An intelligent scheduling method and system for service industry part-time workers

Abstract

The application relates to the technical field of intelligent scheduling, and provides an intelligent scheduling method and system for time-sharing labor in the service industry, which comprises the following steps: obtaining historical comprehensive data of target employees, a fragmented shift set and manpower demand data of each service period; determining a minimum interval threshold according to the work intensity of a previous fragmented shift and the commuting time across stores, and determining a maximum interval threshold according to the income of the target employees; calculating an individualized interval time window corresponding to each target employee according to the historical comprehensive data, the minimum interval threshold and the maximum interval threshold; jointly scheduling the fragmented shift set by taking the interval time window as a main constraint and the manpower demand of each service period as an auxiliary constraint, obtaining a candidate scheduling scheme and outputting the same; and the application can dynamically optimize the interval time window between adjacent fragmented shifts, can maximize the coverage of manpower demand of each period while guaranteeing the rest and recovery of employees and the income bottom line, and can realize the collaborative optimization of employee experience and enterprise operation efficiency.

Description

Technical Field

[0001] This application relates to the field of intelligent scheduling, and in particular to an intelligent scheduling method and system for time-sharing employment in the service industry. Background Technology

[0002] In the service industry, especially in chain operations such as catering, retail, and hotels, labor demand exhibits significant tidal and fragmented characteristics. A single employee often needs to work multiple short, unconnected shifts within a day, and these shifts may be distributed across different stores within the same company. This fragmented employment model places unique technical demands on scheduling: the scheduling system not only needs to allocate suitable on-duty personnel to each independent shift, but also needs to rationally plan the transition periods between adjacent shifts for the same employee to ensure that the employee can complete the necessary rest and recovery and commuting between stores between shifts.

[0003] Existing scheduling technologies typically handle fragmented shifts using one of two methods: either setting a uniform fixed interval for all adjacent shifts, or completely ignoring transitional constraints between shifts and focusing solely on meeting the manpower needs of each individual shift as the sole optimization objective. Both methods have significant technical limitations in the context of fragmented employment in the service industry. Regarding the fixed interval method, its core problem lies in its inability to adapt to the dynamic changes in workload and commuting conditions between different shifts. Within the same scheduling cycle, the workload varies significantly between shifts. For example, the workload of shifts during peak lunch hours is much higher than that of off-peak make-up shifts, and the physiological recovery time required for employees after completing high-intensity shifts is significantly longer than that of light shifts. Furthermore, if consecutive shifts are located in different stores, the time spent commuting between stores must also be factored in. Fixed intervals cannot provide differentiated responses to the above factors: if the interval is set too short, employees on high-intensity shifts or those commuting between stores will not have enough recovery time, leading to fatigue and decreased service quality, and even operational safety accidents; if the interval is set too long, employees with lower workloads or those on consecutive shifts at the same store will experience excessively long idle waiting times. These waiting times are not counted in working hours and are not compensated, diluting employees' actual income per unit of time and dampening their work enthusiasm.

[0004] In the approach of completely ignoring transition constraints, scheduling algorithms only focus on whether the manpower needs of each time period are met, without imposing any restrictions on the time relationship between adjacent shifts of the same employee. While this approach increases the flexibility of scheduling solutions, it can lead to extreme situations in actual operation: an employee may be assigned to two shifts that are too close together, with insufficient intervals for necessary rest and recovery or cross-store commuting. Such scheduling results may be feasible from a manpower coverage perspective, but are unacceptable from the perspective of employee physiological safety and labor law compliance.

[0005] Therefore, this invention discloses an intelligent scheduling method and system for part-time employment in the service industry to solve the above-mentioned technical problems. Summary of the Invention

[0006] This application provides an intelligent scheduling method and system for time-sharing employment in the service industry. It solves the technical problem in the existing technology of how to dynamically optimize the interval time window between adjacent fragmented shifts without affecting the manpower required for each independent shift when jointly scheduling multiple fragmented shifts of the same employee within a day, so that the interval length can ensure that the employee can rest and recover, while avoiding employee idleness and income dilution.

[0007] To achieve the above objectives, this application adopts the following technical solution: Firstly, a smart scheduling method for time-sharing employment in the service industry is provided, including: S1: Obtain historical comprehensive data of the target employees, the fragmented shifts of the target employees within the target period, and the manpower demand data for each service period; among which, the historical comprehensive data includes the historical number of times the target employees arrived on time and the historical number of times they were absent; S2: Determine the minimum interval threshold based on the workload of the target employee's previous fragmented shifts and the commuting time between stores, in order to ensure that the target employee can rest and recover between adjacent fragmented shifts; S3: Determine the maximum interval threshold based on the target employee's income to avoid income dilution caused by idle waiting between adjacent fragmented shifts for the target employee; S4: Calculate the individualized interval time window for each target employee based on historical comprehensive data, the minimum interval threshold, and the maximum interval threshold; wherein, the interval time window is used to limit the interval duration range between adjacent fragmented shifts of the same target employee within the target period, and the interval duration range is [minimum interval threshold, maximum interval threshold]; S5: Using the interval time window as the primary constraint and the manpower demand of each service period as the secondary constraint, a joint scheduling solution is performed on the fragmented shift set to obtain candidate scheduling schemes; wherein, in the joint scheduling solution process, when the interval time window constraint conflicts with the manpower demand constraint of each service period, the interval time window constraint is given priority, and the manpower demand of each service period is allowed to have a preset deviation through slack variables, so as to cover the manpower demand of each service period to the greatest extent while satisfying the interval time window constraint. S6: Verify the candidate scheduling schemes to obtain the target scheduling scheme.

[0008] In conjunction with the first aspect mentioned above, one possible implementation involves acquiring the target employee's historical comprehensive data, the target employee's fragmented shifts within the target period, and the manpower demand data for each service period, including: Within the historical statistical period, the number of times the target employees were on time was recorded daily. Compared with historical absences The statistical criteria are as follows: clocking in before the start of the shift is counted as "on-time arrival", and clocking in after the start of the shift or not clocking in is counted as "absence". The historical statistical period is set manually and is generally 90 calendar days. Read all fragmented shifts of the target employee within the target period to form a fragmented shift set. For each fragmented shift in the set Record the following fields: shift identifier, store identifier, and service period start time. Service hours end time Work intensity value The target period is set manually, for example, the next 7 calendar days; the work intensity value... The value is obtained by normalizing the range of shift duration and peak service duration and then calculating it by equal weighting. The larger the value, the greater the physical exertion. The target period is divided into m service periods according to a preset time granularity; each service period is read. Minimum number of employees on duty This constitutes human resource demand data. .

[0009] In conjunction with the first aspect above, one possible implementation involves determining the minimum interval threshold based on the workload of the target employee's preceding fragmented shifts and cross-store commuting time, including: Aggregate fragmented shifts of employees with the same goal By service period start time Sort in ascending order; pair up two adjacent time-separated shifts; where the start time of the service period in the adjacent shift pair is... The former is recorded as the preceding fragmented shift, and the latter as the subsequent fragmented shift. If the target employee has only 0 or 1 fragmented shifts within the target period, there are no adjacent shift pairs, so this step is skipped directly, and the manually set standard low interval is used as the minimum interval threshold. Read the store identifiers of the preceding fragmented shifts and the subsequent fragmented shifts; if they are different, calculate the cross-store commuting time based on the map distance between the two stores and the preset average commuting speed. If the two are the same, let ; Read the workload value of the preceding fragmented shift from the adjacent shift pair. Based on the formula Determine the minimum interval threshold ;in, To preset the basic rest and recovery time, The workload conversion factor is set based on experience, and The range of values ​​for is (0,1]; Standard work intensity set manually.

[0010] In conjunction with the first aspect above, one possible implementation involves determining the maximum interval threshold based on the target employee's income, including: Read the expected income of target employees within the target period. If no expected income is set for the target employee, then the average daily income of the target employee during the historical statistical period will be used instead. The total historical income and total historical working hours of the target employee within the historical statistical period are retrieved from historical comprehensive data, and the unit time income is determined based on the total historical income and total working hours. Among them, income per unit time It is calculated by dividing the total historical revenue by the total historical working hours. Read the preset income dilution tolerance factor ;in, This represents the maximum permissible percentage of expected income lost due to idle waiting, with a value range of (0,1). Based on the calculation formula Calculate the highest interval threshold; where, The standard maximum interval threshold set manually; The duration within the target period, in hours; The duration is expressed in hours.

[0011] In conjunction with the first aspect mentioned above, one possible implementation involves calculating the individualized interval time window for each target employee based on historical aggregate data, the minimum interval threshold, and the maximum interval threshold, including: A1: Extract the target employee's historical on-time arrival count Compared with historical absences Based on computation Calculate the historical attendance reliability of the target employee , The larger the value, the more reliable the historical attendance record; A2: Based on calculation formula For the minimum interval threshold Make corrections; among them, This is the first correction factor set based on experience, with a value range of (0,1]. A3: Based on calculation formula Maximum Interval Threshold Make corrections; among them, This is a preset second correction coefficient, with a value range of (0,1]. A4: Determine the corrected minimum interval threshold Is it less than or equal to the corrected highest interval threshold? If true, proceed to A5; if false, decrease the value sequentially with a fixed step size. Re-execute A2 and A3 until... Established; A5: Will This serves as an individualized time window for the current target employees during their corresponding service periods.

[0012] In conjunction with the first aspect mentioned above, in one possible implementation, a joint scheduling solution is performed on the fragmented shift set, using the interval time window as the primary constraint and the manpower demand of each service period as a secondary constraint, to obtain candidate scheduling schemes, including: B1: Construct decision variables; where the decision variables are: extracting the target employee set as... For each fragmented shift With each target employee Define binary decision variables , Indicates fragmented schedules Assigned to target employees Simultaneously, a constraint is set: each fragmented shift can be assigned to at most one target employee, i.e. ; B2: Construct the master constraints; B3: Construct auxiliary constraints; B4: For each service period Introducing nonnegative relaxation variables Relax the constraint of B3 to ;in, The value represents the preset deviation of manpower demand allowed during this service period, with a maximum value of the preset value. ; B5: The objective function is set to minimize the sum of weighted slack variables for all service periods, i.e. ;in, Service hours The preset weights; in terms of solution priority, primary constraints are hard constraints with the highest priority, while secondary constraints are soft constraints, allowing for pass / fail. Deviation occurs; B6: Solve the models constructed in B1 to B5 using the integer linear programming branch and bound algorithm to obtain all decision variables. The possible values ​​of ; B7: Convert the decision variables obtained from the solution into structured scheduling records. Each record contains employee ID, shift ID, store ID, service period start time, service period end time, and interval between adjacent shifts. All records constitute candidate scheduling schemes.

[0013] In conjunction with the first aspect mentioned above, in one possible implementation, the master constraint is constructed, including: For target employees Any two adjacent fragmented train services , ,like and Then the interval between adjacent trains between the two Must meet ;in, For the preceding fragmented shifts, For subsequent fragmented service schedules; interval between adjacent service schedules The formula for calculation is: The principal constraint is a hard constraint, and no feasible solution may violate it.

[0014] In conjunction with the first aspect mentioned above, one possible implementation involves constructing auxiliary constraints, including: For each service period The total number of employees assigned to work during this period must be calculated and must not be less than the minimum manpower requirement for that period. ,Right now The summation range is the service period. Coverage of fragmented shifts and target employees.

[0015] In conjunction with the first aspect above, in one possible implementation, verifying the candidate scheduling scheme to obtain the target scheduling scheme includes: Perform integrity checks on candidate scheduling schemes; the integrity checks include: checking one by one whether there are duplicate schedulings for the same target employee in the same service period; checking one by one whether the interval between adjacent fragmented shifts for each target employee falls within its individualized interval time window; If the verification fails, return to B6 and solve again; The verified candidate scheduling schemes are converted into target scheduling tables according to the hierarchical structure of "target employee - date - service period - store", and the interval time window and actual interval duration of each target employee are noted. The target scheduling table is pushed to the scheduling terminal and stored in the scheduling management system, and then issued for execution as the target scheduling plan.

[0016] Secondly, an intelligent scheduling system for time-sharing employment in the service industry is provided, comprising: a communication unit and a processing unit; the communication unit is used to acquire historical comprehensive data of target employees, fragmented shifts of target employees within a target period, and manpower demand data for each service period; wherein, the historical comprehensive data includes the historical number of times the target employee arrived on time and the historical number of times he was absent. The processing unit is used to determine the minimum interval threshold based on the workload of the target employee's preceding fragmented shifts and the commuting time between stores, and to determine the maximum interval threshold based on the target employee's income; to calculate the individualized interval time window corresponding to each target employee based on historical comprehensive data, the minimum interval threshold, and the maximum interval threshold; to perform joint scheduling on the fragmented shift set with the interval time window as the primary constraint and the manpower demand of each service period as the secondary constraint, to obtain candidate scheduling schemes; and to verify the candidate scheduling schemes to obtain the target scheduling scheme.

[0017] This application provides an intelligent scheduling method and system for time-sharing employment in the service industry, with the following advantages: 1. This application provides an intelligent scheduling method and system for time-sharing employment in the service industry. By constructing an individualized interval time window jointly defined by a minimum interval threshold and a maximum interval threshold, the decision granularity of shift scheduling is improved from independent allocation of single shifts to overall collaborative optimization of the fragmented shift sequence of the same employee throughout the day. In determining the minimum interval threshold, this application uses the workload of the target employee's preceding fragmented shifts and cross-store commuting time as input variables. A workload conversion coefficient is used to convert shift duration and peak load into an equivalent incremental rest demand. Simultaneously, commuting time calculation or reset is automatically triggered based on the difference in store identification between preceding and following shifts, achieving differentiated responses to different workloads and commuting conditions. This mechanism ensures that employees with high-intensity, long commutes automatically receive more sufficient recovery time between adjacent shifts, matching their physical exertion, reducing service quality degradation and operational safety hazards caused by fatigue accumulation, and solving the problem of insufficient recovery or resource waste caused by the inability to adjust recovery time as needed under fixed-interval methods. Regarding the determination of the maximum interval threshold, this application uses the target employee's unit-time income and expected income as constraint variables, introduces an income dilution tolerance coefficient to quantify the employee's upper limit of income loss due to idle waiting, and transforms economic factors into time-dimensional constraints through formulaic expression. This applies a reverse constraint on the maximum idle time between adjacent shifts, preventing the dilution of employees' effective working hours due to excessively long shift intervals and ensuring employees' unit-time economic benefits. Through the aforementioned multi-dimensional and multi-parameter dynamic calculation and correction mechanism, this application generates an individualized interval time window for each target employee that matches their work intensity, commuting conditions, income expectations, and attendance records, achieving differentiated and precise scheduling and significantly improving the fit between the scheduling plan and the individual employee's situation.

[0018] 2. In the joint scheduling process, this application constructs a two-tiered priority system with the interval time window as the primary constraint and the service period manpower demand as the secondary constraint. By introducing slack variables and a weighted objective function, it achieves a structured and flexible solution under constraint conflict conditions. The interval time window primary constraint is set as a hard constraint, which must not be violated by any feasible solution, ensuring that the scheduling scheme rigidly satisfies the bottom-line constraints of employee rest and recovery and income protection. The service period manpower demand secondary constraint is set as a soft constraint. By assigning preset weights to non-negative slack variables for each service period and constructing an optimization objective that minimizes the sum of weighted slack variables, the solver prioritizes the complete execution of the primary constraint when it cannot simultaneously satisfy both the primary and secondary constraints. Simultaneously, it allows for deviations in manpower demand within a preset range in a controllable and quantifiable manner. This priority system avoids the problem of missing feasible solutions caused by excessively strong constraints in traditional all-hard constraint methods from a mathematical modeling perspective, enabling the scheduling system to maintain maximum coverage of manpower demand for each time period while prioritizing employee protection. In terms of implementation, this application employs an integer linear programming branch and bound algorithm to solve the aforementioned model. By constructing binary decision variables, it achieves precise matching constraints between fragmented shifts and target employees, and utilizes a branch and bound search strategy to converge to a feasible solution within a finite time. After solving, this application also performs a two-layer verification of candidate scheduling schemes through an integrity check: checking for duplicate scheduling for each employee and verifying whether the interval between adjacent shifts falls within their individualized interval time window. If the verification fails, a re-solution is triggered. The verified scheme is converted into an executable target scheduling table according to a hierarchical structure, with annotations for each employee's interval time window and actual interval time, providing scheduling managers with complete decision-making basis and traceability information. Overall, the primary and secondary constraint priority system and slack variable mechanism of this application establish a structured balance framework between rigid guarantees and flexible adaptation, effectively improving the success rate and computational efficiency of scheduling solutions. While ensuring a minimum level of employee experience, it maximizes coverage of manpower needs across different time periods, achieving synergistic improvement in employee satisfaction and enterprise operational efficiency.

[0019] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0020] Figure 1 A schematic diagram illustrating the steps of an intelligent scheduling method for time-sharing employment in the service industry, provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating the steps for obtaining a candidate scheduling scheme provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an intelligent scheduling system for time-sharing employment in the service industry, provided as an embodiment of this application. Detailed Implementation

[0021] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0022] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0023] like Figure 1 As shown in the embodiment of this application, an intelligent scheduling method for time-sharing employment in the service industry includes: S1: Obtain historical comprehensive data of the target employees, the fragmented shifts of the target employees within the target period, and the manpower demand data for each service period; among which, the historical comprehensive data includes the historical number of times the target employees arrived on time and the historical number of times they were absent; S2: Determine the minimum interval threshold based on the workload of the target employee's previous fragmented shifts and the commuting time between stores, in order to ensure that the target employee can rest and recover between adjacent fragmented shifts; S3: Determine the maximum interval threshold based on the target employee's income to avoid income dilution caused by idle waiting between adjacent fragmented shifts for the target employee; S4: Calculate the individualized interval time window for each target employee based on historical comprehensive data, the minimum interval threshold, and the maximum interval threshold; wherein, the interval time window is used to limit the interval duration range between adjacent fragmented shifts of the same target employee within the target period, and the interval duration range is [minimum interval threshold, maximum interval threshold]; S5: Using the interval time window as the primary constraint and the manpower demand of each service period as the secondary constraint, a joint scheduling solution is performed on the fragmented shift set to obtain candidate scheduling schemes; wherein, in the joint scheduling solution process, when the interval time window constraint conflicts with the manpower demand constraint of each service period, the interval time window constraint is given priority, and the manpower demand of each service period is allowed to have a preset deviation through slack variables, so as to cover the manpower demand of each service period to the greatest extent while satisfying the interval time window constraint. S6: Verify the candidate scheduling schemes to obtain the target scheduling scheme.

[0024] It is worth noting that this invention addresses the long-standing technical pain point in the fragmented employment scenario of the service industry, where the same employee is split into multiple unconnected short shifts within a day, and existing scheduling technologies generally adopt fixed intervals or completely ignore excessive constraints between shifts, leading to accumulated employee fatigue or excessively long idle time. This invention proposes a dual-threshold joint scheduling mechanism with individualized interval time windows as its core. First, this invention elevates the granularity of scheduling from "independent arrangement of single shifts" to "overall coordination of fragmented shift sequences for the same employee throughout the day." By constructing an interval time window jointly defined by a minimum and a maximum interval threshold, it simultaneously resolves the inherent contradiction between rest and recovery and income dilution within the same technical framework for the first time. The minimum interval threshold is dynamically calculated based on the workload of the employee's previous shift and commuting time across stores, automatically reserving more recovery time for employees with high-intensity, long commutes, thus fundamentally reducing the decline in service quality and safety hazards caused by fatigue. The maximum interval threshold, on the other hand, conversely constrains the upper limit of idle time based on the employee's income level, preventing excessive dilution of income per unit time due to excessive waiting time between shifts. This achieves a dynamic balance between the sustainability of employee physical strength and economic benefits, taking into account both the protection of workers' rights and the control of enterprise labor costs. Secondly, this invention incorporates individualized historical comprehensive data such as the number of times employees arrive on time and the number of times they are absent. This makes the interval time window no longer a uniform fixed value, but rather a differentiated customization based on each employee's reliability, stability, and physiological recovery ability, achieving true "one person, one window" precise scheduling. This mechanism not only improves the adaptability of the scheduling plan to the individual status of employees, but also guides employees to improve their attendance discipline through positive incentive effects. In the long run, it can significantly reduce the absenteeism rate, lateness rate, and employee turnover rate, stabilizing the basic foundation of the company's human resource supply. Furthermore, this invention constructs a priority system in the joint scheduling solution, with "interval time window as the main constraint and service period manpower demand as the secondary constraint." When the two conflict, priority is given to ensuring employees' rest and minimum income. At the same time, by using slack variables, preset deviations in manpower demand are allowed, so that the solution process has both the certainty of rigid constraints and the adaptive ability of flexible compromise. This design effectively avoids the problem of missing feasible solutions caused by excessive constraints in mathematical modeling, and significantly improves the success rate and computational efficiency of scheduling solutions in engineering implementation. It can cover manpower demand in all time periods to the greatest extent while ensuring employee experience, and achieve Pareto improvement in employee satisfaction and enterprise operational efficiency.

[0025] In one possible implementation of this application embodiment, the above-mentioned S1 can be implemented by the following S101, which will be described in detail below: S101: Obtain historical comprehensive data of the target employees, the fragmented shifts of the target employees within the target period, and the manpower demand data for each service period, including: Within the historical statistical period, the number of times the target employees were on time was recorded daily. Compared with historical absences Specifically, the statistical criteria are as follows: clocking in before the start of the daily shift is counted as "on-time attendance," while clocking in after the start of the shift or not clocking in is counted as "absence." The historical statistical period can be manually set, generally set to 90 calendar days, but can be adjusted to 60 or 120 calendar days depending on the company's actual work schedule. The clock-in time is determined based on the clock-in timestamp recorded by the attendance management system, accurate to the minute. For example, if an employee has 85 shift records within the historical statistical period, of which 78 days were clocked in before the start of the shift, 5 days were clocked in after the start of the shift, and 2 days were not clocked in, then this employee's historical on-time attendance count is calculated as follows. The number of historical absences is 78. The value is 7. This binary statistical approach, based on clock-in time, can objectively reflect employees' time compliance habits and provide a consistent data foundation for subsequent calculations of attendance reliability.

[0026] Read all fragmented shifts of the target employee within the target period to form a fragmented shift set. For each fragmented shift in the set Record the following fields: shift identifier, store identifier, and service period start time. Service hours end time Work intensity value The target period is set manually, for example, the next 7 calendar days; the work intensity value... The result is calculated by normalizing the range of shift duration and peak service duration, then weighting the values ​​with equal weights. A larger value indicates greater physical exertion. It should be noted that the range normalization specifically refers to the shift duration normalization value: BD i =(D i Dmin) / (Dmax Dmin), D i The current target employee's shift duration, where Dmax is the maximum shift duration among the target employees, and Dmin is the minimum shift duration among the target employees; normalized peak service duration: BP i =(P i Pmin) / (Pmax Pmin), P i The peak service duration for the current target employees is given by Pmax, where Pmax is the maximum peak service duration among the target employees, and Pmin is the minimum peak service duration among the target employees. The equal weights are the normalized shift duration values ​​BD. iNormalized value of peak service duration (BP) i The corresponding weights are all 0.5; the normalized value of shift duration BD i The cumulative dimension reflecting fatigue (the longer the total on-duty hours, the greater the cumulative physical exertion), and the normalized value of peak service hours (BP). i The instantaneous load dimension reflecting fatigue (the higher the service density during peak hours, the greater the physical exertion per unit time) and the physical exertion mechanism of the two dimensions are similar, and there is no prior evidence to suggest that one dimension contributes more systematically. The above equal weighting is a preferred implementation method. When the enterprise has accumulated enough historical attendance and churn data, the weights can be replaced with differentiated weights determined by the entropy weight method or the analytic hierarchy process. This invention is not limited to this. The target period is divided into m service periods according to a preset time granularity; each service period is read. Minimum number of employees on duty This constitutes human resource demand data. The time granularity can be set to 1 hour, or it can be set to 30 minutes or 2 hours according to the refined management needs of store operations. The minimum number of staff on duty for each service period, dt, is preset by the store operation management system based on factors such as historical customer flow data and promotional activities, reflecting the minimum manpower required to maintain normal service quality during that period.

[0027] In one possible implementation of this application embodiment, the above-mentioned S2 can be implemented by the following S201, which will be described in detail below: S201: Determine the minimum interval threshold based on the workload of the target employee's preceding fragmented shifts and cross-store commuting time, including: Aggregate fragmented shifts of employees with the same goal By service period start time Sort in ascending order; pair up two adjacent time-separated shifts; where the start time of the service period in the adjacent shift pair is... The former is recorded as the preceding fragmented shift, and the latter as the subsequent fragmented shift. If the target employee has only 0 or 1 fragmented shifts within the target period, there are no adjacent shift pairs, so this step is skipped directly, and the manually set standard low interval is used as the minimum interval threshold. The standard low interval can be obtained manually, for example, a value of 30 minutes, or it can be set to 20 minutes or 45 minutes according to industry practice.

[0028] Read the store identifiers of the preceding fragmented shifts and the subsequent fragmented shifts; if they are different, calculate the cross-store commuting time based on the map distance between the two stores and the preset average commuting speed. If the two are the same, let Map distances can be obtained by calling third-party map service interfaces, such as actual walking or cycling distances based on road networks, rather than the straight-line distance between two points, to ensure that commuting time estimates are closer to reality. Preset average commuting speeds can be set according to city traffic conditions and employees' usual commuting methods; for example, a walking speed of 5 km / h, a cycling speed of 12 km / h, and a public transportation speed of 20 km / h. For instance, if the previous shift is at store A and the subsequent shift is at store B, and the walking distance between the two stores is 1.5 km, with a preset walking speed of 5 km / h, then... minute.

[0029] Read the workload value of the preceding fragmented shift from the adjacent shift pair. Based on the formula Determine the minimum interval threshold ;in, To preset the basic rest and recovery time, The workload conversion factor is set based on experience, and The value range is (0,1], which is used to convert the degree of deviation of work intensity from the standard value into the increase or decrease of rest time; Standard work intensity set manually.

[0030] In one possible implementation of this application embodiment, the above-mentioned S3 can be implemented by the following S301, which will be described in detail below: S301: Determine the maximum interval threshold based on the target employee's income, including: Read the expected income of target employees within the target period. If a target employee has not set an expected income, their average daily income during the historical statistical period will be used as a substitute. Expected income can be set by the employee themselves through the scheduling terminal, reflecting their desired income level within the target period. If the employee does not actively set an expected income, the system automatically retrieves the total income during the historical statistical period, divides it by the number of days in that period, and uses the resulting average daily income as a substitute value to ensure that each employee has a reasonable income benchmark. The total historical income and total historical working hours of the target employee within the historical statistical period are retrieved from historical comprehensive data, and the unit time income is determined based on the total historical income and total working hours. Among them, income per unit time It is calculated by dividing the total historical revenue by the total historical working hours. Read the preset income dilution tolerance factor ;in, This represents the maximum permissible percentage of expected income lost due to idle waiting time, with a value ranging from (0,1); income dilution tolerance coefficient. This reflects employees' level of acceptance of downtime between shifts. The smaller the value, the more sensitive employees are to revenue dilution, and the system will compress the maximum interval threshold accordingly. The value can be set differently based on the company's employment strategy and the characteristics of the employee group. For example, for an employee group whose main source of income is part-time work, It can be set to 0.3 to 0.5; for employees who rely on part-time work for supplementary income, It can be set to 0.6 to 0.8. The verification process of the 0.6 minimum reduction coefficient is as follows: Based on the scheduling execution data of the enterprise for 90 natural days in history, select the set of shifts with a reduction item lower than 0.6, and count the decrease of the scheduling execution rate (actual attendance times / number of shifts) relative to the benchmark (the set of shifts with a reduction item ≥ 0.8). A total of XXXX samples were collected, with a confidence interval of 95%. The results show that the decrease in execution rate exceeds 15% - this decrease threshold is preset by the enterprise's operation and management goals, and below this threshold is considered that the scheduling quality is acceptable.

[0031] Based on the calculation formula Calculate the highest interval threshold; where, The maximum standard interval threshold is set manually by operators based on the service industry's time-sharing work scheduling standards and labor management practices, with a value range of [2,6] hours, preferably 4 hours; The duration within the target period, in hours; The duration of a unit of time, in hours; The setting serves as a lower limit protection mechanism, preventing the maximum interval threshold from being excessively compressed when the expected income is far lower than the benchmark income, and ensuring that employees have at least a certain buffer time between adjacent shifts. Statistical verification of historical shift data shows that when the reduction item is less than 0.6, employees' adjacent shifts are too close together and they do not get enough rest. The shift execution rate within the target period decreases by more than the preset proportion relative to the benchmark. Therefore, 0.6 is used as the minimum reduction coefficient.

[0032] In one possible implementation of this application embodiment, the above-mentioned S4 can be implemented by the following S401, which will be described in detail below: S401: Calculate the individualized interval time window for each target employee based on historical aggregate data, the minimum interval threshold, and the maximum interval threshold, including: A1: Extract the target employee's historical on-time arrival count Compared with historical absences Based on computation Calculate the historical attendance reliability of the target employee , The larger the value, the more reliable the historical attendance record; A2: Based on calculation formula For the minimum interval threshold Make corrections; among them, This is the first correction factor set based on experience, with a value range of (0,1]. A3: Based on calculation formula Maximum Interval Threshold Make corrections; among them, This is a preset second correction coefficient, with a value range of (0,1]. First correction coefficient Second correction coefficient The value of is related to historical attendance reliability. Related to the distribution characteristics: for Employees with a score below 0.7 Take a value of 0.3~0.5. Take a value of 0.2~0.4; for Employees with a value higher than 0.9 Take a value of 0.5 to 0.8. The value is set between 0.4 and 0.6; the specific value is determined by optimizing historical scheduling data using a grid search method, with the goal of minimizing the total manpower shortage in the scheduling plan. The fixed step size is set to 0.05. At that time, decrease synchronously in steps of 0.05. and until If the condition is still not met after synchronous reduction, then fix it. Remain unchanged, then decrease individually in steps of 0.05. Continue until the desired result is achieved. If the value has been reduced to 0.05 and still cannot be satisfied, then... and All values ​​are set to 0.05 to preserve the correction effect to the greatest extent possible while ensuring the time window is effective.

[0033] A4: Determine the corrected minimum interval threshold Is it less than or equal to the corrected highest interval threshold? If true, proceed to A5; if false, decrease the value sequentially with a fixed step size. Re-execute A2 and A3 until... Established; A5: Will This serves as an individualized time window for the current target employees during their corresponding service periods.

[0034] The following numerical example illustrates the attendance reliability correction and iteration mechanism. Continuing the previous example, suppose an employee's... minute, minute, , , .but minute; Minutes. At this time. If the condition is met, the individualized interval time window for this employee is [60.0 minutes, 139.9 minutes].

[0035] Let's assume another employee with lower attendance reliability, whose , minute, minute, , .but minute; Minutes. At this time. , the condition is established.

[0036] The engineering significance of iterative correction in A4 lies in the fact that when extremely low attendance reliability causes the corrected minimum interval threshold to exceed the maximum interval threshold, the interval time window becomes "inverted," failing to form an effective constraint interval. In this case, by gradually reducing the correction coefficients β and γ, a compromise is made on the strength of the attendance reliability correction, prioritizing the effectiveness of the interval time window, ensuring that a feasible interval constraint space always exists in the scheduling solution. This adaptive iterative mechanism avoids the entire scheduling model becoming unsolvable due to extreme data from individual employees, demonstrating the system's dynamic balance between individual constraints and global feasibility.

[0037] In one possible implementation of this application embodiment, the above-mentioned S5 can be implemented by the following S501, S502 and S503, which are described in detail below: like Figure 2 As shown, S501: Using the interval time window as the primary constraint and the manpower demand of each service period as a secondary constraint, a joint scheduling solution is performed on the fragmented shift set to obtain candidate scheduling schemes, including: B1: Construct decision variables; where the decision variables are: extracting the target employee set as... For each fragmented shift With each target employee Define binary decision variables , Indicates fragmented schedules Assigned to target employees Simultaneously, a constraint is set: each fragmented shift can be assigned to at most one target employee, i.e. ; B2: Construct the master constraints; B3: Construct auxiliary constraints; B4: For each service period Introducing nonnegative relaxation variables Relax the constraint of B3 to ;in, The value represents the preset deviation of manpower demand allowed during this service period, with a maximum value of the preset value. ; B5: The objective function is set to minimize the sum of weighted slack variables for all service periods, i.e. ;in, Service hours The preset weights; in terms of solution priority, primary constraints are hard constraints with the highest priority, while secondary constraints are soft constraints, allowing for pass / fail. Deviation occurs; B6: Solve the models constructed in B1 to B5 using the integer linear programming branch and bound algorithm to obtain all decision variables. The possible values ​​of ; B7: Convert the decision variables obtained from the solution into structured scheduling records. Each record contains employee ID, shift ID, store ID, service period start time, service period end time, and interval between adjacent shifts. All records constitute candidate scheduling schemes.

[0038] S502: Construct the main constraints, including; For target employees Any two adjacent fragmented train services , ,like and Then the interval between adjacent trains between the two Must meet ;in, For the preceding fragmented shifts, For subsequent fragmented service schedules; interval between adjacent service schedules The formula for calculation is: The principal constraint is a hard constraint, and no feasible solution may violate it.

[0039] S503: Construct secondary constraints, including: For each service period The total number of employees assigned to work during this period must be calculated and must not be less than the minimum manpower requirement for that period. ,Right now The summation range is the service period. Coverage of fragmented shifts and target employees.

[0040] In one possible implementation of this application embodiment, the above-mentioned S6 can be implemented by the following S601, which will be described in detail below: S601: Verify the candidate scheduling schemes to obtain the target scheduling scheme, including: Perform integrity checks on candidate scheduling schemes; the integrity checks include: checking one by one whether there are duplicate schedulings for the same target employee in the same service period; checking one by one whether the interval between adjacent fragmented shifts for each target employee falls within its individualized interval time window; If the verification fails, return to B6 and solve again; The verified candidate scheduling schemes are converted into target scheduling tables according to the hierarchical structure of "target employee - date - service period - store", and the interval time window and actual interval duration of each target employee are noted. The target scheduling table is pushed to the scheduling terminal and stored in the scheduling management system, and then issued for execution as the target scheduling plan.

[0041] The above primarily describes the solutions of the embodiments of this application from the perspective of device implementation. It is understood that each device, for example, an intelligent scheduling system for time-sharing employment in the service industry, includes at least one of the hardware structures and software modules corresponding to the execution of each function in order to achieve the above-mentioned functions. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software-driven hardware manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0042] This application embodiment can divide a service industry time-sharing intelligent scheduling system into functional units based on the above method example. For example, each function can be divided into separate functional units, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0043] When using integrated units, Figure 3 The above embodiment shows a possible structural diagram of an intelligent scheduling system for time-sharing employment in the service industry (denoted as communication system 30). The communication system 30 includes a processing unit 301 and a communication unit 302, and may also include a storage medium (denoted as storage unit 303). Figure 3 The structural diagram shown can be used to illustrate the structure of an intelligent scheduling system for time-sharing employment in the service industry involved in the above embodiments.

[0044] when Figure 3The schematic diagram shown illustrates the structure of an intelligent scheduling system for time-sharing employment in the service industry involved in the above embodiments. The processing unit 301 is used to control and manage the actions of the intelligent scheduling system for time-sharing employment in the service industry, the communication unit 302 is used for the intelligent scheduling system for time-sharing employment in the service industry to communicate with other devices, and the storage unit 303 is used to store the program code and data of the intelligent scheduling system for time-sharing employment in the service industry.

[0045] For example, communication unit 302 is used to acquire historical comprehensive data of the target employee, the fragmented shift set of the target employee within the target period, and the manpower demand data for each service period; among which, the historical comprehensive data includes the target employee's historical on-time attendance and historical absences. Processing unit 301: is used to determine the minimum interval threshold based on the workload of the target employee's preceding fragmented shifts and the commuting time between stores, and to determine the maximum interval threshold based on the target employee's income; calculate the individualized interval time window corresponding to each target employee based on historical comprehensive data, the minimum interval threshold, and the maximum interval threshold; perform joint scheduling solution on the fragmented shift set with the interval time window as the main constraint and the manpower demand of each service period as the secondary constraint to obtain candidate scheduling schemes; and verify the candidate scheduling schemes to obtain the target scheduling scheme.

[0046] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0047] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely illustrative descriptions of the application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

[0048] Some of the data in the above calculation formula are obtained by removing dimensions and taking their numerical values. The calculation formula is a calculation formula that is closest to the real situation, obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the calculation formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

Claims

1. An intelligent scheduling method for service industry time-sharing labor, characterized in that, include: S1: Obtain historical comprehensive data of the target employees, the fragmented shifts of the target employees within the target period, and the manpower demand data for each service period; among which, the historical comprehensive data includes the historical number of times the target employees arrived on time and the historical number of times they were absent; S2: Determine the minimum interval threshold based on the workload of the target employee's previous fragmented shifts and cross-store commuting time; S3: Determine the maximum interval threshold based on the target employee's income; S4: Calculate the individualized interval time window for each target employee based on historical comprehensive data, the minimum interval threshold, and the maximum interval threshold; wherein, the interval time window is used to limit the interval duration range between adjacent fragmented shifts of the same target employee within the target period, and the interval duration range is [minimum interval threshold, maximum interval threshold]; S5: Using the interval time window as the primary constraint and the manpower demand of each service period as the secondary constraint, perform a joint scheduling solution on the fragmented shift set to obtain candidate scheduling schemes. S6: Verify the candidate scheduling schemes to obtain the target scheduling scheme. 2.The intelligent scheduling method for time-sharing labor in service industry according to claim 1, characterized in that, The acquisition of historical comprehensive data of the target employees, the fragmented shifts of the target employees within the target period, and the manpower demand data for each service period includes: The historical punctual arrival times of the target employee are counted day by day in a historical statistical period and the historical absence times ; wherein the statistical range is that the check-in completed before the starting time of the current shift is recorded as "punctual arrival", and the check-in completed after the starting time of the current shift or the check-in not completed is recorded as "absence". Reading all the fragmented shifts of the target employee in the target period, forming a fragmented shift set ; for each fragmented shift in the set Record the following fields: shift identifier, store identifier, service period start time , service period end time , work intensity value ; wherein the work intensity value is calculated by equal weight weighting after normalization of shift length and service peak length range dividing the target period into m service periods according to a preset time granularity; reading the minimum on-site personnel lower limit of each service period , , . 3.The intelligent scheduling method for time-sharing labor in service industry according to claim 1, characterized in that, The determination of the minimum interval threshold based on the workload of the target employee's previous fragmented shifts and cross-store commuting time includes: Fragmented shift sets for the same target employee are collected By service period start time In ascending order; form adjacent shift pairs from two fragmented shifts that are adjacent in time; wherein the service period start time The former is denoted as the preceding fragmented shift, and the latter is denoted as the succeeding fragmented shift; if the target employee has only 0 or 1 fragmented shifts in the target period, there is no adjacent shift pair, and this step is directly skipped, and the standard low interval is adopted as the minimum interval threshold. Read the store identifiers of the preceding fragmented shifts and the subsequent fragmented shifts; if they are different, calculate the cross-store commuting time based on the map distance between the two stores and the preset average commuting speed. If the two are the same, let ; Read the workload value of the preceding fragmented shift from the adjacent shift pair. Based on the formula Determine the minimum interval threshold ;in, To preset the basic rest and recovery time, The workload conversion factor is set based on experience, and The range of values ​​for is (0,1]; This is the standard working intensity.

4. The intelligent scheduling method for time-sharing employment in the service industry according to claim 1, characterized in that, The determination of the maximum interval threshold based on the target employee's income includes: Read the expected income of target employees within the target period. If no expected income is set for the target employee, then the average daily income of the target employee during the historical statistical period will be used instead. The total historical income and total historical working hours of the target employee within the historical statistical period are retrieved from historical comprehensive data, and the unit time income is determined based on the total historical income and total working hours. Among them, income per unit time It is calculated by dividing the total historical revenue by the total historical working hours. Read the preset income dilution tolerance factor ;in, This represents the maximum permissible percentage of expected income lost due to idle waiting, with a value range of (0,1). Based on the calculation formula Calculate the highest interval threshold; where, The standard maximum interval threshold; The duration within the target period, in hours; The duration is expressed in hours.

5. The intelligent scheduling method for time-sharing employment in the service industry according to claim 1, characterized in that, The calculation of the individualized interval time window for each target employee based on historical aggregate data, the minimum interval threshold, and the maximum interval threshold includes: A1: Extract the target employee's historical on-time arrival count Compared with historical absences Based on computation Calculate the historical attendance reliability of the target employee , The larger the value, the more reliable the historical attendance record; A2: Based on calculation formula For the minimum interval threshold Make corrections; among them, This is the first correction factor set based on experience, with a value range of (0,1]. A3: Based on calculation formula Maximum Interval Threshold Make corrections; among them, This is a preset second correction coefficient, with a value range of (0,1]. A4: Determine the corrected minimum interval threshold Is it less than or equal to the corrected highest interval threshold? If true, proceed to A5; if false, decrease the value sequentially with a fixed step size. or Re-execute A2 and A3 until... Established; A5: Will This serves as an individualized time window for the current target employees during their corresponding service periods.

6. The intelligent scheduling method for time-sharing employment in the service industry according to claim 1, characterized in that, The method of using the time interval window as the primary constraint and the manpower demand of each service period as a secondary constraint to jointly solve the fragmented shift set yields candidate shift schemes, including: B1: Construct decision variables; where the decision variables are: extracting the target employee set as... For each fragmented shift With each target employee Define binary decision variables , Indicates fragmented schedules Assigned to target employees Simultaneously, a constraint is set: each fragmented shift can be assigned to at most one target employee, i.e. ; B2: Construct the master constraints; B3: Construct auxiliary constraints; B4: For each service period Introducing nonnegative relaxation variables Relax the constraint of B3 to ;in, The value represents the preset deviation of manpower demand allowed during this service period, with a maximum value of the preset value. ; B5: The objective function is set to minimize the sum of weighted slack variables for all service periods, i.e. ;in, Service hours The preset weights; in terms of solution priority, primary constraints are hard constraints with the highest priority, while secondary constraints are soft constraints, allowing for pass / fail. Deviation occurs; B6: Solve the models constructed in B1 to B5 using the integer linear programming branch and bound algorithm to obtain all decision variables. The possible values ​​of ; B7: Convert the decision variables obtained from the solution into structured scheduling records. Each record contains employee ID, shift ID, store ID, service period start time, service period end time, and interval between adjacent shifts. All records constitute candidate scheduling schemes.

7. The intelligent scheduling method for time-sharing employment in the service industry according to claim 6, characterized in that, The construction of the main constraint includes: For target employees Any two adjacent fragmented train services , ,like and Then the interval between adjacent trains between the two Must meet ;in, For the preceding fragmented shifts, For subsequent fragmented service schedules; interval between adjacent service schedules The formula for calculation is: The principal constraint is a hard constraint, and no feasible solution may violate it.

8. The intelligent scheduling method for time-sharing employment in the service industry according to claim 7, characterized in that, The construction of auxiliary constraints includes: For each service period The total number of employees assigned to work during this period must be calculated and must not be less than the minimum manpower requirement for that period. ,Right now The summation range is the service period. Coverage of fragmented shifts and target employees.

9. The intelligent scheduling method for time-sharing employment in the service industry according to claim 1, characterized in that, The process of verifying the candidate scheduling schemes to obtain the target scheduling scheme includes: Perform integrity checks on candidate scheduling schemes; the integrity checks include: checking one by one whether there are duplicate schedulings for the same target employee in the same service period; checking one by one whether the interval between adjacent fragmented shifts for each target employee falls within its individualized interval time window; If the verification fails, return to B6 and solve again; The verified candidate scheduling schemes are converted into target scheduling tables according to the hierarchical structure of "target employee - date - service period - store", and the interval time window and actual interval duration of each target employee are noted. The target scheduling table is pushed to the scheduling terminal and stored in the scheduling management system, and then issued for execution as the target scheduling plan.

10. An intelligent scheduling system for time-sharing employment in the service industry, characterized in that, include: Communication unit and processing unit; The communication unit includes a communication interface for data interaction with the scheduling management system, attendance system and store management terminal to obtain historical comprehensive data of the target employee, the fragmented shift set of the target employee within the target period, and manpower demand data for each service period. The historical comprehensive data includes the target employee's historical on-time attendance count and historical absence count; the processing unit includes one or more processors and a non-volatile memory, the non-volatile memory storing a computer program, which, when executed by the one or more processors, implements the method as described in any one of claims 1 to 9.

Images