Automatic personnel scheduling method and device, computer equipment and storage medium

Abstract

The invention discloses an automatic personnel scheduling method and device, computer equipment and a storage medium. The method comprises the following steps: determining a target feature through a target feature factor and a data label; training a corresponding preset case quantity prediction model through the target features to obtain a successfully trained preset case quantity prediction model; obtaining to-be-predicted personnel time data, and inputting the to-be-predicted personnel time data into the successfully trained preset case quantity prediction model to obtain a case prediction quantity in a preset time period; and obtaining the daily average processing case quantity, and determining the required scheduling personnel according to the daily average processing case quantity and the case prediction quantity of the preset time period. According to the scheme, the original manual scheduling mode can be replaced by the automatic scheduling system, the scheduling plan of the scheduling personnel can be automatically adjusted based on the case prediction amount of the preset time period predicted in real time, and the personnel configuration is ensured to be matched with the case amount demand, so that the optimal configuration of human resources is realized.

Description

Technical Field

[0001] The present invention relates to the fields of intelligent decision-making technology and financial technology, and in particular to a method, device, computer equipment and storage medium for automatic personnel scheduling. Background Art

[0002] In the insurance and finance industries, the scheduling of remote workers is currently mainly based on manual experience and traditional working methods. However, the current scheduling method has not yet been combined with the automatic scheduling system, which has led to a series of problems as follows: The current manual scheduling lacks flexibility and cannot be adjusted intelligently according to the real-time fluctuations in the caseload. This limitation significantly affects the effectiveness of remote human resources. When the caseload surges, there may be a shortage of staff, and when the caseload decreases, there may be an oversupply of staff, resulting in a waste of human resources; More specifically, due to the lack of accurate predictions and flexible adjustment mechanisms, there are problems of over-allocation or under-allocation of remote claims personnel, which may lead to unnecessary costs or reduced service quality. During emergencies or unpredictable case peaks, the manual scheduling system may not be able to quickly adjust resources and staffing, resulting in longer response times and affecting customer experience.

[0003] Therefore, those skilled in the art are in urgent need of finding a new technical solution to solve the above problems. Summary of the invention

[0004] Based on this, a method, device, computer equipment and storage medium for automatic personnel scheduling are provided to solve the technical problems in the prior art that lack flexibility and cannot be intelligently adjusted according to real-time fluctuations in case volume.

[0005] A method for automatically scheduling personnel, comprising: Determine the target feature through the target feature factor and the data label; The preset case volume prediction model corresponding to the target feature is trained to obtain a successfully trained preset case volume prediction model; Obtaining personnel time data to be predicted, and inputting the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume for the preset time period; The average number of cases handled per day is obtained, and the required scheduling personnel are determined based on the average number of cases handled per day and the predicted number of cases in the preset time period.

[0006] A personnel automatic scheduling device, comprising: A feature processing module is used to determine target features through target feature factors and data labels; A training module, used for training the corresponding preset case volume prediction model through the target feature to obtain a successfully trained preset case volume prediction model; A case volume prediction module is used to obtain the personnel time data to be predicted, and input the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume for a preset time period; The intelligent scheduling module is used to obtain the average number of cases handled per day, and determine the required scheduling personnel based on the average number of cases handled per day and the predicted number of cases in the preset time period.

[0007] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned automatic personnel scheduling method when executing the computer program.

[0008] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned automatic personnel scheduling method is implemented.

[0009] The above-mentioned personnel automatic scheduling method, device, computer equipment and storage medium determine the target feature through the target feature factor and data label; train the corresponding preset case volume prediction model through the target feature to obtain the successfully trained preset case volume prediction model; obtain the personnel time data to be predicted, and input the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the case prediction volume of the preset time period; obtain the average daily case volume, and determine the required scheduling personnel through the average daily case volume and the case prediction volume of the preset time period. In this way, the automatic scheduling system can replace the original manual scheduling method, and can automatically adjust the scheduling personnel's scheduling plan based on the real-time predicted case prediction volume of the preset time period to ensure that the personnel allocation matches the case volume demand, thereby achieving the optimal configuration of human resources. Specifically, the automated scheduling plan can greatly improve the flexibility of scheduling, enhance cost-effectiveness and significantly improve the emergency response speed, and at the same time can realize a rapid response to case volume fluctuations, reduce the required costs, and thus improve the overall service efficiency of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0011] Figure 1This is a schematic diagram of an application environment of a method for automatic staff scheduling in one embodiment of the present invention; Figure 2 is a process diagram of a method for automatic staff scheduling in one embodiment of the present invention; Figure 3 2 is a schematic diagram of the structure of an automatic personnel scheduling device in one embodiment of the present invention; Figure 4 is a schematic diagram of a computer device in one embodiment of the present invention; Figure 5 It is a structural diagram of an automatic scheduling system in one embodiment of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] The automatic staff scheduling method provided by the present invention can be applied in Figure 1 In an application environment, a client communicates with a server via a network. The client may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server may be implemented as an independent server or a server cluster consisting of multiple servers.

[0014] In one embodiment, if Figure 2 As shown, a method for automatic staff scheduling is provided, and the method is applied in Figure 1 The following steps are used to illustrate the automatic shift scheduling system where the server is located: S10, determining the target feature through the target feature factor and the data label.

[0015] Understandably, the target characteristic factors are key factors screened out from the data that affect the caseload. For example, the data that affects the caseload is data information that may affect the number of cases collected from surveys of remote claims adjusters. In addition, the screened out key factors can be obtained through manual screening and / or data preprocessing. In the claims scenario, the key factors can be whether it is a weekend (weekend, weekday), time period (morning: 8:00-10:00, noon: 11:00-14:00, afternoon: 15:00-18:00, evening: 19:00-21:00) and institution level (secondary institution, tertiary institution and Four-level institutions); data labels are based on the status information marked on the case volume information. For example, based on the case volume, in the case volume information in recent years, the number of cases in each hour period from 8:00 to 21:00 is labeled as small, less, more or multiple. The target feature is the feature imported into the preset case volume prediction model. The target feature factor and the data label are used together as features such as [whether it is a weekend feature, time period feature, institution level feature, label], and the feature is input into the preset XGBoost model for training to extract the key feature affecting the case volume - the target feature; The determination of the target features of this embodiment is performed in the feature processing module of the system corresponding to this method. The feature processing module can predetermine the feature data required for model training to improve the availability of data in the prediction model.

[0016] S20, training the corresponding preset case volume prediction model through the target feature to obtain a successfully trained preset case volume prediction model.

[0017] Understandably, the structure of the preset case volume prediction model is the Autoregressive Integrated Moving Average (ARIMA) model, where the full name of the ARIMA model is the Autoregressive Integrated Moving Average Model, which is one of the time series forecasting and analysis methods. Its basic principle is to establish a forecast of future values ​​by integrating different components in the time series (such as trends, seasonality, and residuals). It comprehensively considers multiple components of the time series to achieve accurate predictions. During the training process, the model parameters will be continuously adjusted until the model parameters that meet the actual application of the model are obtained. In addition, the successfully trained preset caseload prediction model can be determined according to the number of target characteristic factors, such as training three preset caseload prediction models for handling weekends, time periods, and agency levels, where all three target characteristic factors affect the caseload, and the differences between different agencies are particularly prominent; In this embodiment, the selection of the preset case volume prediction model has a strict mathematical derivation and theoretical basis to predict the target features through its model structure.

[0018] S30, obtaining the personnel time data to be predicted, and inputting the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume for the preset time period.

[0019] It can be understood that the personnel time data to be predicted refers to the data needed to predict the scheduling situation in actual application, which can be the specific working time, such as weekends / weekdays; after the personnel time data to be predicted is input into the successfully trained preset case volume prediction model, the output is the predicted case volume for the preset time period determined by the model based on the personnel, such as the predicted number of final cases is F, F can specifically represent the predicted case volume per hour output by the case prediction module; the predicted case volume for the preset time period refers to the number of cases that each scheduled personnel can handle within a period of time, such as the predicted case volume per hour mentioned above; This embodiment uses the successfully trained model to directly obtain accurate prediction data - the predicted number of cases in a preset time period.

[0020] S40, obtaining the average number of cases processed per day, and determining the required scheduling personnel based on the average number of cases processed per day and the predicted number of cases in the preset time period.

[0021] Understandably, the average number of cases handled per day refers to the average number of cases handled per shift staff member per day; there is a mathematical relationship between the average number of cases handled per day and the predicted number of cases in a given time period; This embodiment determines the required staff for scheduling within a certain time period through a mathematical operation relationship between two data.

[0022] It should be noted that if Figure 5 As shown, the automatic scheduling system corresponding to the present method is mainly composed of a feature processing module, a caseload prediction module and an intelligent scheduling module. Each module can execute the corresponding processing process to finally obtain the required scheduling personnel.

[0023] In the embodiment of step S10 to step S40, the original manual scheduling method can be replaced by an automatic scheduling system. The system can automatically adjust the scheduling plan of the scheduling personnel based on the real-time predicted case volume for a preset time period to ensure that the staffing is matched with the case volume demand, thereby achieving the optimal allocation of human resources. Specifically, the automated scheduling plan can greatly improve the flexibility of scheduling, enhance cost-effectiveness and significantly improve the emergency response speed. At the same time, it can achieve a rapid response to fluctuations in case volume, reduce the required costs, and thus improve the overall service efficiency of users.

[0024] Furthermore, before determining the target feature through the target feature factor and the data label, the method further includes: Acquire first training feature data corresponding to the feature factor; Converting the first training feature data into numerical feature data by a preset conversion method; The coefficient between the numerical characteristic data and the target variable is calculated by a preset coefficient calculation method, and a preset number of characteristic factors are selected as the target characteristic factors according to the size relationship between the coefficient and the preset coefficient.

[0025] It can be understood that the characteristic factors refer to the majority of factors corresponding to the data affecting the case volume (without screening); the first training characteristic data is the data used to train and determine the target characteristics, including the data affecting the case volume; the preset conversion methods include the z-score method and the one-hot encoding method. Encoding) method, converting data into standardized numerical feature data by z-score method (for each numerical feature, calculating its mean and standard deviation in the data set, using z-score standardization method is to convert data into a distribution with mean 0 and standard deviation 1) and using one-hot encoding to convert non-numeric features into numerical feature data (for each non-numeric feature, create a new one-hot encoding matrix. If a feature has three categories A, B, and C, it is converted into three new binary features, representing A, B, and C respectively); the preset coefficient calculation method is the Pearson correlation coefficient calculation method, where the Pearson correlation coefficient formula calculates the sum of the products of the differences between two variables and their means, and then divides it by the product of their respective standard deviations to obtain the correlation coefficient. After obtaining the coefficient, the first training feature data is sorted according to the absolute value of the coefficient, and the top three features with the strongest correlation with the target variable are selected; This embodiment reduces the impact of noise data and redundant data on model training by selecting the features most relevant to the target variable, thereby improving the accuracy and generalization ability of the model; selecting the features most relevant to the target variable can make the model easier to understand and explain, which is beneficial to business prediction and model tuning.

[0026] Furthermore, before determining the target feature through the target feature factor and the data label, the method further includes: Obtaining the quantile corresponding to the first training feature data; Determining a data threshold using the quantile; After the first training feature data is classified by the data threshold, a data label corresponding to each category of the first training feature data is marked.

[0027] Understandably, the quantile is the quartile. The quartile refers to the values ​​at three dividing points in statistics, which are arranged from small to large and divided into four equal parts. The three values ​​are, for example, 25%, 50%, and 75%. The quantile will be used as the basis for dividing data labels; the data threshold is defined according to the quartile for each data label, including: less: data with a case volume lower than the first quartile (25%), less: data with a case volume between the first quartile and the second quartile (50%), more: data with a case volume between the second quartile and the third quartile (75%), and more: data with a case volume higher than the third quartile; the data labels are the above-mentioned less, less, more, and more. The various first training feature data are classified according to the above-mentioned data thresholds. After successful classification, the corresponding data labels are marked; This embodiment determines the data threshold by calculating the quantile of the first training feature data. This method is data-driven and can automatically adapt to the distribution characteristics of the data. Using the quantile as the threshold for classification can make a more detailed division based on the statistical characteristics of the data. By setting multiple data thresholds corresponding to the quantiles, the first training feature data can be divided into multiple categories, which helps to expand the prediction range of the model.

[0028] Further, the preset case volume prediction model corresponding to the target feature is trained to obtain a successfully trained preset case volume prediction model, including: Acquire pre-classified second training feature data; After inputting the second training feature data into the preset case volume prediction model, determining the first model parameter by a preset model parameter adjustment method; Under the preset case volume prediction model of the first model parameters, the initial case prediction volume output by the model is adjusted by different adjustment parameters to obtain the successfully trained preset case volume prediction model.

[0029] Understandably, the second training feature data is the institutional data in the historical case, and the historical case data can be divided into three categories according to the institutional level: secondary, tertiary and quaternary institutional data, and each data record can be formatted as [specific time (weekday / weekend, time period), case volume], for example, [2024-01-01 08:00, 22]; the preset case volume prediction model refers to the regression integral moving average (ARIMA) model; the first model parameters are the best model parameters p, d, q determined by the grid search algorithm, where p represents the order of autoregression (AR), that is, the number of autoregression terms in the model, d represents the number of differencing, that is, the number of difference terms in the model, and q represents the order of moving average (MA), that is, the number of moving average terms in the model; the adjustment parameter is a hyperparameter, and the initial case prediction volume output by the preset case volume prediction model can be adjusted by the adjusted hyperparameter, such as multiplying the hyperparameter by the initial case prediction volume to obtain a case prediction volume for a preset time period of a final output result; In this embodiment, by acquiring the pre-classified second training feature data, the model can learn more precise classification features and rules, thereby improving the accuracy and reliability of the prediction; using the classified data for training can greatly shorten the model training time and improve the training efficiency. At the same time, since the adjustment of the model parameters is based on scientific methods, it also reduces unnecessary trial and error costs.

[0030] Furthermore, the adjustment of the initial case prediction quantity output by the model through different adjustment parameters includes: Obtain multiple adjustment parameters that are successfully adjusted; The case prediction quantity for a preset time period output by the successfully trained preset case quantity prediction model is determined by the adjustment parameters and the initial case prediction quantity.

[0031] Understandably, the adjustment parameters will have different values ​​corresponding to different time characteristics of case occurrence (weekdays / weekends, time periods). Different adjustment parameters can be used to obtain different case volume forecasts. The specific parameter values ​​and their applications are as follows (when the initial case forecast volume is f): (a) When the feature is [weekday, morning] (for example, the input time is 2024-09-06 09:25), the calculation parameter is 1.18, that is, after the model prediction, it is multiplied by 1.18, and the final predicted number of cases is output: 1.18f.

[0032] (b) When the feature is [weekday, noon], the calculation parameter is 0.6, that is, the final predicted number of cases is 0.6f.

[0033] (c) When the feature is [weekday, afternoon], the calculation parameter is 1.32, that is, the final predicted number of cases is 1.32f.

[0034] (d) When the feature is [weekdays, evening], the calculation parameter is 1.10, that is, the final predicted number of cases is 1.10f.

[0035] (e) When the feature is [weekend, morning], the calculation parameter is 0.79, that is, the predicted number of cases is 0.79f.

[0036] (f) When the feature is [weekend, noon], the calculation parameter is 0.72, that is, the predicted number of cases is 0.72f.

[0037] (g) When the feature is [weekend, afternoon], the calculation parameter is 1.25, that is, the predicted number of cases is 1.25f.

[0038] (h) When the feature is [weekend, night], the calculation parameter is 1.31, that is, the predicted number of cases is 1.31f; This embodiment adjusts the initial case prediction volume originally output by the preset case volume prediction model by adjusting parameters to obtain a case prediction volume suitable for a preset time period under different time scenarios; the eight adjustment parameters mentioned above can be designed (in view of the fact that the basic model only considers the characteristics of the organization and fails to fully consider the impact of holidays and peak and trough periods on the case volume, the introduction of eight adjustment parameters can enhance the accuracy of the prediction) to dynamically correct the case volume prediction value output by the model. This parameter adjustment method not only improves the prediction accuracy, but also enhances the adaptability and generalization ability of the model to different scenarios.

[0039] Furthermore, the obtaining of multiple successfully adjusted adjustment parameters includes: Obtaining initial values ​​of characteristic parameters and pre-set hyperparameters corresponding to the preset case volume prediction model; Obtaining a label value through the initial value of the feature parameter, the hyperparameter, and the initial case prediction quantity output by the preset case quantity prediction model; New hyperparameters are obtained by adjusting the preset model parameter adjustment method between the minimized cumulative difference between the label value and the initial case prediction amount.

[0040] Understandably, the initial value of the characteristic parameter is the parameter required to run the preset case volume prediction model at the beginning. Specifically, the influence range of each parameter can be preliminarily determined through the importance of the features output by the model, that is, the initial value of the characteristic parameter is determined, wherein the value of each feature importance does not exceed 1 and the sum of all is equal to 1; the hyperparameters in the 8 models are pre-set, w1\w2\w3\w4\w5\w6\w7\w8 (the value of each hyperparameter can be between 0-2), and the prediction result f output by the preset case volume prediction model is multiplied by the hyperparameter and the initial value of the characteristic parameter to obtain the label value; the preset model parameter adjustment method-grid search algorithm is used to adjust the value of the hyperparameter to control the value of the initial case prediction amount f. The main process is to minimize the cumulative value of [label value-initial case prediction amount f] of all training data, that is, min(sum[label value-initial case prediction amount f]), that is, to obtain the new hyperparameter; This embodiment can significantly improve the prediction accuracy and generalization ability of the model through reasonable hyperparameter adjustment, which helps the model to better cope with various complex situations in practical applications; effective hyperparameter adjustment methods can reduce unnecessary model training time and save computing resource consumption.

[0041] Furthermore, obtaining the average number of cases processed per day, and determining the required scheduling personnel according to the average number of cases processed per day and the predicted number of cases in the preset time period, includes: Get the personnel's daily service time; Under the preset scheduling calculation formula, the required scheduled personnel are determined by the personnel's daily service time, the average number of cases handled per day, and the predicted number of cases in the preset time period; the preset scheduling calculation formula = the average number of cases handled per day / (predicted number of cases in the preset time period / personnel's daily service time).

[0042] Understandably, the daily service time of personnel refers to the time required for personnel to handle the average number of cases per day, such as the definition of 14 hours per day for providing claims services; the average number of cases handled per day refers to the average number of cases handled by personnel per day, such as the average number of cases handled by remote claims personnel per day, which data can be collected and obtained through pre-survey by the organization; the preset scheduling calculation formula can be used to determine the number of personnel required to be scheduled per hour, or the number of personnel required to be scheduled at other time points, which is specifically determined by the daily service time of personnel. When the daily service time of personnel is hours, the formula output result is the number of personnel required to be scheduled per hour, and the same principle applies to other times; This embodiment performs intelligent scheduling based on the predicted case volume and the average processing capacity of remote claims adjusters. The determined schedulers are closer to the actual application scenarios. At the same time, the responsible personnel of the organization will refer to the number of schedulers required per hour and arrange the schedulers per hour in sequence according to the personnel list, so as to be able to specify the scheduling of a single person.

[0043] In summary, the above provides a method for automatic personnel scheduling, which determines the target characteristics through target characteristic factors and data labels; trains the corresponding preset case volume prediction model through the target characteristics to obtain a successfully trained preset case volume prediction model; obtains the time data of the personnel to be predicted, and inputs the time data of the personnel to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume for a preset time period; obtains the average daily case volume processed, and determines the required scheduled personnel through the average daily case volume processed and the predicted case volume for the preset time period. In this way, the original manual scheduling method can be replaced by an automatic scheduling system. The system can automatically adjust the scheduling plan of the scheduling personnel based on the real-time predicted case volume in the preset time period to ensure that the staffing matches the case volume demand, thereby achieving the optimal allocation of human resources. Specifically, the automated scheduling plan can greatly improve the flexibility of scheduling, enhance cost-effectiveness and significantly improve the emergency response speed. At the same time, it can quickly respond to fluctuations in case volume, reduce the required costs, and thus improve the overall service efficiency of users; in addition, the intelligent scheduling system can automatically adjust the scheduling plan according to the real-time prediction results to ensure that the staffing matches the case volume demand, thereby achieving the optimal utilization of resources; in addition, three target characteristic factors that significantly affect the case volume prediction are identified, which provide key inputs for the construction of the preset case volume prediction model, significantly improving the accuracy and reliability of the prediction; in addition, based on regional characteristics, the healing case volume prediction model can be trained to capture the case volume change characteristics in different regions.

[0044] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0045] In one embodiment, a personnel automatic scheduling device is provided, and the personnel automatic scheduling device corresponds to the personnel automatic scheduling method in the above embodiment. Figure 3 As shown, the automatic personnel scheduling device includes an acquisition module 11, a determination module 12 and an execution module 13. The detailed description of each functional module is as follows: The feature processing module 11 is used to determine the target feature through the target feature factor and the data label; A training module 12, used to train the corresponding preset case volume prediction model through the target feature to obtain a successfully trained preset case volume prediction model; The case volume prediction module 13 is used to obtain the personnel time data to be predicted, and input the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume in the preset time period; The intelligent scheduling module 14 is used to obtain the average number of cases processed per day, and determine the required scheduling personnel based on the average number of cases processed per day and the predicted number of cases in the preset time period.

[0046] Furthermore, the automatic personnel scheduling device also includes: A first acquisition module, used to acquire first training feature data corresponding to the feature factor; A conversion module, used to convert the first training feature data into numerical feature data by a preset conversion method; The screening module is used to screen the coefficients between the numerical feature data and the target variable calculated by a preset coefficient calculation method, and screen out a preset number of feature factors as the target feature factors according to the size relationship between the coefficients and the preset coefficients.

[0047] Furthermore, the automatic personnel scheduling device also includes: A second acquisition module is used to obtain the quantile corresponding to the first training feature data; A determination module, used for determining a data threshold value according to the quantile; The marking module is used to mark the data label corresponding to each category of the first training feature data after classifying the first training feature data according to the data threshold.

[0048] Furthermore, the training module includes: A first acquisition submodule is used to acquire pre-classified second training feature data; A first determination submodule, configured to input the second training feature data into the preset case volume prediction model and determine the first model parameter by a preset model parameter adjustment method; The adjustment submodule is used to adjust the initial case prediction quantity output by the model through different adjustment parameters under the preset case volume prediction model of the first model parameters to obtain the successfully trained preset case volume prediction model.

[0049] Furthermore, the adjustment submodule includes: An acquisition unit, used for acquiring a plurality of adjustment parameters that are successfully adjusted; A determination unit is used to determine the case prediction volume for a preset time period output by the successfully trained preset case volume prediction model through the adjustment parameter and the initial case prediction volume.

[0050] Furthermore, the acquisition unit includes: An acquisition subunit, used to acquire initial values ​​of characteristic parameters and pre-set hyperparameters corresponding to the preset case volume prediction model; An output subunit, used to obtain a label value through the initial value of the feature parameter, the hyperparameter, and the initial case prediction quantity output by the preset case quantity prediction model; The adjustment subunit is used to obtain new hyperparameters after adjusting the minimized cumulative difference between the label value and the initial case prediction amount through the preset model parameter adjustment method.

[0051] Furthermore, the intelligent scheduling module includes: The second acquisition submodule is used to obtain the daily service time of personnel; The second determination submodule is used to determine the required scheduled personnel according to the personnel's daily service time, the average number of cases handled per day, and the predicted number of cases in the preset time period under a preset scheduling calculation formula; the preset scheduling calculation formula = the average number of cases handled per day / (the predicted number of cases in the preset time period / personnel's daily service time).

[0052] For the specific definition of the automatic personnel scheduling device, please refer to the definition of the automatic personnel scheduling method above, which will not be repeated here. Each module in the above-mentioned automatic personnel scheduling device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0053] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data involved in the automatic personnel scheduling method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for automatic personnel scheduling is implemented.

[0054] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for automatic staff scheduling in the above embodiment are implemented, such as Figure 2 Steps S10 to S40 shown include the following steps: Determine the target feature through the target feature factor and the data label; The preset case volume prediction model corresponding to the target feature is trained to obtain a successfully trained preset case volume prediction model; Obtaining personnel time data to be predicted, and inputting the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume for the preset time period; The average number of cases handled per day is obtained, and the required scheduling personnel are determined based on the average number of cases handled per day and the predicted number of cases in the preset time period.

[0055] Alternatively, when the processor executes the computer program, the functions of each module / unit of the automatic personnel scheduling device in the above embodiment are realized, for example Figure 3 The functions of modules 11 to 14 are shown in Figure 1. To avoid repetition, they will not be described here.

[0056] In one embodiment, a computer-readable storage medium is provided on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for automatically scheduling personnel in the above embodiment are implemented, such as Figure 2 Steps S10 to S40 shown include the following steps: Determine the target feature through the target feature factor and the data label; The preset case volume prediction model corresponding to the target feature is trained to obtain a successfully trained preset case volume prediction model; Obtaining personnel time data to be predicted, and inputting the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume for the preset time period; The average number of cases handled per day is obtained, and the required scheduling personnel are determined based on the average number of cases handled per day and the predicted number of cases in the preset time period.

[0057] Alternatively, when the computer program is executed by the processor, the functions of each module / unit of the automatic personnel scheduling device in the above embodiment are realized, such as Figure 3 The functions of modules 11 to 14 are shown in Figure 1. To avoid repetition, they will not be described here.

[0058] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0059] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0060] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for automatic staff scheduling, characterized in that: include: Determine the target feature through the target feature factor and the data label; The preset case volume prediction model corresponding to the target feature is trained to obtain a successfully trained preset case volume prediction model; Obtaining personnel time data to be predicted, and inputting the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume for the preset time period; The average number of cases handled per day is obtained, and the required scheduling personnel are determined based on the average number of cases handled per day and the predicted number of cases in the preset time period.

2. The automatic staff scheduling method according to claim 1, characterized in that: Before determining the target feature by using the target feature factor and the data label, the method further includes: Acquire first training feature data corresponding to the feature factor; Converting the first training feature data into numerical feature data by a preset conversion method; The coefficient between the numerical characteristic data and the target variable is calculated by a preset coefficient calculation method, and a preset number of characteristic factors are selected as the target characteristic factors according to the size relationship between the coefficient and the preset coefficient.

3. The automatic staff scheduling method according to claim 1 or 2, characterized in that: Before determining the target feature by using the target feature factor and the data label, the method further includes: Obtaining the quantile corresponding to the first training feature data; Determining a data threshold using the quantile; After the first training feature data is classified by the data threshold, a data label corresponding to each category of the first training feature data is marked.

4. The automatic staff scheduling method according to claim 1, characterized in that: The method of training the corresponding preset case volume prediction model through the target feature to obtain a successfully trained preset case volume prediction model includes: Acquire pre-classified second training feature data; After inputting the second training feature data into the preset case volume prediction model, determining the first model parameter by a preset model parameter adjustment method; Under the preset case volume prediction model of the first model parameters, the initial case prediction volume output by the model is adjusted by different adjustment parameters to obtain the successfully trained preset case volume prediction model.

5. The automatic staff scheduling method according to claim 4, characterized in that: The initial case prediction quantity output by the model is adjusted by different adjustment parameters, including: Obtain multiple adjustment parameters that are successfully adjusted; The case prediction quantity for a preset time period output by the successfully trained preset case quantity prediction model is determined by the adjustment parameters and the initial case prediction quantity.

6. The automatic staff scheduling method according to claim 5, characterized in that: The multiple adjustment parameters successfully adjusted are obtained, including: Obtaining initial values ​​of characteristic parameters and pre-set hyperparameters corresponding to the preset case volume prediction model; Obtaining a label value through the initial value of the feature parameter, the hyperparameter, and the initial case prediction quantity output by the preset case quantity prediction model; New hyperparameters are obtained by adjusting the preset model parameter adjustment method between the minimized cumulative difference between the label value and the initial case prediction amount.

7. The automatic staff scheduling method according to claim 1, characterized in that: The obtaining of the average number of cases processed per day, and determining the required scheduling personnel according to the average number of cases processed per day and the predicted number of cases in the preset time period, includes: Get the personnel's daily service time; Under the preset scheduling calculation formula, the required scheduled personnel are determined by the personnel's daily service time, the average number of cases handled per day, and the predicted number of cases in the preset time period; the preset scheduling calculation formula = the average number of cases handled per day / (predicted number of cases in the preset time period / personnel's daily service time).

8. An automatic personnel scheduling device, characterized in that: include: A feature processing module is used to determine target features through target feature factors and data labels; A training module, used for training the corresponding preset case volume prediction model through the target feature to obtain a successfully trained preset case volume prediction model; A case volume prediction module is used to obtain the personnel time data to be predicted, and input the personnel time data to be predicted into the successfully trained preset case volume prediction model to obtain the predicted case volume for a preset time period; The intelligent scheduling module is used to obtain the average number of cases handled per day, and determine the required scheduling personnel based on the average number of cases handled per day and the predicted number of cases in the preset time period.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the automatic personnel scheduling method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the automatic personnel scheduling method according to any one of claims 1 to 7 is implemented.

Images