Scheduling data processing method and device

Abstract

The invention relates to the technical field of machine learning, and particularly discloses a scheduling data processing method and device.The method comprises the steps that according to business volume influencing factor data and business volume data in a historical time period and business volume influencing factor data in a future target time period, the business volume influencing factor data in the future target time period are calculated; determining target business volume data in a future target time period; according to the scheduling data, the employee work record data and the attendance influencing factor data in the historical time period and the attendance influencing factor data in the future target time period, determining availability data of each employee in the plurality of employees in the future target time period; based on the target business volume data in the future target time period and the availability data of each employee in the plurality of employees, generating a preliminary scheduling scheme in the future target time period; and optimizing the preliminary scheduling scheme by adopting a multi-objective optimization algorithm to obtain a target scheduling scheme in a future target time period. According to the scheme, the scheduling efficiency and the management level can be improved.

Description

Technical Field

[0001] The present invention relates to the field of machine learning technology, and more particularly to a method and device for processing shift scheduling data. Background Art

[0002] At present, the employee scheduling of enterprises and institutions can be carried out through the automatic scheduling system. The existing automatic scheduling system has problems such as insufficient data processing capabilities, low accuracy of prediction models, and single optimization strategies. Specifically, in terms of data processing, due to the system's lack of attention to data collection, cleaning, and integration, a large amount of valuable information is omitted or ignored, resulting in insufficient processing depth and breadth. The accuracy of the prediction model is one of the important indicators to measure the performance of the automatic scheduling system. However, the existing system often has large errors and uncertainties when predicting employee behavior. In addition, the existing automatic scheduling system often lacks flexibility and diversity in optimization strategies, making it difficult to meet the diverse actual needs of enterprises.

[0003] To address the above problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of this specification provide a method and device for processing shift scheduling data to solve the problem of insufficient accuracy of automatic shift scheduling in the prior art.

[0005] The embodiment of this specification provides a method for processing shift scheduling data, including:

[0006] Acquire data on factors affecting business volume and business volume data in a historical time period and data on factors affecting business volume in a future target time period; determine target business volume data in a future target time period based on the data on factors affecting business volume and business volume data in a historical time period and data on factors affecting business volume in a future target time period;

[0007] Obtaining the scheduling data, employee work record data, and data on factors affecting attendance in a historical time period, as well as the data on factors affecting attendance in a future target time period; determining availability data of each of a plurality of employees in a future target time period based on the scheduling data, employee work record data, and data on factors affecting attendance in a historical time period, as well as the data on factors affecting attendance in a future target time period;

[0008] generating a preliminary shift scheduling plan for the future target time period based on the target business volume data in the future target time period and the availability data of each of the multiple employees;

[0009] A multi-objective optimization algorithm is used to optimize the preliminary scheduling plan to obtain a target scheduling plan for a future target time period; the objective function of the multi-objective optimization algorithm includes multiple target items, and the multiple target items include at least two of the following: a workload matching target item, a workload balance target item, a skill matching target item, and a personal preference and operational demand target item.

[0010] In one embodiment, the data of factors influencing business volume includes at least one of the following: weather data, business activity data and business indicator data; the data of factors influencing attendance includes at least one of the following: weather data and employee personal information.

[0011] In one embodiment, the target traffic volume data for a future target time period is determined based on the data of factors influencing traffic volume and traffic volume data for the historical time period and the data of factors influencing traffic volume for a future target time period, including: preprocessing the data of factors influencing traffic volume and traffic volume data for the historical time period to obtain a training sample set; training a preset prediction model based on the training sample set to obtain a target traffic volume prediction model; inputting the data of factors influencing traffic volume for the future target time period into the target traffic volume prediction model to obtain the target traffic volume data for the future target time period.

[0012] In one embodiment, availability data of each employee among multiple employees in a future target time period is determined based on the scheduling data, employee work record data and data on factors affecting attendance in the historical time period and data on factors affecting attendance in a future target time period, including: preprocessing the scheduling data, employee work record data and data on factors affecting attendance in the historical time period to obtain a training sample set; training a long short-term memory model using the training sample set to obtain a target long short-term memory model; inputting the data on factors affecting attendance in the future target time period into the target long short-term memory model to obtain availability data of each employee among the multiple employees in the future target time period.

[0013] In one embodiment, a preliminary scheduling plan for the future target time period is generated based on the target business volume data in the future target time period and the availability data of each of the multiple employees, including: determining at least one scheduling time period and the business volume data of each scheduling time period in the at least one scheduling time period according to the target business volume data in the future target time period; determining the number of employees required for each scheduling time period according to the business volume data of each scheduling time period in the at least one scheduling time period; generating a preliminary scheduling plan for the future target time period based on the availability data of each of the multiple employees in each scheduling time period in the at least one scheduling time period and the number of employees required for each scheduling time period.

[0014] In one embodiment, a multi-objective optimization algorithm is used to optimize the preliminary scheduling plan to obtain a target scheduling plan for a future target time period, including: using a particle swarm algorithm to optimize the preliminary scheduling plan to obtain a target scheduling plan for a future target time period.

[0015] The embodiment of this specification also provides a scheduling data processing device, including:

[0016] The first determination module is used to obtain the data of factors affecting business volume and business volume data in the historical time period and the data of factors affecting business volume in the future target time period; and determine the target business volume data in the future target time period according to the data of factors affecting business volume and business volume data in the historical time period and the data of factors affecting business volume in the future target time period;

[0017] The second determination module is used to obtain the scheduling data, employee work record data and attendance factor data in the historical time period and the attendance factor data in the future target time period; determine the availability data of each employee among the multiple employees in the future target time period according to the scheduling data, employee work record data and attendance factor data in the historical time period and the attendance factor data in the future target time period;

[0018] A generating module, configured to generate a preliminary shift scheduling plan for a future target time period based on the target business volume data in the future target time period and the availability data of each of the multiple employees;

[0019] An optimization module is used to optimize the preliminary scheduling plan using a multi-objective optimization algorithm to obtain a target scheduling plan for a future target time period; the objective function of the multi-objective optimization algorithm includes multiple target items, and the multiple target items include at least two of the following: a workload matching target item, a workload balance target item, a skill matching target item, and a personal preference and operation demand target item.

[0020] The embodiments of the present specification also provide a computer device, including a processor and a memory for storing instructions executable by the processor, and the processor implements the steps of the scheduling data processing method described in any of the above embodiments when executing the instructions.

[0021] The embodiments of the present specification also provide a computer-readable storage medium on which computer instructions are stored. When the instructions are executed by a processor, the steps of the scheduling data processing method described in any of the above embodiments are implemented.

[0022] The embodiments of this specification also provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the scheduling data processing method described in any of the above embodiments.

[0023] In an embodiment of the present specification, a scheduling data processing method is provided, which determines the target business volume data in the future target time period according to the data of factors affecting business volume and business volume in the historical time period and the data of factors affecting business volume in the future target time period, determines the availability data of each of multiple employees in the future target time period according to the scheduling data, employee work record data and data of factors affecting attendance in the historical time period and data of factors affecting attendance in the future target time period, generates a preliminary scheduling plan for the future target time period based on the target business volume data in the future target time period and the availability data of each of multiple employees, and takes into account the changes in business volume and the availability data of employees at the same time, ensures that there are enough people during high workload periods, and reduces overtime during low workload periods, so as to achieve the best balance between efficiency and cost, and can also give priority to scheduling employees with low leave probability and willingness to work overtime as predicted results. Afterwards, the preliminary scheduling plan is optimized using a multi-objective optimization algorithm to obtain a target scheduling plan for the future target time period. When looking for the optimal scheduling plan, we can comprehensively consider multiple aspects such as workload matching, employee workload balance, skill matching, personal preferences, and corporate operation needs, so as to improve the rationality and scientificity of the scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of this specification, constitute a part of this specification, and do not constitute a limitation of this specification. In the drawings:

[0025] Figure 1 A flow chart of a method for processing shift data in one embodiment of this specification is shown;

[0026] Figure 2 A schematic diagram of modules used for a method for processing shift data in an embodiment of this specification is shown;

[0027] Figure 3 A flow chart of a method for processing shift data in one embodiment of this specification is shown;

[0028] Figure 4 A flowchart of a PSO algorithm for optimizing a shift scheduling scheme in an embodiment of this specification is shown;

[0029] Figure 5 A schematic diagram of a shift scheduling data processing device in an embodiment of the present specification is shown;

[0030] Figure 6A schematic diagram of a computer device in an embodiment of the present specification is shown. DETAILED DESCRIPTION

[0031] The principles and spirit of this specification will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement this specification, and are not intended to limit the scope of this specification in any way. On the contrary, these embodiments are provided to make this specification more thorough and complete, and to fully convey the scope of this disclosure to those skilled in the art.

[0032] Those skilled in the art will appreciate that the embodiments of this specification may be implemented as a system, device, method, or computer program product. Therefore, this specification may be implemented in the following forms: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0033] It should be noted that the user-related information and data involved in the embodiments of this specification are all information and data authorized by the user or fully authorized by relevant parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users or relevant parties to choose to authorize or refuse.

[0034] It should also be noted that in the embodiments of this specification, certain software, components, models and other existing solutions in the industry may be mentioned. They should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0035] The embodiments of this specification provide a method for processing shift scheduling data. Figure 1 A flow chart of a method for processing shift data in an embodiment of the present specification is shown. Although the present specification provides method operation steps or device structures as shown in the following embodiments or drawings, more or fewer operation steps or module units may be included in the method or device based on routine or without creative labor. In the steps or structures where there is no necessary causal relationship logically, the execution order of these steps or the module structure of the device is not limited to the execution order or module structure described in the embodiments of the present specification and shown in the drawings. When the method or module structure is applied to an actual device or terminal product, it can be connected according to the method or module structure shown in the embodiments or drawings for sequential execution or parallel execution (for example, a parallel processor or a multi-threaded processing environment, or even a distributed processing environment).

[0036] Specifically, Figure 1 As shown, a method for processing shift data provided by an embodiment of the present specification may include the following steps.

[0037] Step S101, obtaining data on factors influencing business volume and business volume data in a historical time period and data on factors influencing business volume in a future target time period; determining target business volume data in a future target time period based on the data on factors influencing business volume and business volume data in a historical time period and data on factors influencing business volume in a future target time period.

[0038] The method in this embodiment can be applied to a computer device or an application in a computer device. The computer device can be a server or a client. Data of factors affecting business volume and the amount of business volume in a historical time period and data of factors affecting business volume in a future target time period can be obtained. Among them, data of factors affecting business volume refers to data of factors affecting the business volume of enterprises and institutions, which can be selected according to specific actual conditions. Data of factors affecting business volume can be actual data of factors affecting business volume in a historical time period, or predicted data or actual data of factors affecting business volume in a future target time period.

[0039] In one embodiment, the data of factors affecting business volume may include: weather data, business activity data and business indicator data. Business activity data may be some business activities of enterprises and institutions. For example, for bank branches, there may be various activities such as card application lucky draw activities and deposit and financial management gift activities. Business indicator data may refer to the business indicator requirements of enterprises and institutions, and business indicator data may be positively correlated with business volume.

[0040] After acquiring the data, the target traffic volume data in the future target time period can be determined based on the traffic volume influencing factor data and traffic volume data in the historical time period and the traffic volume influencing factor data in the future target time period. Specifically, the correlation between the traffic volume influencing factor data and traffic volume data can be determined based on the traffic volume influencing factor data and traffic volume data in the historical time period, and then based on the correlation, combined with the traffic volume influencing factor data in the future target time period, the target traffic volume data in the future target time period can be determined.

[0041] Step S102, obtaining the scheduling data, employee work record data and attendance factor data within the historical time period and the attendance factor data within the future target time period; determining the availability data of each employee among multiple employees within the future target time period based on the scheduling data, employee work record data and attendance factor data within the historical time period and the attendance factor data within the future target time period.

[0042] It is also possible to obtain the scheduling data, employee work record data, and data on factors affecting attendance in the historical time period, as well as data on factors affecting attendance in the future target time period. Among them, the scheduling data may be data such as the scheduling time and the number of schedulings of multiple employees in the historical time period. The employee work record data may include vacation records, leave applications, attendance, overtime records, work task information, etc. of multiple employees in the historical time period. Data on factors affecting attendance refer to data on factors that affect the scheduling of employees of enterprises and institutions, and can be selected according to specific actual conditions. Data on factors affecting attendance may be actual data on factors affecting scheduling in the historical time period, or may be predicted data or actual data on factors affecting scheduling in the future target time period. Data on factors affecting attendance include at least one of the following: weather data and employee personal information. Employee personal information may include information such as the employee's address, commuting time, and commuting method.

[0043] After acquiring the data, the availability data of each of the multiple employees in the future target time period can be determined based on the scheduling data, employee work record data and data on factors affecting attendance in the historical time period and data on factors affecting attendance in the future target time period. Specifically, the correlation between the data on factors affecting attendance and employee attendance can be determined based on the scheduling data, employee work record data and data on factors affecting attendance in the historical time period, and then based on the correlation and combined with the data on factors affecting attendance in the future target time period, the availability data of each of the multiple employees in the future target time period can be determined. The employee's availability data refers to the employee's availability and work status in each scheduling time period in the future target time period.

[0044] Step S103: generating a preliminary shift scheduling plan for the future target time period based on the target business volume data in the future target time period and the availability data of each of the multiple employees.

[0045] Step S104, using a multi-objective optimization algorithm to optimize the preliminary scheduling plan to obtain a target scheduling plan for the future target time period; the objective function of the multi-objective optimization algorithm includes multiple target items, and the multiple target items include at least two of the following: a workload matching target item, a workload balance target item, a skill matching target item, and a personal preference and operation demand target item.

[0046] After obtaining the target business volume data and the availability data of each employee in the future target time period, a preliminary scheduling plan for the future target time period can be generated. The preliminary scheduling plan may include data such as multiple scheduling time periods, employee identification corresponding to each scheduling time period, etc. A multi-objective optimization algorithm can be used to optimize the preliminary scheduling plan to obtain a target scheduling plan for the future target time period. The objective function of the multi-objective optimization algorithm includes multiple objective items, and the multiple objective items include at least two of the following: a workload matching objective item, a workload balance objective item, a skill matching objective item, and a personal preference and operation demand objective item. The workload matching objective item is used to measure the degree of matching between the scheduled workload and the predicted workload. The workload balance objective item is used to measure the degree of balance between the workloads of different employees. The skill matching objective item is used to measure the degree of matching between employees and the job requirements to which they are assigned. The personal preference and operation demand objective item is used to measure the degree of balance between the personal preferences of employees and the operation requirements of enterprises and institutions. By setting multiple target items, we can comprehensively consider multiple aspects such as workload matching, employee workload balance, skill matching, personal preferences, and corporate operation needs, thereby improving the rationality of scheduling.

[0047] The method in the above embodiment determines the target business volume data in the future target time period according to the data of factors affecting business volume and business volume in the historical time period and the data of factors affecting business volume in the future target time period, determines the availability data of each employee in the future target time period according to the scheduling data, employee work record data and data of factors affecting attendance in the historical time period and data of factors affecting attendance in the future target time period, generates a preliminary scheduling plan for the future target time period based on the target business volume data in the future target time period and the availability data of each employee in the multiple employees, and considers the change of business volume and the availability data of employees at the same time, ensures that there are enough people during the high workload period, and reduces overtime during the low workload period, so as to achieve the best balance between efficiency and cost, and can also give priority to scheduling employees with low leave probability and willing to work overtime. Afterwards, the preliminary scheduling plan is optimized by using a multi-objective optimization algorithm to obtain the target scheduling plan for the future target time period. When looking for the optimal scheduling plan, multiple aspects such as workload matching, employee workload balance, skill matching, personal preference and enterprise operation needs can be comprehensively considered, so as to improve the rationality and scientificity of scheduling.

[0048] In some embodiments of the present specification, target traffic volume data for a future target time period is determined based on the data of factors influencing traffic volume and traffic volume data for the historical time period and the data of factors influencing traffic volume for a future target time period, including: preprocessing the data of factors influencing traffic volume and traffic volume data for the historical time period to obtain a training sample set; training a preset prediction model based on the training sample set to obtain a target traffic volume prediction model; inputting the data of factors influencing traffic volume for the future target time period into the target traffic volume prediction model to obtain the target traffic volume data for the future target time period.

[0049] Specifically, the data of factors affecting business volume and business volume data in the historical time period can be preprocessed to obtain a training sample set. Among them, the preprocessing can include cleaning the data, eliminating invalid data and correcting erroneous information. Then, fill in the missing values. Subsequently, these data are integrated according to a unified standard and format to form a data set for analysis. Finally, in order to meet the requirements of the machine learning model, the system converts the data into a numerical format and performs normalization. After preprocessing, a training sample set is obtained. After that, the preset prediction model can be trained based on the training sample set. Among them, the preset prediction model can be a machine learning regression model. The preset prediction model can be trained based on the training sample set to obtain a target business volume prediction model. After that, the data of factors affecting business volume in the future target time period can be input into the target business volume prediction model to obtain the target business volume data in the future target time period. The target business volume data may include the workload of each of multiple positions in the future target time period. The workload can be measured by working hours and / or the number of work tasks. In the above manner, the business volume can be predicted by machine learning, which can improve the efficiency and accuracy of business volume prediction.

[0050] In some embodiments of the present specification, target business volume data for a future target time period is determined based on the data of factors affecting business volume and business volume data for the historical time period and the data of factors affecting business volume for the future target time period, including: performing time series analysis on the data of factors affecting business volume and business volume data for the historical time period and the data of factors affecting business volume for the future target time period to obtain the target business volume data for the future target time period.

[0051] In some embodiments of the present specification, availability data of each employee among multiple employees in a future target time period is determined based on the scheduling data, employee work record data and data on factors affecting attendance in the historical time period and data on factors affecting attendance in the future target time period, including: preprocessing the scheduling data, employee work record data and data on factors affecting attendance in the historical time period to obtain a training sample set; training a long-short-term memory model using the training sample set to obtain a target long-short-term memory model; inputting the data on factors affecting attendance in the future target time period into the target long-short-term memory model to obtain availability data of each employee among the multiple employees in the future target time period.

[0052] In this embodiment, the scheduling data, employee work record data and attendance factor data in the historical time period can be preprocessed to obtain a training sample set. The collected data can be cleaned to remove invalid data and correct erroneous information, for example, to correct the inconsistency of employee ID numbers. Next, fill in missing values, such as using the average or median to fill in the blank data in the employee attendance record. Subsequently, the system integrates these data according to a unified standard and format to form a data set for analysis. Finally, in order to meet the requirements of the deep learning model, the system converts the data into a numerical format and performs normalization processing so that all data are at the same level, thereby ensuring that the data quality meets the high standards for model training. Afterwards, the training sample set can be used to train the long-term and short-term memory model to obtain a target long-term and short-term memory model. The attendance factor data in the future target time period is input into the target long-term and short-term memory model to obtain the availability data of each of the multiple employees in the future target time period. The availability data may include information such as the availability and work status of each of the multiple employees in each of the multiple scheduling time periods.

[0053] In some embodiments of the present specification, a preliminary scheduling plan for the future target time period is generated based on the target business volume data in the future target time period and the availability data of each of the multiple employees, including: determining at least one scheduling time period and the business volume data of each scheduling time period in the at least one scheduling time period according to the target business volume data in the future target time period; determining the number of employees required for each scheduling time period according to the business volume data of each scheduling time period in the at least one scheduling time period; generating a preliminary scheduling plan for the future target time period based on the availability data of each of the multiple employees in each scheduling time period in the at least one scheduling time period and the number of employees required for each scheduling time period.

[0054] Specifically, at least one scheduling time period and the business volume data of each scheduling time period within the at least one scheduling time period can be determined based on the target business volume data within the future target time period. The business volume data may include the positions and the corresponding business volumes within each scheduling time period. According to the business volume data of each scheduling time period within the at least one scheduling time period, the number of employees required for each position in each scheduling time period is determined. Based on the availability data of each employee among the multiple employees in each scheduling time period in the at least one scheduling time period and the number of employees required for each scheduling time period, a preliminary scheduling plan for the future target time period can be generated. In the above manner, preliminary scheduling can be performed.

[0055] In some embodiments of the present specification, a multi-objective optimization algorithm is used to optimize the preliminary scheduling plan to obtain a target scheduling plan for a future target time period, including: using a particle swarm algorithm to optimize the preliminary scheduling plan to obtain a target scheduling plan for a future target time period.

[0056] In this embodiment, the particle swarm algorithm can be used to optimize the preliminary scheduling plan to obtain the target scheduling plan for the future target time period. The PSO algorithm searches for the optimal solution by simulating the foraging behavior of bird flocks. The algorithm comprehensively considers multiple goals such as the workload balance, skill matching, personal preferences, and business operation needs of employees. For example, when predicting whether an employee is suitable for a certain shift, the system will not only consider whether the employee's technical expertise matches the job requirements, but also evaluate his recent workload to ensure that an employee will not work continuously for too long, thereby affecting his work efficiency and health. Through multiple iterative adjustments and optimizations of the PSO algorithm, the system can find a balance point so that the scheduling plan not only meets the business's operational needs, but also takes into account the personal situation of employees. In this way, a balance point can be found between multiple goals, such as the workload balance, skill matching, personal preferences, and business operation needs of employees, ensuring the rationality and scientificity of the scheduling plan.

[0057] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. For details, please refer to the description of the above-mentioned related processing related embodiments, and no further description is given here.

[0058] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0059] The above method is described below in conjunction with a specific embodiment. However, it should be noted that the specific embodiment is only for better illustrating the present specification and does not constitute an improper limitation on the present specification.

[0060] This specific embodiment provides a method for processing shift scheduling data. The purpose of this embodiment is to achieve refined automatic shift scheduling based on deep learning and big data analysis technology, build a high-precision employee behavior prediction model by deeply mining the hidden information in historical shift scheduling data, and formulate a scientific and reasonable shift scheduling optimization strategy on this basis to achieve a comprehensive upgrade and transformation of enterprise shift scheduling management.

[0061] The method in this embodiment can adopt a modular design. Figure 2 , shows a schematic diagram of modules for implementing the scheduling data processing method in this embodiment. Figure 2 As shown in the figure, it can be mainly implemented by the following modules: data collection and preprocessing module, deep learning model training module, shift prediction and optimization module, shift plan generation and adjustment module, user interaction and feedback module. Please refer to Figure 3 , shows a flow chart of the method for processing shift data in this specific embodiment. Figure 3 As shown, the method in this embodiment may include the following steps.

[0062] Step 1: Data collection and preprocessing.

[0063] The scheduling system first needs to comprehensively collect and integrate multi-source heterogeneous data, including historical scheduling records, employee personal information (such as vacation records, leave applications, attendance, overtime records, etc.), and work task information. These data are distributed in different departments and databases of the company. For example, employee vacation records may be stored in the human resources management system, while overtime records may exist in the financial system. In order to ensure the comprehensiveness and accuracy of the data, the system automatically captures data from these scattered systems regularly through the data acquisition module and stores them in a unified data warehouse. In the preprocessing stage, the system first cleans the collected data, removes invalid data and corrects erroneous information, for example, corrects the inconsistency of employee ID numbers. Then, the system fills in missing values, such as using the mean or median to fill in blank data in employee attendance records. Subsequently, the system integrates these data according to unified standards and formats to form a data set for analysis. Finally, in order to meet the requirements of the deep learning model, the system converts the data into a numerical format and normalizes it so that all data are at the same level, thereby ensuring that the data quality meets the high standards required for model training.

[0064] Step 2: Deep learning model training.

[0065] In the deep learning model training module, the system deeply applies the LSTM (Long Short-Term Memory) neural network architecture to cope with the time series characteristics in the scheduling data. The core of LSTM lies in its unique unit structure, including the forget gate, input gate and output gate, which work together on the unit state to capture the time dependency of employee work patterns.

[0066] The system will notice that an employee works less overtime on weekends and more on weekdays. During model training, the data is first divided into training, validation, and test sets to ensure the generalization ability of the model. Initial weights and biases are set when the model is initialized, and then these parameters are adjusted through multiple iterations of training to improve prediction accuracy. Specifically, the forget gate determines which old state information should be retained, and the input gate determines which new information should be added to the current state. The unit state update combines the results of the forget gate and the input gate to update the current state. The output gate determines what information to output based on the updated unit state. For example, in a typical training cycle, the system will continuously adjust the parameters of the forget gate and the input gate so that the model can better remember the information that is useful for future predictions and forget the irrelevant information. In this way, the system can more accurately predict the availability and work status of employees at a certain point in the future.

[0067] Once the LSTM model predicts which employee is most likely to work overtime or take leave in the coming week, the shift prediction and optimization module will adjust the shift schedule based on these predictions. For example, if the model predicts that employee A is likely to work overtime in the coming week, while employee B is more likely to take leave, the system will initially reduce employee B's shift hours and arrange more overtime hours for employee A in case of emergency.

[0068] The system will generate a preliminary draft schedule for each employee based on information such as employee performance, skill matching, and workload balance in historical data, combined with prediction results. A preliminary scheduling model can be used to process a large amount of feature data by building a decision tree to randomly sample the training set. A preliminary draft schedule can be generated using the preliminary scheduling model based on information such as employee performance, skill matching, and workload balance in historical data, combined with prediction results. This draft will take into account employee availability and work needs to ensure that the scheduling plan meets both the business's operational needs and the employees' personal circumstances.

[0069] For example, if the prediction results show that employee C has an 80% probability of needing to work overtime every day in the next week, the system will use this predicted probability as an important factor in scheduling, and may arrange employee C to work during peak working hours and reserve overtime time to cope with the possible workload.

[0070] The specific algorithm formula is as follows:

[0071] (1) Forget gate: determines which information needs to be forgotten.

[0072] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0073] Among them, f t is the output of the forget gate, σ is the sigmoid activation function, W f and b f

[0074] is the weight and bias of the forget gate, h t-1 is the hidden state of the previous time step, x t is the input for the current time step.

[0075] (2) Input gate: determines what new information needs to be added to the unit state.

[0076]

[0077] Among them, i t is the candidate cell state, Wi , W C , b i and b C are the weights and biases of the input gate and candidate cell state.

[0078] (3) Unit state update: The unit state is updated by combining the results of the forget gate and the input gate.

[0079]

[0080] Among them, C t is the cell state at the current time step.

[0081] (4) Output gate: Determines what information to output based on the updated cell state.

[0082] o t =σ(W o ·[h t-1 ,x t ]+b o )][h t =o t tanh(C t )

[0083] Among them, t is the output of the output gate, h t is the hidden state of the current time step, W o and b o are the weight and bias of the output gate.

[0084] Model training includes data partitioning, model initialization, parameter tuning, and training iteration. Through multiple iterations of training, the system continuously optimizes model parameters, improves prediction accuracy, and uses validation sets to ensure the generalization of the model on unknown data. Use the trained LSTM model to make preliminary shift predictions. The system can predict which employee is most likely to work overtime or take leave in the next week, so as to prepare for staff deployment in advance.

[0085] In addition to predicting the likelihood of employees working overtime or taking leave, machine learning algorithms can also be used to predict the next workload to determine the number of employees who need to work overtime and the positions that need to work overtime. For example, time series analysis or machine learning regression models can be used to predict future workloads.

[0086] For time series analysis algorithms, ARIMA (autoregressive integrated moving average model) or seasonal decomposition time series forecasting models (such as SARIMA) can be used to predict workload. Input data may include historical workload data, holiday information, special events, etc., and the output data is the predicted value of future workload.

[0087] For regression models, linear regression or logistic regression models can be used to predict workload and overtime positions. Input data may include historical workload, employee skill level, project requirements, etc. The output data is the predicted value of workload and the prediction of overtime positions.

[0088] When making preliminary shift scheduling based on the LSTM prediction results and workload prediction results, the system will consider the employee's availability (including overtime and leave predictions) and the predicted workload to generate a preliminary scheduling plan. For example, if the forecast shows that the workload of a position will increase in the next week, the system will prioritize those employees who are predicted to have a low probability of leaving and are willing to work overtime.

[0089] When optimizing PSO, the objective function does need to further consider the workload objective. You can add an objective related to workload, such as: This objective will help the optimization algorithm take into account the workload changes when finding the optimal scheduling plan, ensuring that there are enough people during high workload periods and reducing overtime during low workload periods to achieve the best balance between efficiency and cost.

[0090] Step 3: Schedule prediction and optimization.

[0091] To further optimize the shift schedule, the system uses a multi-objective optimization algorithm - particle swarm optimization (PSO). The PSO algorithm searches for the optimal solution by simulating the foraging behavior of bird flocks. The algorithm comprehensively considers multiple objectives such as employee workload balance, skill matching, personal preferences, and corporate operational needs. Please refer to Figure 4 , shows a flow chart of using PSO algorithm to optimize the shift scheduling scheme. Figure 4 As shown in the figure, when predicting whether an employee is suitable for a certain shift, the system will not only consider whether the employee's technical expertise matches the job requirements, but also evaluate his recent workload to ensure that an employee will not be asked to work for too long in a row, thereby affecting his work efficiency and health. Through multiple iterative adjustments and optimizations of the PSO algorithm, the system can find a balance point so that the shift scheduling plan not only meets the company's operational needs, but also takes into account the personal situation of employees.

[0092] The core lies in the update formula of particle speed and position:

[0093]

[0094] in, and are the velocity and position of particle i at the kth iteration in the dth dimension, w is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers, is the individual best historical position of particle i in the dth dimension, gbest kis the best historical position of the group, which can be achieved through the following specific methods:

[0095] The objective function construction includes the following contents.

[0096] 1. Construction of target items.

[0097] (1) Workload balance:

[0098] In order to measure the balance of workload, a cost function related to the working hours and intensity of employees can be defined. represents the working hours of employee i in time period j, represents the work intensity of employee i in time period j (which can be comprehensively evaluated by factors such as workload and difficulty), then the overall workload of employee i can be expressed as:

[0099]

[0100] where w t and w i are the weights of working hours and work intensity, respectively. One of the goals of the system is to minimize the variance of all employees’ workloads to ensure balance: Minimize(Var(WL i ))

[0101] (2) Skill matching:

[0102] Skill matching can be quantified by evaluating the degree of matching between employee skills and job requirements. ik is the score of employee i on skill k, R jk is the demand degree of job j for skill k, then the skill matching degree SkillMatchij can be expressed as:

[0103]

[0104] The system seeks to maximize the sum of the skill matching of all employees:

[0105]

[0106] Among them, x ij is a decision variable, indicating whether employee i is assigned to position j.

[0107] (3) Personal preferences and operational needs:

[0108] Personal preferences and operational needs can be reflected by setting preference coefficients and demand coefficients. For example, for employee i’s preference for shift j, a preference coefficient P can be set. ij ; For the operation demand of the enterprise in shift j, a demand coefficient D can be set jOperational demand refers to the company's demand for employees in a specific shift, such as specific skills, a specific number of employees, etc. Combining these two, we can construct the following target item PreferenceAndDemand:

[0109]

[0110] Among them, λ is the balance coefficient, which is used to adjust the weight between personal preference and operational demand. This objective item is to balance the personal preference of employees and the operational demand of the enterprise, which is adjusted by the preference coefficient and the demand coefficient. The optimization goal is determined according to the specific situation, which may be to maximize employee satisfaction (or minimize dissatisfaction) or minimize operating costs (or maximize operating efficiency).

[0111] (4) Workload matching target items:

[0112] You can define an objective term to minimize the difference between the effort and the forecasted effort. For example:

[0113]

[0114] Among them, f 工作量 (x) is the workload matching target item. wi is the actual workload, is the predicted workload, and n is the number of positions.

[0115] 2. Construction of Constraints

[0116] (1) Each employee must be assigned to a shift. Shifts can include normal shifts, overtime shifts, and special shifts. Normal shifts: such as morning shifts, mid-day shifts, and evening shifts. Overtime shifts: overtime on weekdays and overtime on weekends. Special shifts: such as holiday shifts, night shifts, etc.

[0117]

[0118] (2) The needs of each shift must be met (assuming there are minimum and maximum employee limits):

[0119]

[0120] (3) Skill matching threshold. SkillMatchij is the skill matching of employee i to shift j. Thresholdij is the skill matching threshold of employee i to shift j. This means that only when the matching degree between the employee's skills and the job requirements exceeds this threshold, the employee is considered suitable for the job.

[0121] SkillMatchij≥Thresholdij, if x ij =1

[0122] (4) Workload constraints. Ensure that the workload of each position does not exceed its maximum capacity, for example:

[0123]

[0124] Among them, w i is the workload of the ith position, W max It is the maximum workload of the position.

[0125] In the PSO algorithm, each particle represents a possible scheduling solution, and its position is determined by all x ij The fitness function (i.e., objective function) of the algorithm can be constructed by integrating all the above-mentioned objective items and constraints. By iteratively updating the speed and position of particles, the PSO algorithm can gradually approach the optimal scheduling solution.

[0126] In this scheme, the fitness function needs to comprehensively consider multiple objectives and constraints such as employee workload balance, skill matching, personal preferences, enterprise operation needs, workload, etc.

[0127] In one embodiment, the fitness function can be determined based on multiple target items and / or multiple constraints and multiple preset weights. The multiple preset weights can be the weights corresponding to the penalty items corresponding to each target item in the multiple target items and each constraint in the multiple constraints. The sum of the multiple preset weights is 1. When constructing the fitness function, it is necessary to ensure that all target items are in the same optimization direction (all maximized or all minimized). If the target items have different optimization directions, the directions can be unified by negation or normalization. Weights can be used to balance the importance of different target items. Constraints can be added to the fitness function through penalty items to ensure the feasibility of the scheduling plan.

[0128] In one embodiment, if a constraint is that each employee must be assigned to a shift, the penalty term can be defined as:

[0129]

[0130] Among them, a i is a binary variable, which is 1 if employee i is assigned to a shift, otherwise 0. If ai = 0, the contribution of the penalty term to the fitness function is 1, otherwise 0. In this way, employees who are not assigned to a shift will lead to a decrease in the value of the fitness function.

[0131] In one embodiment, if the system predicts that an employee has a high risk of overtime in the next week, the system will try to adjust the employee's schedule under the guidance of the PSO algorithm to reduce the probability of overtime. During the adjustment process, the system will comprehensively consider factors such as the employee's technical expertise, recent workload, and personal preferences. If the system finds that placing the employee on a relatively easy shift can both meet the job requirements and reduce his workload, then this option will be selected as one of the optimized schedules.

[0132] Step 4: Generate and adjust the shift schedule.

[0133] During the generation and adjustment of the shift plan, the system will automatically generate a detailed shift plan based on the optimized shift results and present it to the user in the form of a table or graph. For example, the system can generate a duty table for the next month, clearly marking the duty time and location of each employee. At the same time, the system provides a visual interface through which users can intuitively understand the shift situation of each employee and make necessary adjustments to the plan. For example, if an employee needs to temporarily change his shift due to family reasons, the manager can adjust it directly on the interface and submit a modification request. The system will re-optimize and adjust the shift plan based on the new request to ensure that the shift plan meets the latest requirements. In addition, the system also supports functions such as historical data comparison and trend analysis to help users better understand the changes and trends in the shift situation. For example, by comparing the shift records of different months, managers can find that the number of overtime hours of employees in certain time periods has increased significantly, so as to prepare for the allocation of human resources in advance.

[0134] Step 5: User interaction and feedback.

[0135] Users can view the shift schedule through a visual interface and edit and adjust it. The system collects user feedback information to further optimize the scheduling model and process. If an employee finds that he is scheduled to work in an unsuitable time period, he can submit an adjustment request through the interface. The system will record user feedback information and use it to further optimize the scheduling model and process.

[0136] The beneficial effects of this program are mainly reflected in the following aspects:

[0137] (1) Integration of deep learning and big data analysis technology: Deep learning technology (especially long short-term memory network LSTM) is closely combined with big data analysis technology and used in the automatic scheduling system. The LSTM model can capture the time dependency in the employee's work mode, helping the system to identify the employee's work habits, preferences and ability change trends, so as to make more reasonable arrangements when scheduling. In this way, the system can not only conduct in-depth mining of historical scheduling data, but also accurately predict the employee's behavior patterns, providing reliable data support for scheduling decisions.

[0138] (2) Application of multi-objective optimization algorithms: Multi-objective optimization algorithms, especially particle swarm optimization (PSO), are introduced to optimize shift scheduling. This algorithm can find a balance between multiple objectives, such as employee workload balance, skill matching, personal preferences, and enterprise operational needs, ensuring the rationality and scientific nature of the shift scheduling.

[0139] (3) Flexible adjustment mechanism: The solution has designed a flexible adjustment mechanism that allows users to adjust the shift schedule according to actual conditions after it is generated, and the system can quickly respond to these adjustments and re-optimize the shift schedule to make it more in line with user expectations. This mechanism ensures that even in the face of emergencies (such as employees taking temporary leave or urgent tasks), the system can respond quickly and adjust the schedule to ensure that business operations are not affected.

[0140] The method in this embodiment not only solves the problems of insufficient data processing capacity, low prediction model accuracy, and single optimization strategy in the existing system, but also creates greater value and benefits for the enterprise. Through the intelligent analysis of historical data and the application of multi-objective optimization algorithms, this solution not only improves the scheduling efficiency and management level, but also enhances the flexibility and adaptability of the system, bringing substantial competitive advantages to the enterprise.

[0141] Based on the same inventive concept, a scheduling data processing device is also provided in the embodiments of this specification, as described in the following embodiments. Since the principle of solving the problem by the scheduling data processing device is similar to that of the scheduling data processing method, the implementation of the scheduling data processing device can refer to the implementation of the scheduling data processing method, and the repeated parts will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived. Figure 5 is a structural block diagram of a scheduling data processing device according to an embodiment of this specification, such as Figure 5 As shown, it includes: a first determination module 501, a second determination module 502, a generation module 503 and an optimization module 504. The structure is described below.

[0142] The first determination module 501 is used to obtain the data of factors affecting business volume and business volume data in the historical time period and the data of factors affecting business volume in the future target time period; and determine the target business volume data in the future target time period based on the data of factors affecting business volume and business volume data in the historical time period and the data of factors affecting business volume in the future target time period.

[0143] The second determination module 502 is used to obtain the scheduling data, employee work record data and attendance factor data within the historical time period and the attendance factor data within the future target time period; based on the scheduling data, employee work record data and attendance factor data within the historical time period and the attendance factor data within the future target time period, determine the availability data of each employee among multiple employees within the future target time period.

[0144] The generating module 503 is used to generate a preliminary shift scheduling plan for the future target time period based on the target business volume data in the future target time period and the availability data of each of the multiple employees.

[0145] The optimization module 504 is used to optimize the preliminary scheduling plan using a multi-objective optimization algorithm to obtain a target scheduling plan for the future target time period; the objective function of the multi-objective optimization algorithm includes multiple target items, and the multiple target items include at least two of the following: a workload matching target item, a workload balance target item, a skill matching target item, and a personal preference and operational demand target item.

[0146] In some embodiments of the present specification, the data of factors affecting business volume include at least one of the following: weather data, business activity data, and business indicator data; the data of factors affecting attendance include at least one of the following: weather data and employee personal information.

[0147] In some embodiments of the present specification, the first determination module is specifically used to: pre-process the data on factors affecting business volume and business volume data within the historical time period to obtain a training sample set; train a preset prediction model based on the training sample set to obtain a target business volume prediction model; input the data on factors affecting business volume within the future target time period into the target business volume prediction model to obtain the target business volume data within the future target time period.

[0148] In some embodiments of the present specification, the second determination module is specifically used to: pre-process the scheduling data, employee work record data and data on factors affecting attendance in the historical time period to obtain a training sample set; use the training sample set to train the long-short-term memory model to obtain a target long-short-term memory model; input the data on factors affecting attendance in the future target time period into the target long-short-term memory model to obtain availability data of each of the multiple employees in the future target time period.

[0149] In some embodiments of the present specification, the generation module is specifically used to: determine at least one scheduling time period and the business volume data of each scheduling time period within the at least one scheduling time period according to the target business volume data within the future target time period; determine the number of employees required for each scheduling time period according to the business volume data of each scheduling time period within the at least one scheduling time period; generate a preliminary scheduling plan for the future target time period based on the availability data of each of the multiple employees in each scheduling time period in the at least one scheduling time period and the number of employees required for each scheduling time period.

[0150] In some embodiments of the present specification, the optimization module is specifically used to optimize the preliminary scheduling plan using a particle swarm algorithm to obtain a target scheduling plan for a future target time period.

[0151] From the above description, it can be seen that the embodiments of this specification achieve the following technical effects: according to the data of factors affecting business volume and business volume in the historical time period and the data of factors affecting business volume in the future target time period, the target business volume data in the future target time period is determined; according to the scheduling data, employee work record data and data of factors affecting attendance in the historical time period and data of factors affecting attendance in the future target time period, the availability data of each employee among multiple employees in the future target time period is determined; based on the target business volume data in the future target time period and the availability data of each employee among multiple employees, a preliminary scheduling plan for the future target time period is generated, and the changes in business volume and the availability data of employees are considered at the same time, ensuring that there are enough people during high workload periods and reducing overtime during low workload periods to achieve the best balance between efficiency and cost, and employees with low leave probability and willingness to work overtime can also be given priority for scheduling. Afterwards, the preliminary scheduling plan is optimized using a multi-objective optimization algorithm to obtain a target scheduling plan for the future target time period. When looking for the optimal scheduling plan, we can comprehensively consider multiple aspects such as employee workload balance, skill matching, personal preferences, and corporate operational needs, so as to improve the rationality and scientific nature of the scheduling.

[0152] This specification also provides a computer device, which can be found in Figure 6The computer device structure diagram of the scheduling data processing method provided by the embodiment of this specification is shown, and the computer device may specifically include an input device 61, a processor 62, and a memory 63. Among them, the memory 63 is used to store processor executable instructions. When the processor 62 executes the instructions, the steps of the scheduling data processing method described in any of the above embodiments are implemented.

[0153] In this embodiment, the input device may specifically be one of the main devices for information exchange between the user and the computer system. The input device may include a keyboard, a mouse, a camera, a scanner, a light pen, a handwriting input board, a voice input device, etc.; the input device is used to input the original data and the program for processing these numbers into the computer. The input device can also obtain and receive data transmitted from other modules, units, and devices. The processor can be implemented in any appropriate manner. For example, the processor can take the form of a computer-readable medium, a logic gate, a switch, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a programmable logic controller, and an embedded microcontroller, etc., such as a microprocessor or a processor and a computer-readable program code (such as software or firmware) that can be executed by the (micro) processor. The memory may specifically be a memory device used to store information in modern information technology. The memory may include multiple levels. In a digital system, anything that can store binary data can be a memory; in an integrated circuit, a circuit with a storage function without a physical form is also called a memory, such as a RAM, a FIFO, etc.; in a system, a storage device with a physical form is also called a memory, such as a memory stick, a TF card, etc.

[0154] In this embodiment, the functions and effects specifically realized by the computer device can be explained in comparison with other embodiments and will not be described in detail here.

[0155] A computer storage medium based on the scheduling data processing method is also provided in the implementation mode of this specification. The computer storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the steps of the scheduling data processing method described in any of the above embodiments are implemented.

[0156] In this embodiment, the storage medium includes, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a cache, a hard disk (HDD) or a memory card. The memory may be used to store computer program instructions. The network communication unit may be an interface for network connection communication set in accordance with the standard specified by the communication protocol.

[0157] In this embodiment, the functions and effects specifically implemented by the program instructions stored in the computer storage medium can be explained in comparison with other embodiments and will not be described in detail here.

[0158] The embodiments of this specification also provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the scheduling data processing method described in any of the above embodiments.

[0159] Obviously, those skilled in the art should understand that the modules or steps of the above-mentioned embodiments of this specification can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. In this way, the embodiments of this specification are not limited to any specific combination of hardware and software.

[0160] It should be understood that the above description is for illustration and not for limitation. Many embodiments and many applications beyond the examples provided will be apparent to those skilled in the art upon reading the above description. Therefore, the scope of this specification should not be determined with reference to the above description, but should be determined with reference to the preceding claims and the full scope of equivalents to which these claims belong.

[0161] The above description is only the preferred embodiment of this specification and is not intended to limit this specification. For those skilled in the art, the embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification shall be included in the protection scope of this specification.

Claims

1. A method for processing shift scheduling data, characterized in that: include: Obtain data on factors affecting business volume and business volume data in a historical time period and data on factors affecting business volume in a future target time period; Determine the target traffic volume data in the future target time period according to the traffic volume influencing factor data and traffic volume data in the historical time period and the traffic volume influencing factor data in the future target time period; Obtaining the scheduling data, employee work record data, and data on factors affecting attendance in a historical time period, as well as the data on factors affecting attendance in a future target time period; determining availability data of each of a plurality of employees in a future target time period based on the scheduling data, employee work record data, and data on factors affecting attendance in a historical time period, as well as the data on factors affecting attendance in a future target time period; generating a preliminary shift scheduling plan for the future target time period based on the target business volume data in the future target time period and the availability data of each of the multiple employees; A multi-objective optimization algorithm is used to optimize the preliminary scheduling plan to obtain a target scheduling plan for a future target time period; the objective function of the multi-objective optimization algorithm includes multiple target items, and the multiple target items include at least two of the following: a workload matching target item, a workload balance target item, a skill matching target item, and a personal preference and operational demand target item.

2. The method for processing shift scheduling data according to claim 1, characterized in that: The data of factors affecting business volume includes at least one of the following: weather data, business activity data and business indicator data; The data of factors affecting attendance include at least one of the following: weather data and employee personal information.

3. The method for processing shift scheduling data according to claim 1, characterized in that: Determining target traffic volume data in the future target time period according to the traffic volume influencing factor data and traffic volume data in the historical time period and the traffic volume influencing factor data in the future target time period includes: Preprocessing the data of factors influencing the traffic volume and the traffic volume data in the historical time period to obtain a training sample set; Training a preset prediction model based on the training sample set to obtain a target business volume prediction model; The data of factors influencing the business volume in the future target time period are input into the target business volume prediction model to obtain the target business volume data in the future target time period.

4. The method for processing shift scheduling data according to claim 1, characterized in that: According to the shift scheduling data, employee work record data and attendance factor data in the historical time period and attendance factor data in the future target time period, availability data of each of the multiple employees in the future target time period is determined, including: Preprocessing the shift scheduling data, employee work record data, and attendance factor data within the historical time period to obtain a training sample set; Using the training sample set to train the long short-term memory model to obtain a target long short-term memory model; The data of factors affecting attendance in the future target time period are input into the target long short-term memory model to obtain the availability data of each employee among the multiple employees in the future target time period.

5. The method for processing shift scheduling data according to claim 1, characterized in that: Based on the target business volume data in the future target time period and the availability data of each of the multiple employees, a preliminary shift scheduling plan for the future target time period is generated, including: Determine at least one scheduling time period and the business volume data of each scheduling time period in the at least one scheduling time period according to the target business volume data in the future target time period; Determining the number of employees required for each scheduling time period according to the business volume data of each scheduling time period within the at least one scheduling time period; A preliminary scheduling plan for a future target time period is generated based on the availability data of each employee among the multiple employees in each scheduling time period in the at least one scheduling time period and the number of employees required for each scheduling time period.

6. The method for processing shift scheduling data according to claim 1, characterized in that: The preliminary shift scheduling scheme is optimized by using a multi-objective optimization algorithm to obtain a target shift scheduling scheme for the future target time period, including: The preliminary scheduling plan is optimized using a particle swarm algorithm to obtain a target scheduling plan for a future target time period.

7. A shift scheduling data processing device, characterized in that: include: The first determination module is used to obtain data on factors affecting business volume and business volume data in a historical time period and data on factors affecting business volume in a future target time period; Determine the target traffic volume data in the future target time period according to the traffic volume influencing factor data and traffic volume data in the historical time period and the traffic volume influencing factor data in the future target time period; The second determination module is used to obtain the scheduling data, employee work record data and attendance factor data in the historical time period and the attendance factor data in the future target time period; determine the availability data of each employee among the multiple employees in the future target time period according to the scheduling data, employee work record data and attendance factor data in the historical time period and the attendance factor data in the future target time period; A generating module, configured to generate a preliminary shift scheduling plan for a future target time period based on the target business volume data in the future target time period and the availability data of each of the multiple employees; An optimization module is used to optimize the preliminary scheduling plan using a multi-objective optimization algorithm to obtain a target scheduling plan for a future target time period; the objective function of the multi-objective optimization algorithm includes multiple target items, and the multiple target items include at least two of the following: a workload matching target item, a workload balance target item, a skill matching target item, and a personal preference and operation demand target item.

8. A computer device, characterized in that: The method comprises a processor and a memory for storing processor-executable instructions, wherein the processor implements the steps of the method according to any one of claims 1 to 6 when executing the instructions.

9. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images