling number;

and generating personnel data to be selected according to the personnel information of the website, and screening according to the scheduling number and the personnel data to be selected to obtain scheduling data.

8. An intelligent shift arrangement device, characterized in that, the device includes:

the acquisition module is used for acquiring real-time website data acquired by the operation and maintenance system;

the analysis module is used for analyzing the real-time network point data to obtain the scheduling number through a scheduling prediction model constructed by time sequence decomposition and a graph neural network algorithm;

and the processing module is used for generating the network point scheduling data according to the scheduling number.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method of any one of claims 1 to 7.

11. A computer program product comprising computer program/instructions, characterized in that said computer program/instructions, when executed by a processor, implement the steps of the method of any of claims 1 to 7.