Personnel scheduling management method and system based on PPMH index

Abstract

The invention relates to the technical field of personnel scheduling management, in particular to a personnel scheduling management method and system based on a PPMH index. The method comprises the following steps: collecting historical business data and carrying out business density estimation, generating business density data and complete man-hour data, carrying out business demand estimation based on the business density data to obtain an estimated product quantity, generating an efficiency demand standard, carrying out man-hour simulation inference according to the efficiency demand standard, and carrying out man-hour simulation inference according to the efficiency demand standard. The method comprises the steps of obtaining pre-estimated required basic working hours, generating pre-estimated PPMH data, performing personnel configuration mapping based on the pre-estimated PPMH data to obtain an initial scheduling scheme, performing business simulation to generate simulated business data, reconstructing the initial scheduling scheme according to the simulated business data to obtain an alternate scheduling sequence, and sending the alternate scheduling sequence to a terminal for personnel management. According to the invention, a more flexible and more accurate personnel scheduling management method is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of personnel scheduling management, and in particular, to a personnel scheduling management method and system based on the PPMH index. Background Art

[0002] In current enterprise operations, the importance of personnel scheduling management has become increasingly prominent. Reasonable scheduling can significantly improve work efficiency and the rationality of resource allocation. However, traditional scheduling methods often rely on experience and intuition, making it difficult to adapt to rapidly changing demands, resulting in resource waste and uneven work burdens on employees. Scheduling methods lacking scientific basis have significantly affected the overall operation effect. With the progress of data analysis technology, enterprises have begun to attempt to use historical data for scheduling decisions. However, in practical applications, there are still problems in the effective integration and analysis of data. The accuracy of business density estimation and demand prediction is affected by various factors, resulting in the personnel allocation plan being unable to reflect the real demand, thereby affecting the operation efficiency of the enterprise. Traditional scheduling management methods cannot cope with complex and dynamically changing working environments, and there is an urgent need for new solutions to improve the management level. Summary of the Invention

[0003] Based on this, it is necessary to provide a personnel scheduling management method and system based on the PPMH index to solve at least one of the above technical problems.

[0004] To achieve the above object, a personnel scheduling management method based on the PPMH index includes the following steps:

[0005] Step S1: Collect historical business data; estimate the business density based on the historical business data to obtain business density data; perform working hour mapping on the business density data to generate complete working hour data;

[0006] Step S2: Estimate the business demand based on the business density data to obtain the estimated product quantity; calibrate the demand for the complete working hour data according to the estimated product quantity to generate the efficiency demand standard;

[0007] Step S3: Perform working hour simulation inference on the estimated product quantity according to the efficiency demand standard to obtain the estimated basic working hours required; perform work efficiency measurement processing on the estimated product quantity and the estimated basic working hours required to generate the estimated PPMH data;

[0008] Step S4: Perform personnel allocation mapping on the estimated product quantity based on the estimated PPMH data to obtain the initial scheduling plan; perform business simulation according to the initial scheduling plan to generate simulated business data;

[0009] Step S5: Reconstruct the rotation scheduling order of the initial scheduling plan according to the simulated business data to obtain the rotation scheduling order, and send the rotation scheduling order to the terminal to perform personnel scheduling management.

[0010] By collecting historical business data, the present invention can establish a comprehensive business foundation. The business density estimation provides an in-depth understanding of demand fluctuations. The generated business density data provides a scientific basis for subsequent man-hour mapping and demand prediction. The generation of complete man-hour data ensures an accurate grasp of resource utilization. The business demand prediction can quantify future product demands. The formulation of the efficiency demand standard provides a clear reference for personnel allocation. The man-hour simulation inference realizes a scientific estimation of the basic man-hour. The generated predicted PPMH data (PPMH (Product Per Man Hour) index definition: the number of products created per person per hour) provides a quantitative index for scheduling decisions. The personnel allocation mapping ensures the rationality of the initial scheduling plan. The business simulation verifies the feasibility of the scheduling plan. The reconstruction of the rotation scheduling order improves the flexibility and adaptability of scheduling. The final implementation of scheduling management realizes the efficient allocation of human resources. The overall process improves the scientific nature and accuracy of management, providing more efficient operation support for enterprises.

[0011] Preferably, step S1 includes the following steps:

[0012] Step S11: Collect historical business data; perform time series decomposition on the historical business data to obtain a product time series;

[0013] Step S12: Perform category merging on the product time series to obtain category distribution data; perform peak detection on the category distribution data to generate peak period feature data;

[0014] Step S13: Perform business density estimation on the peak period feature data to obtain business density data;

[0015] Step S14: Perform man-hour mapping on the business density data to generate basic man-hour data; perform auxiliary task calculation on the basic man-hour data to obtain auxiliary man-hour data;

[0016] Step S15: Integrate the basic man-hour data and the auxiliary man-hour data to generate complete man-hour data.

[0017] The present invention provides a dynamic change perspective of product demand through the collection and time series decomposition of historical business data. The generation of category merging and category distribution data lays a foundation for subsequent analysis. The identification of peak period characteristic data enables enterprises to accurately grasp business peaks. The implementation of business density estimation provides an in-depth understanding of resource utilization rate. The generation of basic working hour data ensures the quantitative analysis of major business demands. The auxiliary task calculation adds flexibility and comprehensiveness to the overall working hour management. The realization of working hour integration generates complete working hour data, providing a scientific basis for personnel scheduling. The overall process improves the efficiency and accuracy of enterprises in personnel allocation and resource management, and supports the ability to quickly respond to market changes.

[0018] Preferably, step S2 includes the following steps:

[0019] Step S21: Obtain the current business hour parameter; perform trend extrapolation on the business density data to obtain business trend prediction data;

[0020] Step S22: Perform periodic adjustment on the business trend prediction data based on the current business hour parameter to generate adjusted predicted business data;

[0021] Step S23: Perform product production mapping on the adjusted predicted business data to obtain the estimated product quantity;

[0022] Step S24: Expand the standard processes for the estimated product quantity based on the preset product process data to obtain process load data;

[0023] Step S25: Perform cross-mapping on the process load data and the complete working hour data to generate demand calibration data; perform efficiency modeling on the demand calibration data to generate efficiency demand standards.

[0024] The present invention provides a time basis for subsequent analysis through the acquisition of the current business hour parameter. The trend extrapolation of the business density data improves the predictability of future business changes. The generation of the adjusted predicted business data ensures the adaptation to periodic fluctuations. The mapping of the estimated product quantity realizes the quantification of production demands. The expansion of the process load data provides a specific basis for resource allocation. The demand calibration data generated by cross-mapping ensures the reasonable allocation of various resources. The implementation of efficiency modeling provides a scientific standard for the matching of personnel allocation and process load. The overall process enhances the operational flexibility and efficiency of enterprises in a dynamic environment and supports accurate personnel scheduling decisions.

[0025] Preferably, step S25 includes the following steps:

[0026] Perform multi-dimensional space projection on the process load data and the complete working hour data to obtain projection calibration data; perform fractal feature decomposition on the projection calibration data to generate fractal calibration data;

[0027] Perform manifold surface fitting on the fractal calibration data to obtain manifold calibration data; perform redundancy compression processing on the manifold calibration data to generate required calibration data;

[0028] Extract the fluctuation characteristics of the required calibration data to obtain the required fluctuation characteristics; capture the periodic changes of the required fluctuation characteristics to generate the required fluctuation period data;

[0029] Perform dynamic efficiency tensor transformation on the required calibration data based on the required fluctuation period data to generate the efficiency requirement standard.

[0030] The present invention provides a new perspective for the integration of process load data and complete working hours data through multi-dimensional space projection. The fractal feature decomposition reveals the complex patterns in the data. The manifold surface fitting optimizes the expression of the data. The redundancy compression processing improves the processing efficiency of the data. The extraction of the required fluctuation characteristics enables a deeper understanding of the demand changes. The capture of periodic changes ensures an accurate grasp of the demand fluctuation law. The dynamic efficiency tensor transformation provides real-time adaptability for the generation of the efficiency requirement standard. The overall process improves the enterprise's analysis ability and decision support for complex data environments, and enhances the scientificity and flexibility of personnel scheduling.

[0031] Preferably, step S3 includes the following steps:

[0032] Step S31: Perform process mapping on the estimated product quantity to obtain the estimated required processes;

[0033] Step S32: Derive the process working hours of the estimated required processes according to the efficiency requirement standard to generate the estimated required basic working hours;

[0034] Step S33: Perform site distribution mapping on the estimated product quantity according to the preset workstation data to obtain the site distribution data;

[0035] Step S34: Calculate the required auxiliary working hours of the estimated required basic working hours according to the site distribution data to generate the estimated required auxiliary working hours;

[0036] Step S35: Integrate the estimated required basic working hours and the estimated required auxiliary working hours to obtain the estimated complete working hours; perform work efficiency measurement processing on the estimated product quantity based on the estimated complete working hours to generate the estimated PPMH data.

[0037] The present invention provides a clear production process guide for estimating the quantity of products through process mapping. The derivation of process working hours ensures a scientific estimation of the time requirements for processes. The mapping of station distribution provides a basis for spatial optimization of resource allocation. The calculation of auxiliary working hours supplements the integrity of working hour data. The integration of working hours realizes the effective combination of basic working hours and auxiliary working hours. The generation of estimated complete working hours provides an accurate reference for shift scheduling decisions. The implementation of work efficiency measurement processing improves the quantitative analysis of production efficiency. The overall process enhances the scientific nature and efficiency of the enterprise in production scheduling and personnel management, making production operations more flexible and adaptable.

[0038] Preferably, step S34 includes the following steps:

[0039] Step S341: Conduct a statistical analysis of the basic working hour density of stations for the estimated required basic working hours based on the station distribution data to obtain a collection of station working hour densities;

[0040] Step S342: Make an initial judgment on the auxiliary working hours based on the collection of station working hour densities to generate initial auxiliary working hours;

[0041] Step S343: Identify the collaborative working hours for the initial auxiliary working hours to obtain the collaborative auxiliary working hours; conduct a station collaboration simulation on the station distribution data based on the collaborative auxiliary working hours to generate the omissible auxiliary working hours;

[0042] Step S344: Calculate the working hours for the initial auxiliary working hours and the omissible auxiliary working hours to obtain the estimated required auxiliary working hours.

[0043] The present invention provides a quantitative analysis of the working hour requirements of each station through the statistical analysis of the basic working hour density of stations. The judgment of the initial auxiliary working hours lays a foundation for subsequent working hour calculations. The identification of collaborative working hours improves the resource utilization efficiency. The station collaboration simulation provides a basis for optimizing the operation process. The generated omissible auxiliary working hours reduce unnecessary time waste. The calculation combining the initial auxiliary working hours and the omissible auxiliary working hours ensures the accuracy of the estimated required auxiliary working hours. The overall process enhances the fine management and allocation ability of working hour data, providing strong support for the scientific nature and efficiency of personnel shift scheduling management.

[0044] Preferably, step S343 includes the following steps:

[0045] Extract the working hour types from the initial auxiliary working hours to obtain the auxiliary working hour types; conduct an analysis of station restrictions on the auxiliary working hour types to generate the station restriction data for each working hour;

[0046] Traverse the work-hour site limit data. When the work-hour site limit data is the restricted site work-hour data, access the next work-hour site limit data; when the work-hour site limit data is the unrestricted site work-hour data, package the unrestricted site work-hour data to obtain the collaborative auxiliary work-hours.

[0047] Enumerate adjacent sites for the site distribution data based on the collaborative auxiliary work-hours to obtain adjacent site data; perform collaborative auxiliary work-hour digestion on the collaborative auxiliary work-hours based on the adjacent site data to obtain the collaborative digestion auxiliary work-hours.

[0048] Conduct work-hour quantization comparison on the collaborative digestion auxiliary work-hours and the collaborative auxiliary work-hours to generate the omissible auxiliary work-hours.

[0049] The present invention provides a clear basis for the classification of auxiliary work-hours by extracting work-hour types, ensures reasonable supervision of work-hour allocation through site limit analysis, enhances the comprehensive understanding of the working environment by traversing each work-hour site limit data, provides the possibility for collaborative work through packaging the unrestricted site work-hour data, optimizes the spatial layout of resource scheduling through adjacent site enumeration, improves work efficiency and resource utilization rate through collaborative auxiliary work-hour digestion, provides data support for determining the omissible auxiliary work-hours through work-hour quantization comparison, and the overall process improves the flexibility and accuracy of work-hour management, providing strong support and basis for scientific decision-making in personnel scheduling.

[0050] Preferably, step S4 includes the following steps:

[0051] Step S41: Perform product-manpower matching mapping on the estimated PPMH data to generate the manpower required for each product.

[0052] Step S42: Allocate personnel for the estimated product quantity according to the manpower required for each product to obtain the initial scheduling plan.

[0053] Step S43: Extract the business model from the historical business data to obtain the historical business model; simulate the business scenario for the historical business model to obtain the simulated business field.

[0054] Step S44: Perform business simulation on the initial scheduling plan based on the simulated business field to generate the simulated business data.

[0055] The present invention provides specific human resource guidance for the production requirements of each product through product-human matching mapping. The personnel allocation ensures the rationality and effectiveness of the initial shift scheduling plan. The business model extraction provides a structured perspective for the analysis of historical data. The business scenario simulation enhances the response ability in different business situations. The business simulation provides data basis for the feasibility evaluation of the initial shift scheduling plan. The overall process improves the scientificity and accuracy of personnel scheduling, supports the dynamic adjustment and optimization of future production plans, and enhances the adaptability and resource allocation efficiency of the enterprise in a complex environment.

[0056] Preferably, step S5 includes the following steps:

[0057] Step S51: Perform time slice processing on the simulated business data to obtain business time period data;

[0058] Step S52: Identify the extreme values of personnel requirements for the business time period data based on the estimated PPMH data to obtain the extreme values of personnel requirements for the time period;

[0059] Step S53: Analyze the rotation order constraints for the extreme values of personnel requirements for the time period to obtain personnel rotation constraints;

[0060] Step S54: Reconstruct the scheduling order of the initial shift scheduling plan based on the personnel rotation constraints to obtain the rotated scheduling order, and send the rotated scheduling order to the terminal to perform personnel scheduling management.

[0061] The present invention provides a more detailed time analysis basis for business data through time slice processing. The identification of extreme values of personnel requirements ensures the accurate grasp of the manpower requirements during peak hours. The analysis of rotation order constraints provides rationality and flexibility for personnel scheduling. The reconstruction of the scheduling order optimizes the execution effect of the initial shift scheduling plan. Sending the rotated scheduling order to the terminal realizes an immediate management response. The overall process improves the scientificity and efficiency of personnel scheduling, and supports the rapid adaptation and reasonable resource allocation of the enterprise in a dynamic business environment.

[0062] The present invention also provides a personnel scheduling management system based on the PPMH index for executing the personnel scheduling management method based on the PPMH index as described above. The personnel scheduling management system based on the PPMH index includes:

[0063] A data collection module for collecting historical business data; estimating business density based on the historical business data to obtain business density data; performing working hour mapping on the business density data to generate complete working hour data;

[0064] A demand estimation module for estimating business demand based on the business density data to obtain the estimated product quantity; calibrating the demand for the complete working hour data according to the estimated product quantity to generate the efficiency demand standard;

[0065] The man-hour inference module is used to perform man-hour simulation inference on the estimated product quantity according to the efficiency requirement standard to obtain the estimated required basic man-hours; perform ergonomic measurement processing on the estimated product quantity and the estimated required basic man-hours to generate estimated PPMH data;

[0066] The shift scheduling mapping module is used to perform personnel allocation mapping on the estimated product quantity based on the estimated PPMH data to obtain an initial shift scheduling plan; perform business simulation according to the initial shift scheduling plan to generate simulated business data;

[0067] The shift scheduling reconstruction module is used to reconstruct the rotation shift scheduling order of the initial shift scheduling plan according to the simulated business data to obtain the rotation shift scheduling order, and send the rotation shift scheduling order to the terminal to execute personnel shift scheduling management.

[0068] Through the implementation of the data acquisition module, the present invention ensures the comprehensive acquisition of historical business data. The business density estimation provides a quantitative analysis of the changes in business requirements, and the generated business density data lays a scientific foundation for man-hour mapping. The generation of complete man-hour data ensures the effective management of resource utilization. The demand estimation module realizes the accurate prediction of future product demands. The efficiency requirement standard provides a clear guiding basis for personnel allocation. The application of the man-hour inference module makes the estimation of basic man-hours more scientific. The estimated PPMH data provides a quantitative reference for shift scheduling decisions. The application of the shift scheduling mapping module ensures the rationality and feasibility of the initial shift scheduling plan. The implementation of business simulation verifies the effectiveness of the shift scheduling plan in actual operation. The shift scheduling reconstruction module improves the flexibility and adaptability of shift scheduling. The final rotation shift scheduling order is sent to the terminal to execute the high efficiency of personnel management. The overall system improves the enterprise's management level and resource allocation efficiency, providing strong technical support for optimizing operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a schematic step flow diagram of a personnel shift scheduling management method based on the PPMH index;

[0070] Figure 2 is Figure 1 a detailed implementation step flow diagram of step S2 in

[0071] Figure 3 is Figure 1 a detailed implementation step flow diagram of step S3 in

[0072] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0073] The technical method of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those skilled in the art within the scope of the present invention without creative work belong to the scope of protection of the present invention.

[0074] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0075] It should be understood that although terms such as "first" and "second" may be used here to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit can be called the second unit, and similarly the second unit can be called the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed associated items.

[0076] To achieve the above object, please refer to Figures 1 to 3 , a personnel scheduling management method based on the PPMH index, including the following steps:

[0077] Step S1: Collect historical business data; estimate business density based on the historical business data to obtain business density data; perform working hour mapping on the business density data to generate complete working hour data;

[0078] Step S2: Estimate business demand based on the business density data to obtain the estimated product quantity; calibrate the demand for the complete working hour data according to the estimated product quantity to generate the efficiency demand standard;

[0079] Step S3: Perform working hour simulation inference on the estimated product quantity according to the efficiency demand standard to obtain the estimated basic working hours required; perform work efficiency measurement processing on the estimated product quantity and the estimated basic working hours required to generate the estimated PPMH data;

[0080] Step S4: Perform personnel allocation mapping on the estimated product quantity based on the estimated PPMH data to obtain the initial scheduling plan; perform business simulation according to the initial scheduling plan to generate simulated business data;

[0081] Step S5: Reconstruct the rotation scheduling order of the initial scheduling plan according to the simulated business data to obtain the rotation scheduling order, and send the rotation scheduling order to the terminal to execute personnel scheduling management.

[0082] Through collecting historical business data, the present invention can establish a comprehensive business foundation. The business density estimation provides an in-depth understanding of demand fluctuations. The generated business density data provides a scientific basis for subsequent working hour mapping and demand prediction. The generation of complete working hour data ensures an accurate grasp of resource utilization. The business demand prediction can quantify future product demands. The formulation of efficiency demand standards provides a clear reference for personnel allocation. The working hour simulation inference realizes a scientific estimation of basic working hours. The generated predicted PPMH data (PPMH (Product Per Man Hour) index definition: the number of products created per person per hour) provides a quantitative index for scheduling decisions. The personnel allocation mapping ensures the rationality of the initial scheduling plan. The business simulation verifies the feasibility of the scheduling plan. The reconstruction of the rotation scheduling order improves the flexibility and adaptability of scheduling. The final implementation of scheduling management realizes the efficient allocation of human resources. The overall process improves the scientificity and accuracy of management, providing more efficient operation support for enterprises.

[0083] In an embodiment of the present invention, the personnel scheduling management method based on the PPMH index includes the following steps:

[0084] Step S1: Collect historical business data; perform business density estimation according to the historical business data to obtain business density data; perform working hour mapping on the business density data to generate complete working hour data;

[0085] In this embodiment, when collecting historical business data, it is necessary to call the transaction record database in the restaurant's point-of-sale (POS) cash register terminal, retrieve each order information in the past at least six months, including timestamp, product category, sales quantity, and turnover, obtain the original transaction log through a data interface (API) or batch export function, convert the original log into a standardized data format, remove invalid data, reconstruct the transaction details of each business period according to the time series, form a structured historical business data table. The business density estimation needs to be based on time series regression modeling, and a smoothing processing method based on kernel density estimation (KDE) is used to fit the probability distribution of the order flow at different times of each day, model the distribution of the order quantity at hourly, half-hourly or finer time intervals, generate the business density data for the corresponding period. The business density data needs to be dynamically weighted in combination with variables such as the number of restaurant seats and equipment availability to generate a more accurate business density distribution table. The working hours mapping needs to be based on the historical working hours allocation data of the restaurant. The business density data is input into a preset mapping matrix, and combined with the product production complexity parameters of different workstations (such as wok cooking, grilling, frying, etc.), the required operating working hours for each period are calculated, and finally a complete working hours data table is output. The complete working hours data includes fields such as timestamp, estimated required working hours, and workstation load ratio.

[0086] Step S2: Estimate the business demand based on the business density data to obtain the estimated product quantity; calibrate the demand for the complete working hours data according to the estimated product quantity to generate the efficiency demand standard;

[0087] In this embodiment, the business demand estimation needs to be based on the business density data, call the autoregressive integrated moving average model (ARIMA) based on the historical product sales trend, perform time series prediction on the product sales trend in the future business period, combine the product category coefficient, and weight the product sales quantities of different categories in each period to output an estimated product quantity table. The demand calibration needs to be based on the complete working hours data, input the estimated product quantity into the PPMH standard library, determine the minimum working hours demand for each business period according to the mapping relationship between the product quantity and the recommended working hours, combine the available human resources data of the restaurant, calculate the working hours load coefficient, and generate the efficiency demand standard. The efficiency demand standard includes the basic working hours required for each business period, the recommended job allocation, and the floating range of human resource demand.

[0088] Step S3: Perform working hours simulation inference on the estimated product quantity according to the efficiency demand standard to obtain the estimated required basic working hours; perform work efficiency measurement processing on the estimated product quantity and the estimated required basic working hours to generate the estimated PPMH data;

[0089] In this embodiment, the man-hour simulation inference needs to be based on the efficiency requirement standard, construct a man-hour requirement distribution model based on Monte Carlo Simulation, use the estimated product quantity as the input variable, simulate the task distribution of different workstations, calculate the man-hour requirements of each workstation, and generate an estimated basic man-hour table. The ergonomic metric processing needs to input the estimated product quantity and the estimated basic man-hours into the PPMH calculation formula, combine the historical PPMH reference data, and use the weighted moving average method for correction to remove the influence of extreme values and generate the estimated PPMH data. The estimated PPMH data includes the estimated production efficiency indicators for each business period and can be used for subsequent shift scheduling optimization.

[0090] Step S4: Perform personnel allocation mapping on the estimated product quantity based on the estimated PPMH data to obtain an initial shift scheduling plan; conduct business simulation based on the initial shift scheduling plan to generate simulated business data;

[0091] In this embodiment, the personnel allocation mapping needs to be based on the estimated PPMH data, call an optimization model based on Integer Linear Programming (ILP), determine the optimal personnel ratio for each business period, consider different job skill matching degrees, minimum man-hour constraints, and human cost control objectives, and output the initial shift scheduling plan. The business simulation needs to be based on the initial shift scheduling plan and use the Discrete Event Simulation (DES) method to construct a complete simulation process including order generation, product processing, meal delivery, and personnel scheduling. Use the estimated product quantity as the input variable to simulate the business operation under different manpower configurations and generate simulated business data. The simulated business data includes the actual man-hour utilization rate, personnel load, and order processing time for each time period.

[0092] Step S5: Reconstruct the rotation scheduling order of the initial shift scheduling plan according to the simulated business data to obtain the rotation scheduling order, and send the rotation scheduling order to the terminal to perform personnel shift scheduling management.

[0093] In this embodiment, the reconstruction of the rotation shift sequence needs to be based on the simulated business data, call the shift adjustment algorithm based on heuristic optimization, calculate the task load and shift continuity of each employee, rotate and adjust the shifts with relatively low working hour utilization rates, and use methods such as bubble sort or binary tree search to find the optimal shift adjustment path, generate a rotation shift sequence table. The rotation shift sequence table includes the final shift arrangements of each employee, job adjustment suggestions, and the optimization results of manpower utilization rate. Finally, send the rotation shift sequence to the terminal. The terminal can be connected to the restaurant's Human Resource Management System (HRMS) to automatically update the shift information and ensure the accurate execution of personnel shift management.

[0094] Preferably, step S1 includes the following steps:

[0095] Step S11: Collect historical business data; perform time series decomposition on the historical business data to obtain product time series;

[0096] Step S12: Perform category merging on the product time series to obtain category distribution data; perform peak detection on the category distribution data to generate peak period feature data;

[0097] Step S13: Perform business density estimation on the peak period feature data to obtain business density data;

[0098] Step S14: Perform working hour mapping on the business density data to generate basic working hour data; perform auxiliary task calculation on the basic working hour data to obtain auxiliary working hour data;

[0099] Step S15: Integrate the basic working hour data and the auxiliary working hour data to generate complete working hour data.

[0100] In this embodiment, when collecting historical business data, it is necessary to use the data export interface in the restaurant's point-of-sale (POS) cashier terminal to batch export the daily transaction details for at least the past 12 months. The data format is a CSV (comma-separated values) file, and the content includes order number, timestamp, product name, quantity, price, and cashier identifier. Use a data cleaning tool such as OpenRefine (an open-source data cleaning tool) to clean the exported raw data, delete duplicate records, repair missing values, and unify the time format. Import the cleaned data into the Python data analysis library Pandas (a data analysis library), sort all transaction records according to the timestamp, and use a time series decomposition method, such as STL decomposition (Seasonal-Trend decomposition using LOESS, a seasonal-trend decomposition method with locally weighted regression smoothing), to decompose the historical business data. Decompose the time series into a long-term trend term, a seasonal term, and a residual term, extract and reconstruct the time series for each product. The time series includes date, time period, sales quantity, and product identifier. When performing category merging on the product time series, use the grouping query function in the database query language SQL (Structured Query Language) to merge the time series of all products according to the product category field, accumulate the sales quantities of products in the same category within the same time period, and output a category distribution data table. The category distribution data table includes time period, category identifier, and total sales volume. The category distribution data is subjected to peak detection by calling the SciPy library (a scientific computing library) in Python, and a spectral analysis method based on the discrete Fourier transform (DFT, Discrete Fourier Transform) is used to analyze the curve of sales quantity changing with time, calculate the sales frequency and amplitude for each time period, and set the amplitude threshold to 1. of the sales average.By 5 times, filter out the time periods higher than the threshold as the sales peak, output the characteristic data of the peak period. The characteristic data of the peak period includes the peak time period, category, and sales quantity. When estimating the business density of the characteristic data of the peak period, use the Kernel Density Estimation (KDE) method for smoothing. Adopt the Gaussian Kernel function, and set the bandwidth parameter to 15 minutes. Call the KernelDensity module through the Scikit-learn library (machine learning library) in Python to perform probability density estimation on the sales data of each time period within the peak period, generate the sales density distribution within the time period, and output the business density data. The business density data includes the time period and the corresponding density value. When performing working hours mapping on the business density data, call the working hours mapping model based on Linear Programming (LP). Input the business density data and the working hours standards of each position in the restaurant. The working hours standards are derived from the historical working hours records of the restaurant. Define the standard production time for each product, set the maximum working hours limit for each position, and calculate the required working hours for each position in each time period by solving the linear programming model, and output the basic working hours data. The basic working hours data includes the time period, position, and required working hours. For the calculation of auxiliary tasks, it is necessary to call the auxiliary task allocation algorithm based on the Rule Engine. Input the basic working hours data into the rule engine, and dynamically allocate auxiliary tasks such as cleaning and ingredient preparation according to the task type and priority, calculate the required auxiliary working hours, and output the auxiliary working hours data. The auxiliary working hours data includes the task type, time period, and required working hours. When integrating the basic working hours data and the auxiliary working hours data, call the Pandas library in Python, import the basic working hours data and the auxiliary working hours data as two data frames, and merge the two data frames according to the time period field through Inner Join operation. After merging, accumulate the working hours within the same time period to generate the complete working hours data. The complete working hours data includes the time period, position, the sum of basic working hours and auxiliary working hours.

[0101] Preferably, step S2 includes the following steps:

[0102] Step S21: Obtain the current business hours parameter; perform trend extrapolation on the business density data to obtain the business trend prediction data;

[0103] Step S22: Perform periodic adjustment on the business trend prediction data based on the current business hours parameter to generate the adjusted predicted business data;

[0104] Step S23: Perform product production mapping on the adjusted predicted business data to obtain the estimated product quantity;

[0105] Step S24: Based on the preset product process data, expand the estimated product quantity according to the standard process to obtain the process load data;

[0106] Step S25: Perform cross-mapping on the process load data and the complete working hour data to generate the demand calibration data; perform efficiency modeling on the demand calibration data to generate the efficiency requirement standard.

[0107] In this embodiment, when obtaining the current business hours parameter, the real-time query module of the Business Management System is called. Through the database query statement SELECT, parameters such as the start time, end time, hour-level time period division, and business hours identifier of the current business day are retrieved from the system database. The parameters are exported in JSON (JavaScript Object Notation) format, and the content includes time period labels, corresponding hours, and durations. This is used as the time benchmark to input into the trend extrapolation model. The Prophet library in Python (a time series prediction library) is called for trend extrapolation processing. The business density data is imported into the Prophet model, and holiday parameters, business hours length, and historical business cycles are set. The Bayesian Sampling method is used to train the parameter distributions of the trend term, seasonal term, and holiday term. The current business hours parameter is input, and the time index of the Prophet model is adjusted to the current business time period. The trend prediction function predict is executed to output the business trend prediction data. The business trend prediction data includes the predicted business density and confidence interval for each future time period. When performing periodic adjustment on the business trend prediction data based on the current business hours parameter, the SARIMA model (Seasonal AutoRegressive Integrated Moving Average) built into the statsmodels library in Python (a statistical modeling library) is called. The business trend prediction data is used as the input time series, and the current business hours parameter is set as the period adjustment cycle. The autocorrelation function (ACF) and partial autocorrelation function (PACF) are used to analyze the periodic characteristics of the prediction data. The seasonal period is set to 7 days, and the differencing order is 1 for seasonal differencing operations. The SARIMA model is fitted, and the fit function is called to perform parameter estimation. The predict function is used to perform periodic adjustment within the current business time period, and the adjusted prediction business data is output. The adjusted prediction business data includes the predicted business density for each adjusted time period. When performing product production mapping on the adjusted prediction business data, the mapping method based on the pivot table is called. The adjusted prediction business data is imported into the pivot table function of Excel (a spreadsheet software). The time period is set as the row label, the product category is set as the column label, and the predicted density value is set as the numerical field. The production requirements for each type of product in each time period are automatically generated through the pivot table. The data generated by the pivot table is read using the openpyxl library in Python (an Excel operation library). According to the average production time and sales frequency of each type of product, the matrix multiplication method is used to calculate the estimated product quantity for each time period, and the estimated product quantity data is output.The data includes time periods, product categories, and estimated quantities. When performing standard process expansion on the estimated product quantities based on the preset product process data, the query module of the database management system MySQL is called to obtain the standard process data for each product category from the preset process database. The standard process data includes process numbers, required time, equipment, and the number of workers. The NumPy library (numerical calculation library) in Python is used to perform a dot product operation on the estimated product quantity matrix and the standard process data matrix, mapping the quantity of each product category to the man-hours and resources required for the corresponding processes, and outputting process load data. The process load data includes time periods, process numbers, man-hours, and the number of workers required. When performing cross-mapping on the process load data and the complete man-hour data, the merge function of the Pandas library is called to perform an inner join on the process load data and the complete man-hour data through the time period field, matching the process load and available man-hours in the same time period, calculating the man-hour difference within each time period, and generating demand calibration data. The demand calibration data includes time periods, process numbers, man-hour requirements, and available man-hours. When performing efficiency modeling on the demand calibration data, the linear regression model (LinearRegression linear regression) of the Scikit-learn library is called, using the man-hour difference as the independent variable and personnel allocation and completion rate as the dependent variables to train the regression model, and calling the predict function to predict the efficiency requirements for the newly input man-hour difference, outputting the efficiency requirement standard. The efficiency requirement standard includes the optimal number of personnel and man-hour allocation within each time period.,

[0108] Preferably, step S25 includes the following steps:

[0109] Perform multi-dimensional space projection on the process load data and the complete man-hour data to obtain projection calibration data; perform fractal feature decomposition on the projection calibration data to generate fractal calibration data;

[0110] Perform manifold surface fitting on the fractal calibration data to obtain manifold calibration data; perform redundancy compression processing on the manifold calibration data to generate demand calibration data;

[0111] Extract the fluctuation characteristics of the demand calibration data to obtain demand fluctuation characteristics; capture the periodic changes of the demand fluctuation characteristics to generate demand fluctuation cycle data;

[0112] Perform dynamic efficiency tensor transformation on the demand calibration data based on the demand fluctuation cycle data to generate the efficiency requirement standard.

[0113] In this embodiment, when performing multi-dimensional space projection on the process load data and the complete working hour data, the PCA (Principal Component Analysis) algorithm in the scikit-learn library (machine learning library) in Python is called to merge the process load data and the complete working hour data into a high-dimensional matrix. Each row of the matrix represents the process and working hour data for a time period, and each column represents different process or time dimensions. The fit_transform function is called to perform principal component decomposition on the data matrix, and the top three principal components with the highest proportion of explained variance are extracted to generate three-dimensional space coordinates. The three-dimensional projection diagram is drawn through the Matplotlib library (plotting library) and the projection calibration data is output. The projection calibration data includes the three-dimensional coordinate values corresponding to each time period. When performing fractal feature decomposition on the projection calibration data, the pywt library (wavelet transform library) in Python is called for multi-scale fractal decomposition. The wavelet basis is set as the Daubechies wavelet, and the decomposition level is 4 layers. The discrete wavelet transform is performed on the projection calibration data to decompose the data detail coefficients and approximation coefficients at different scales. The threshold processing function is called to filter the high-frequency noise part and retain the multi-scale feature components, and the fractal calibration data is output. The fractal calibration data contains the process load and working hour distribution characteristics at different scales. When performing manifold surface fitting on the fractal calibration data, the ISOMAP algorithm (Isometric Mapping) in the manifold library (manifold learning library) in Python is called. The fractal calibration data is input into the ISOMAP algorithm, the number of neighbors is set to 10, and the maximum number of iterations is set to 1000. The fit_transform function is called for dimensionality reduction calculation to generate low-dimensional manifold coordinates. The griddata function of the SciPy library (scientific computing library) is used for triangular interpolation to interpolate the low-dimensional manifold coordinates into a continuous surface, and the manifold calibration data is output. The manifold calibration data contains the manifold coordinates and interpolation surface corresponding to each time period. When performing redundancy compression processing on the manifold calibration data, sklearn. in Python is called.The TruncatedSVD algorithm (Truncated Singular Value Decomposition) of the decomposition library inputs the manifold calibration data matrix into the TruncatedSVD model, sets the number of singular values to be retained as 20, calls the fit_transform function to decompose the data matrix, retains the eigenvectors corresponding to the first 20 singular values, filters out redundant dimensions, and outputs the required calibration data. The required calibration data includes the compressed process load and man-hour demand features. When extracting the fluctuation features of the required calibration data, call the tsfresh library (time series feature extraction library) in Python, input the required calibration data as a time series, set the window size to 5, and call the extract_features function to extract statistical features such as skewness, kurtosis, autocorrelation coefficient, and time delay features of the time series, and output the demand fluctuation features. The demand fluctuation features include the dynamic fluctuation features of the process load and man-hours within the time period. When capturing the periodic changes of the demand fluctuation features, call the STL decomposition algorithm (Seasonal-Trend decomposition using Loess) of the statsmodels library in Python, input the demand fluctuation features into the STL decomposition model, set the seasonal period to 24 hours and the smoothness to 0.05, call the fit function to decompose into periodic components, trend components, and residual components, extract the periodic components as the output, and generate the demand fluctuation cycle data. The demand fluctuation cycle data includes the periodic process load and man-hour demand fluctuations for each time period. When performing dynamic efficiency tensor transformation on the required calibration data based on the demand fluctuation cycle data, call the Tensorly library (tensor calculation library) in Python, merge the required calibration data and the demand fluctuation cycle data into a third-order tensor, and the tensor dimensions are time period, process type, and man-hour demand respectively. Call the PARAFAC algorithm (Parallel Factor Analysis) of the tensorly.decomposition library, set the tensor decomposition rank to 10, call the parafac function to decompose the tensor, extract the factor matrices of different dimensions, call the mode_dot function of the tensorly.tenalg library for mode tensor multiplication, map the demand fluctuation cycle data to the required calibration data, and output the efficiency demand standard. The efficiency demand standard includes the process load, man-hour demand, and the corresponding personnel scheduling efficiency configuration for each time period.

[0114] Preferably, step S3 includes the following steps:

[0115] Step S31: Perform process mapping on the estimated product quantity to obtain the estimated required processes;

[0116] Step S32: Derive the process man-hours for the estimated required processes according to the efficiency demand standard to generate the estimated required basic man-hours;

[0117] Step S33: Perform site distribution mapping on the estimated product quantity according to the preset workstation data to obtain site distribution data;

[0118] Step S34: Calculate the required auxiliary working hours for the estimated required basic working hours according to the site distribution data to generate the estimated required auxiliary working hours;

[0119] Step S35: Integrate the estimated required basic working hours and the estimated required auxiliary working hours to obtain the estimated complete working hours; perform work efficiency measurement processing on the estimated product quantity based on the estimated complete working hours to generate the estimated PPMH data.

[0120] In this embodiment, when performing process mapping on the estimated product quantity, the pandas library (data analysis library) in Python is called to import a preset product process mapping table. The mapping table is stored in the form of a CSV file and contains the process number, process name, and process sequence corresponding to each product. The data table is loaded as a DataFrame data frame through the read_csv function. The merge function is called to associate the estimated product quantity data with the process mapping table according to the product number, and the merged data is sorted according to the process sequence. The groupby function is called to group by the product number, and the process set and quantity required for each product are counted, and the estimated required processes are output. The estimated required processes are the process list required for each product and the corresponding quantity. When deriving the process man-hours for the estimated required processes according to the efficiency requirement standard, the numpy library (numerical calculation library) of Python is called to load the efficiency requirement standard matrix. The matrix is read from an NPY file through the np.load function. The matrix dimensions are the number of processes, time periods, and unit man-hour efficiency. The merge function of the pandas library is called to match the estimated required processes with the efficiency requirement standard matrix, and the unit man-hour efficiency corresponding to each process is extracted. The multiply function of the numpy library is called to multiply element by element, and the unit man-hour efficiency is multiplied by the process quantity to obtain the man-hour requirement for each process. The sum function is called to summarize by product number, and the estimated required basic man-hours are output. The estimated required basic man-hours include the total man-hour requirement corresponding to each product. When performing site distribution mapping on the estimated product quantity according to the preset workstation data, the geopandas library (geographic data processing library) in Python is called to load the workstation location data. The data is stored in the GeoJSON format and contains the geographical coordinates, site number, and production capacity limit of each workstation. It is loaded as a GeoDataFrame through the read_file function. The spatialjoin function is called to perform a spatial join according to the geographical coverage of the product production area and the workstation, and calculate the product quantity distribution of each site. The groupby function is called to count the product quantity that each workstation needs to produce according to the site number, and the site distribution data is output. The site distribution data includes the number, geographical coordinates of each workstation, and the corresponding product production task quantity. When calculating the required auxiliary man-hours for the estimated required basic man-hours according to the site distribution data, the networkx library (network analysis library) of Python is called to load the logistics network data between sites. The data is stored in the GraphML format and contains the transportation time, distance, and transportation capacity of each edge. It is loaded as a directed graph through the read_graphml function. The shortest_path function is called to calculate the shortest transportation path and time from each workstation to the main warehouse. The merge function of the pandas library is called to associate the site distribution data with the transportation time data, and the auxiliary man-hour requirement of each site is calculated by weighting according to the transportation time.Call the sum function to accumulate the auxiliary working hours of all stations, and output the estimated required auxiliary working hours. The estimated required auxiliary working hours are the transportation and handling working hour requirements allocated to each station. When integrating the estimated required basic working hours and the estimated required auxiliary working hours, call the pandas library in Python to load the data frames of the estimated required basic working hours and the estimated required auxiliary working hours, call the merge function to merge the two data frames according to the product number, call the add function to add the basic working hours and the auxiliary working hours of the same product row by row, and output the estimated complete working hours. The estimated complete working hours include the total working hour requirements of each product. When performing work efficiency measurement processing on the estimated product quantity based on the estimated complete working hours, call the linear regression model of the sklearn library in Python, use the estimated product quantity as the independent variable and the estimated complete working hours as the dependent variable, call the fit function to train the model, call the predict function to generate the working hour prediction values of each product, call the pandas library to calculate the production quantity per unit working hour of each product, and output the estimated PPMH data. The estimated PPMH data includes the unit working hour output index of each product.,

[0121] Preferably, step S34 includes the following steps:

[0122] Step S341: Conduct site basic working hour density statistics on the estimated required basic working hours according to the site distribution data to obtain a collection of site working hour densities;

[0123] Step S342: Make an initial auxiliary working hour judgment based on the collection of site working hour densities to generate initial auxiliary working hours;

[0124] Step S343: Identify the collaborative working hours for the initial auxiliary working hours to obtain the collaborative auxiliary working hours; conduct site collaboration simulation on the site distribution data according to the collaborative auxiliary working hours to generate the omissible auxiliary working hours;

[0125] Step S344: Calculate the working hours for the initial auxiliary working hours and the omissible auxiliary working hours to obtain the estimated required auxiliary working hours.

[0126] In this embodiment, when performing site-based man-hour density statistics on the estimated required basic man-hours according to the site distribution data, it is necessary to first import the site distribution data into the time analysis module, and use a multidimensional dataset processing tool such as Apache Kylin to perform data slicing operations on fields such as site ID, estimated product quantity, process type, and standard man-hours. By defining the time dimension granularity as hours, days, and weeks, a multi-level pivot table is established, and a hierarchical aggregation algorithm is used to accumulate and statistically analyze the man-hour requirements of each site within each time dimension. A piecewise clustering algorithm such as K-Means clustering is used to perform density analysis on the man-hour distribution of each site in different time periods, generating the man-hour density statistical results of each site in different time dimensions, and outputting the results as a site man-hour density collection. Each record in the site man-hour density collection contains the site ID, time granularity, process type, cumulative man-hours, and man-hour density value. When making an initial auxiliary man-hour judgment based on the site man-hour density collection, it is necessary to import the man-hour density collection into a data analysis tool such as the Pandas library of Python for data screening and classification operations, and use a conditional filtering function to divide the interval according to the man-hour density value, setting the man-hour density threshold to 0.8 working hours per hour, filter out the sites and time periods with working hour density exceeding the threshold, use a linear regression model to perform fitting analysis on the filtered working hour data, predict the auxiliary working hour demand, use the NumPy library for matrix operations, calculate the initial auxiliary working hour demand of each site in different time periods according to the product of the process standard working hour matrix and the working hour density matrix, and the finally output initial auxiliary working hour data includes site ID, time period, auxiliary working hour demand and process type. When identifying the collaborative working hours of the initial auxiliary working hours, it is necessary to import the initial auxiliary working hour data into a graph theory analysis tool such as the NetworkX library. By constructing a site-process collaborative network graph, each site and process are regarded as nodes, and the working hour demand relationship between the site and the process is regarded as an edge. Use the minimum spanning tree algorithm to optimize the collaborative network graph and find the optimal collaborative path of the working hour distribution. Use a graph cut algorithm such as Karger’s Algorithm to partition the collaborative network, identify the site groups and process groups whose working hours can be shared under collaborative operations, calculate the shared working hour amount and generate the collaborative auxiliary working hour data, and output a CSV file containing the site group ID, shared working hour amount and process type. At the same time, according to the collaborative auxiliary working hour data, use the AnyLogic simulation tool to perform site collaboration simulation on the site distribution data, set the simulation parameters including site coordinates, working hour transfer time, collaborative execution time, etc., and run the simulation to calculate the auxiliary working hour amount that each site can save under collaborative operations. When calculating the working hours of the initial auxiliary working hours and the omissible auxiliary working hours, it is necessary to import the two types of working hour data into a numerical calculation tool such as MATLAB, and use the matrix addition function to add the initial auxiliary working hour matrix and the negative omissible auxiliary working hour matrix to calculate the final estimated required auxiliary working hour matrix. Each element in the matrix represents the net auxiliary working hour demand of the corresponding site in a specific time period, and the finally output estimated required auxiliary working hour data is stored in the form of an SQL database and associated with the site ID, time period and process type.

[0127] Preferably, step S343 includes the following steps:

[0128] Extract the working hour type of the initial auxiliary working hours to obtain the auxiliary working hour type; perform site limit analysis on the auxiliary working hour type to generate the working hour site limit data of each item;

[0129] Traverse the working hour site limit data of each item. When the working hour site limit data is the restricted site working hour data, access the next working hour site limit data; when the working hour site limit data is the unrestricted site working hour data, package the unrestricted site working hour data to obtain the collaborative auxiliary working hours;

[0130] Enumerate adjacent sites for the site distribution data according to the collaborative auxiliary working hours to obtain adjacent site data; perform collaborative auxiliary working hour digestion on the collaborative auxiliary working hours based on the adjacent site data to obtain the collaborative digested auxiliary working hours;

[0131] Perform a man-hour quantification comparison on the collaborative digestion auxiliary man-hours and the collaborative auxiliary man-hours that can be utilized to generate the auxiliary man-hours that can be omitted.

[0132] In this embodiment, when extracting the working hour type of the initial auxiliary working hours, it is necessary to import the initial auxiliary working hour data into a data classification tool such as the Scikit-Learn library in Python, and use KNN (K-Nearest Neighbor algorithm) to classify the type of working hour data. The fields in the working hour data, including process number, station number, working hour value, task nature, etc., are used as feature inputs. Set the value of K to 5, and perform type clustering by calculating the Euclidean distance between the working hour data. Finally, output the auxiliary working hour type data set. Each record in the data set contains a working hour type label, the corresponding process number, station number, and working hour value. When performing station limit analysis on the auxiliary working hour type, it is necessary to use a constraint solving tool such as IBM ILOG CPLEX to analyze the auxiliary working hour type data set. Import the station distribution data and the auxiliary working hour type data, establish a linear constraint model based on multiple constraint conditions such as station physical location, production capacity limit, equipment type, and operation duration limit, and use the CPLEX solver to solve the model. Output the feasibility analysis results of each working hour type at different stations. Mark the working hour data with a feasibility label of "1" as unrestricted station working hour data, and mark the label of "0" as restricted station working hour data. The output station limit data contains the station number, working hour type label, and restriction mark. When traversing the station limit data of each working hour station, it is necessary to use the iterator tool Itertools library in Python to read the station limit data in the SQL database table, and use the for loop statement to access the station limit data one by one. When reading a record with a restriction mark of "0", call the Next() function to access the next data. When reading a record with a restriction mark of "1", call a JSON packaging tool such as the json library in Python to perform a packaging operation on this data in dictionary form to generate collaborative auxiliary working hour data. The packaged data contains the station number, working hour type label, and collaborative working hour value. When enumerating adjacent stations based on the collaborative auxiliary working hours for the station distribution data, it is necessary to use a geographic analysis tool such as QGIS (open source geographic information system). Import the station distribution data and the collaborative auxiliary working hour data, and call the spatial analysis plugin Nearest NeighborAnalysis in QGIS to analyze the geographic coordinates of each station. Filter out the adjacent stations of each station according to the condition that the geographic distance is less than 500 meters. The output adjacent station data contains the station number, adjacent station number, and geographic distance. When performing collaborative auxiliary working hour digestion on the collaborative auxiliary working hours based on the adjacent station data, it is necessary to use a scheduling optimization tool such as Gurobi Optimizer. Import the collaborative auxiliary working hour JSON data and the adjacent station ShapeFile data, establish a multi-objective optimization model based on the station geographic distance, working hour quantity, and task type, and use the cuckoo search algorithm to solve the model. The number of iterations is set to 100, and the maximum iteration error is 0.001. During the solving process, the collaborative working hours of each station are adjusted to achieve optimal collaborative scheduling. Finally, a dataset of collaborative digestion auxiliary working hours is output. The dataset contains the station number, the numerical value of the working hours after collaborative digestion, and the station numbers participating in the collaboration. When making a quantitative comparison of the collaborative digestion auxiliary working hours and the collaborative auxiliary working hours, it is necessary to use the Matplotlib data visualization library for data comparison and analysis. Import the data of collaborative digestion auxiliary working hours and collaborative auxiliary working hours, and use Bar Plot to visually display the two sets of working hours data for each station. By comparing the differences between the collaborative digestion auxiliary working hours and the original collaborative auxiliary working hours of each station, the difference matrix calculation method is used to calculate the working hours difference of each station. When the working hours difference is negative, it means that the auxiliary working hours can be omitted. The finally output data of the auxiliary working hours that can be omitted is stored in CSV format, including the station number, the numerical value of the working hours that can be omitted, and the process type, which is used as the input data for PPMH index calculation.

[0133] Preferably, step S4 includes the following steps:

[0134] Step S41: Perform product-manpower matching mapping on the estimated PPMH data to generate the manpower required for each product;

[0135] Step S42: Allocate personnel according to the manpower required for each product to the estimated product quantity to obtain the initial shift schedule;

[0136] Step S43: Extract the business model from the historical business data to obtain the historical business model; simulate the business scenario for the historical business model to obtain the simulated business field;

[0137] Step S44: Based on the simulated business field, conduct business simulation on the initial shift schedule to generate simulated business data.

[0138] In this embodiment, when performing product-manpower matching mapping on the estimated PPMH data, it is necessary to import the estimated PPMH data into a data mapping tool such as MATLAB and use the matrix mapping method to process the PPMH data. First, the PPMH data is split into three matrices: product dimension, man-hour dimension, and manpower dimension. The man-hour dimension matrix is multiplied by the manpower dimension matrix using the matrix multiplication function mtimes() in MATLAB to obtain the basic manpower demand matrix for each product. Then, through the matrix dot product operation of the product dimension matrix and the basic manpower demand matrix, item-by-item matching of products and manpower requirements is achieved. The manpower data required for each product output is stored in the CSV file format, and the data content includes product number, process number, required manpower value, and corresponding PPMH index value. When staffing the estimated product quantity according to the manpower required for each product, it is necessary to use a scheduling optimization tool such as Gurobi Optimizer, import the manpower data required for each product and the estimated product quantity data, establish an integer linear programming model based on constraints such as personnel skill level, working time limit, production priority, and shift duration, and use Gurobi's optimization solver to perform multiple rounds of iterative optimization. Set the maximum number of iterations to 500 times and the tolerance value to 0.0001. Through solving, the optimal staffing plan that meets all constraints is obtained, and the staffing data is exported in JSON format. The data content includes product number, assigned personnel number, shift time period, and personnel skill level. When extracting the business model from historical business data, it is necessary to use the Pandas library in Python to read the historical business data, call the time series analysis tool such as the Holt-Winters exponential smoothing method in the statsmodels library to decompose the business data into time series. Fields such as time, sales volume, foot traffic, and business hours in the business data are used as input features, and the smoothing parameters alpha is set to 0.3, beta is set to 0.1, and gamma is set to 0.05. Extract three key pattern elements, namely seasonality, trend, and periodicity, through multiple iterations of calculation. The finally output historical business model dataset includes time period tags, seasonal indices, trend coefficients, and periodic cycle numbers. When simulating business scenarios for the historical business model, it is necessary to use the simulation modeling tool AnyLogic, import the historical business model dataset, divide the business hours into time periods with a cycle of 15 minutes, adjust the passenger flow in different time periods based on the seasonal index, adjust the fluctuation range of operating income based on the trend coefficient, set the daily and weekly business peak hours based on the periodic cycle number, generate various business scenarios through the event-driven simulation module of AnyLogic, and the output simulated business scenario data includes time period numbers, estimated passenger flow, turnover, service demand, and service time distribution. When conducting business simulation on the initial scheduling plan based on the simulated business scenario, it is necessary to use a scheduling simulation program written in Java, import the simulated business scenario data and the initial scheduling plan data, use the Discrete Event Simulation (DES) method to simulate the personnel work situation in each time period, set the execution time of each task to 5 minutes, use a priority queue to manage the task execution order, call the thread pool ExecutorService to create 10 parallel working threads to respectively simulate the work processes of personnel in different positions, and record in real time the number of tasks completed, the number of personnel used, and the number of personnel idle in each time period. The finally generated simulated business data is output in the CSV file format, and the data includes time periods, position numbers, the number of tasks completed, the number of personnel used, and the number of personnel idle.

[0139] Preferably, step S5 includes the following steps:

[0140] Step S51: Perform time period slicing processing on the simulated business data to obtain business time period data;

[0141] Step S52: Identify the extreme values of personnel requirements for the business time period data based on the estimated PPMH data to obtain the extreme values of personnel requirements for the time period;

[0142] Step S53: Conduct an analysis of the rotation order constraints on the extreme values of personnel requirements for the time period to obtain personnel rotation constraints;

[0143] Step S54: Reconstruct the scheduling order of the initial scheduling plan based on the personnel rotation constraints to obtain the rotated scheduling order, and send the rotated scheduling order to the terminal to perform personnel scheduling management.

[0144] In this embodiment, when performing time slice processing on simulated business data, first import the simulated business data into the Pandas library in Python, convert the timestamp field of each piece of data into the Datetime format. Then use the resample() method of Pandas to divide the business data by day, week, or month. Set the time slice to 15 minutes, and the input parameter is '15T', indicating that each 15 minutes is a time period. Use the resample() method to resample the data according to this time period, and take the average values of fields such as turnover and passenger flow within each time period as the representative data of that time period. The output data is stored in the DataFrame format. The obtained business time period data includes information such as the start time, end time, average turnover, passenger flow, and personnel requirements of each time period. When identifying the extreme values of personnel requirements for the business time period data based on the estimated PPMH data, first combine the estimated PPMH data with the business time period data, and use the merge() function in Python to merge the two according to the timestamp field. The merged data contains the personnel requirements and PPMH index data for each time period. Then use the groupby() method of Pandas to group the data by time period, calculate the maximum and minimum values of the personnel requirements within each time period, and identify the extreme values of the personnel requirements within each time period. Set a threshold. If the personnel requirement is greater than this threshold, then this time period is considered a high-demand time period; if it is less than this threshold, then it is considered a low-demand time period. The identification results are saved in the CSV file format, including the time period number, maximum personnel requirement, minimum personnel requirement, and associated data of the PPMH index for each time period. When performing the rotation order constraint analysis on the extreme values of the time period personnel requirements, first classify the high-demand time periods and low-demand time periods according to the extreme value results of the personnel requirements, and use the shortest path algorithm (such as the Dijkstra algorithm) in graph theory to analyze the personnel rotation order. Set the nodes to represent each time period, and the edges to represent the personnel rotation constraints between two time periods. The specific constraint condition is: if the personnel requirement of a certain time period exceeds the set threshold, then the time period after this time period needs to be scheduled after a certain rotation period. The algorithm calculates the shortest path between time periods to ensure that the personnel rotation order conforms to the predetermined rules, and finally outputs the personnel rotation constraint data, including the time period number, corresponding rotation order, constraint conditions, and rotation period. When reconstructing the scheduling order of the initial scheduling plan based on the personnel rotation constraints, first import the initial scheduling plan into a scheduling optimization tool such as Gurobi, combine the rotation constraint data output in the previous step, set the scheduling optimization goal to minimize the interval time between personnel rotations, use an integer programming model, and the model constraints include the personnel rotation order, the shortest time interval between working time periods, the upper and lower limits of the number of people in each shift, etc. Use the Gurobi solver to perform multiple rounds of iterative optimization on the scheduling problem, and set the maximum number of iterations to 500,The final scheduled shift result includes the scheduled staff, job number, and working hours for each period. The optimized scheduling order is saved in the XML file format. Finally, the optimization result is sent to the terminal device through the API interface, and the terminal device performs personnel management execution according to the scheduling order.

[0145] The present invention also provides a personnel scheduling management system based on the PPMH index for executing the personnel scheduling management method based on the PPMH index as described above. The personnel scheduling management system based on the PPMH index includes:

[0146] A data collection module for collecting historical business data; estimating business density based on the historical business data to obtain business density data; performing working hour mapping on the business density data to generate complete working hour data;

[0147] A demand estimation module for estimating business demand based on the business density data to obtain an estimated product quantity; calibrating the demand for the complete working hour data according to the estimated product quantity to generate an efficiency demand standard;

[0148] A working hour inference module for performing working hour simulation inference on the estimated product quantity according to the efficiency demand standard to obtain the estimated basic working hours required; performing work efficiency measurement processing on the estimated product quantity and the estimated basic working hours required to generate estimated PPMH data;

[0149] A scheduling mapping module for performing personnel configuration mapping on the estimated product quantity based on the estimated PPMH data to obtain an initial scheduling plan; performing business simulation according to the initial scheduling plan to generate simulated business data;

[0150] A scheduling reconstruction module for reconstructing the rotation scheduling order of the initial scheduling plan according to the simulated business data to obtain the rotation scheduling order, and sending the rotation scheduling order to the terminal for performing personnel scheduling management.

[0151] The implementation of the data acquisition module in the present invention ensures the comprehensive acquisition of historical business data. The business density estimation provides a quantitative analysis of the changes in business demands. The generated business density data lays a scientific foundation for the working hours mapping. The generation of complete working hours data ensures the effective management of resource utilization. The demand prediction module achieves the accurate prediction of future product demands. The efficiency demand standard provides a clear guiding basis for personnel allocation. The application of the working hours inference module makes the estimation of basic working hours more scientific. The predicted PPMH data provides a quantitative reference for the shift scheduling decision. The application of the shift scheduling mapping module ensures the rationality and feasibility of the initial shift scheduling plan. The implementation of the business simulation verifies the effectiveness of the shift scheduling plan in actual operations. The shift scheduling reconstruction module improves the flexibility and adaptability of the shift scheduling. The final rotation shift scheduling order is sent to the terminal to execute the high efficiency of personnel management. The overall system improves the enterprise's management level and resource allocation efficiency, providing strong technical support for optimizing operations.

[0152] Therefore, in any aspect, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Thus, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be encompassed within the present invention.

[0153] The above description is only the specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A personnel scheduling management method based on PPMH index, characterized in that: The following steps are involved: Step S1: Collect historical business data; estimate business density based on the historical business data to obtain business density data; map the business density data to working hours to generate complete working hour data; Step S2: Based on the business density data, business demand is estimated to obtain the estimated product quantity; based on the estimated product quantity, the complete working hour data is calibrated to generate the efficiency demand standard; Step S3: Perform labor time simulation inference on the estimated product quantity according to the efficiency requirement standard to obtain the estimated required basic labor time; perform labor efficiency measurement processing on the estimated product quantity and the estimated required basic labor time to generate estimated PPMH data; Step S4: mapping the estimated product quantity to personnel allocation based on the estimated PPMH data to obtain an initial shift scheduling plan; performing business simulation according to the initial shift scheduling plan to generate simulated business data; Step S5: reconstruct the rotation scheduling order of the initial scheduling plan according to the simulated business data to obtain the rotation scheduling order, and send the rotation scheduling order to the terminal to perform personnel scheduling management.

2. The personnel scheduling management method based on PPMH index according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Collect historical business data; perform time series decomposition on the historical business data to obtain product time series; Step S12: categorize the product time series to obtain category distribution data; perform peak detection on the category distribution data to generate peak period feature data; Step S13: estimating the business density of the peak period characteristic data to obtain business density data; Step S14: Mapping the business density data to working hours to generate basic working hour data; performing auxiliary task calculation on the basic working hour data to obtain auxiliary working hour data; Step S15: integrating the basic working time data and the auxiliary working time data to generate complete working time data.

3. The personnel scheduling management method based on PPMH index according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: obtaining current business hour parameters; performing trend extrapolation processing on business density data to obtain business trend forecast data; Step S22: periodically adjusting the business trend forecast data based on the current business hour parameters to generate adjusted forecast business data; Step S23: Perform product production mapping on the adjusted forecast business data to obtain an estimated product quantity; Step S24: performing standard process expansion on the estimated product quantity based on the preset product process data to obtain process load data; Step S25: cross-map the process load data and the complete working time data to generate demand calibration data; and perform performance modeling on the demand calibration data to generate performance demand standards.

4. The personnel scheduling management method based on PPMH index according to claim 3 is characterized in that: Step S25 includes the following steps: Perform multi-dimensional spatial projection on process load data and complete working time data to obtain projection calibration data; perform fractal feature decomposition on the projection calibration data to generate fractal calibration data; Performing manifold surface fitting on the fractal calibration data to obtain manifold calibration data; performing redundancy compression processing on the manifold calibration data to generate required calibration data; Extract fluctuation characteristics of demand calibration data to obtain demand fluctuation characteristics; capture periodic changes of demand fluctuation characteristics to generate demand fluctuation period data; Based on the demand fluctuation cycle data, the demand calibration data is dynamically transformed into an efficiency tensor to generate an efficiency demand standard.

5. The personnel scheduling management method based on PPMH index according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: Mapping the estimated product quantity to the process to obtain the estimated required process; Step S32: deriving the process hours for the estimated required processes according to the efficiency requirement standard to generate the estimated required basic hours; Step S33: performing site distribution mapping on the estimated product quantity according to the preset workstation data to obtain site distribution data; Step S34: Calculate the required auxiliary man-hours for the estimated required basic man-hours according to the site distribution data to generate the estimated required auxiliary man-hours; Step S35: Integrate the estimated basic working hours and the estimated auxiliary working hours to obtain the estimated complete working hours; perform work efficiency measurement processing on the estimated product quantity based on the estimated complete working hours to generate estimated PPMH data.

6. The personnel scheduling management method based on PPMH index according to claim 5 is characterized in that: Step S34 includes the following steps: Step S341: performing site basic man-hour density statistics on the estimated required basic man-hours according to the site distribution data to obtain a site man-hour density collection; Step S342: determining the initial auxiliary man-hours according to the site man-hour density collection to generate the initial auxiliary man-hours; Step S343: performing collaborative work hours identification on the initial auxiliary work hours to obtain collaborative auxiliary work hours; performing site collaborative simulation on the site distribution data according to the collaborative auxiliary work hours to generate auxiliary work hours that can be omitted; Step S344: Calculate the initial auxiliary working hours and the auxiliary working hours that can be omitted to obtain the estimated auxiliary working hours required.

7. The personnel scheduling management method based on PPMH index according to claim 6 is characterized in that: Step S343 includes the following steps: Extract the working time type of the initial auxiliary working time to obtain the auxiliary working time type; perform site restriction analysis on the auxiliary working time type to generate site restriction data for each working time; Traverse each work time site restriction data, and when the work time site restriction data is the restricted site work time data, access the next work time site restriction data; when the work time site restriction data is the unrestricted site work time data, package the unrestricted site work time data to obtain collaborative auxiliary work time; Enumerate the neighboring sites of the site distribution data according to the collaborative auxiliary working hours to obtain the neighboring site data; digest the collaborative auxiliary working hours based on the neighboring site data to obtain the collaborative digested auxiliary working hours; Make a quantitative comparison of the auxiliary working hours that can be collaboratively digested and the auxiliary working hours that can be collaboratively digested to generate the auxiliary working hours that can be omitted.

8. The personnel scheduling management method based on PPMH index according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: Perform product-manpower matching mapping on the estimated PPMH data to generate the manpower required for each product; Step S42: Allocate personnel for the estimated number of products according to the manpower required for each product to obtain an initial shift scheduling plan; Step S43: extracting the business model from the historical business data to obtain the historical business model; simulating the business scene of the historical business model to obtain a simulated business scene; Step S44: Perform business simulation on the initial shift scheduling plan based on the simulated business site to generate simulated business data.

9. The personnel scheduling management method based on PPMH index according to claim 1 is characterized in that: Step S5 includes the following steps: Step S51: performing time period slicing processing on the simulated business data to obtain business period data; Step S52: identifying the extreme value of personnel demand for business period data based on the estimated PPMH data to obtain the extreme value of personnel demand for the period; Step S53: Performing rotation order constraint analysis on the extreme value of personnel demand in the time period to obtain personnel rotation constraints; Step S54: reconstructing the shift order of the initial shift scheduling plan based on the personnel rotation constraint to obtain the shift rotation order, and sending the shift rotation order to the terminal to perform personnel scheduling management.

10. A personnel scheduling management system based on PPMH index, characterized in that: Used to execute the personnel scheduling management method based on the PPMH index as claimed in claim 1, the personnel scheduling management system based on the PPMH index comprises: The data collection module is used to collect historical business data; estimate the business density based on the historical business data to obtain business density data; and map the business density data to working hours to generate complete working hour data; The demand estimation module is used to estimate business demand based on business density data and obtain the estimated product quantity; the demand is calibrated for the complete working time data according to the estimated product quantity to generate the efficiency demand standard; The man-hour inference module is used to simulate and infer the man-hours of the estimated product quantity according to the efficiency requirement standard to obtain the estimated basic man-hours required; the estimated product quantity and the estimated basic man-hours required are processed for work efficiency measurement to generate the estimated PPMH data; The shift mapping module is used to map the estimated product quantity to the staffing based on the estimated PPMH data to obtain an initial shift plan; perform business simulation based on the initial shift plan to generate simulated business data; The shift scheduling reconstruction module is used to reconstruct the rotation scheduling order of the initial shift scheduling plan according to the simulated business data, obtain the rotation scheduling order, and send the rotation scheduling order to the terminal to execute personnel scheduling management.

Images