Task Scheduling Methods

Abstract

The present application discloses a task scheduling method, which includes: determining the task amount of multiple task time periods; determining a first correspondence between task skills and the number of task undertakers in multiple task time periods at least based on the task amount; determining a third correspondence between multiple task undertakers, multiple task skills and multiple task time periods at least based on the first correspondence and the second correspondence between task undertakers and task skills. By adopting this processing method, working hours can be reasonably arranged for hundreds or thousands of task undertakers, so that task requirements can be fully covered as much as possible, and flexible employment can be achieved to meet task requirements with large dynamic changes, and to achieve task scheduling processing at a fine time granularity level and task skill level; therefore, it can effectively take into account lower labor costs and higher staff satisfaction, that is, it can effectively improve the rationality of task scheduling and improve labor management performance.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a task scheduling method. Background Art

[0002] The service industry (such as customer service, retail enterprises, etc.) faces the pressure of increasing task volume on the one hand, and the pressure of rising labor costs on the other hand. To maintain reasonable profits and per capita task volume is a challenge that all service enterprises have to face. Therefore, the service industry is paying attention to improving the utilization rate of employees and the scientific nature of scheduling, thereby improving the efficiency of labor management.

[0003] With the development of big data technology and the rapid growth of data, more and more service industries are estimating the demand for tasks in the future based on big data. According to the needs of refined operations, the granularity of the estimate can be up to one hour or even half an hour to describe the dynamic fluctuations of demand. Take the catering industry as an example, especially fast food chains, which have a typical peak-valley demand pattern. The demand for staff during meal times is much higher than that during non-meal times. On the other hand, the catering industry has a rich division of positions, and employees' skills may cover multiple positions. With the rise in labor costs, flexible employment has become a trend in the catering industry. How to reduce labor costs and reasonably allocate tasks to employees to improve employee satisfaction and thus improve service quality is one of the core issues in the operation of such companies.

[0004] At present, the commonly used personnel scheduling method is the method based on shift allocation, that is, the time granularity of personnel and tasks is shift. The method based on shift allocation can be implemented by pattern repetition, historical use, rule engine, etc. The following briefly describes the technical characteristics and shortcomings of these commonly used implementation methods.

[0005] 1. Scheduling with repeated patterns. The characteristics of this method are: setting a simple model for repetition, such as: morning-morning-noon-noon-evening-evening-rest. The disadvantages of this method are: it requires manual settings and is difficult to support complex scenarios.

[0006] 2. The historical scheduling method. The characteristic of this method is: copying the schedule of the previous cycle. The disadvantage of this method is: it cannot support changing needs, and the workload of subsequent adjustments may be very large.

[0007] 3. The scheduling method of rule engine. The characteristics of this method are: using multiple rules to screen out suitable personnel in turn and assign tasks. The disadvantage of this method is: it is difficult to comprehensively consider all factors, which often leads to violations of some hard constraints.

[0008] However, in the process of realizing the present invention, the inventors found that the existing technical solutions have at least the following problems: on the one hand, in the business scenarios of refined operations in service industries such as restaurants, the time granularity of task demand is hourly, and the dynamic fluctuation range of task volume in different time periods is large; on the other hand, the existing shift-based dispatching method requires that the time granularity of both personnel and tasks is shift, and does not consider the task's demand for employee skills. This will result in the existing scheduling method being unable to adapt to the task requirements of fine time granularity, large fluctuations in task volume, and different tasks having different requirements for employee skills. Therefore, flexible employment cannot be achieved, thereby reducing the rationality of personnel scheduling. Summary of the invention

[0009] The present application provides a task scheduling method to solve the problem of low rationality of task scheduling in the prior art.

[0010] The present application provides a task scheduling method, comprising:

[0011] Determine the task volume for multiple task time periods;

[0012] At least according to the task amount, determining a first correspondence between task skills and the number of task undertakers in multiple task time periods;

[0013] A third correspondence between multiple task undertakers, multiple task skills and multiple task time periods is determined based at least on the first correspondence and the second correspondence between the task undertakers and the task skills.

[0014] Optionally, the third corresponding relationship is determined by the following steps:

[0015] According to the first corresponding relationship, a plurality of task sequences having the same task skill requirements, number of unit task undertakers and continuous task time periods are generated;

[0016] The third corresponding relationship is determined at least according to the task sequence and the second corresponding relationship.

[0017] Optionally, the task sequence is generated in the following manner:

[0018] According to the first corresponding relationship, a plurality of task sequences are generated that have the same task skill requirements, number of unit task undertakers, continuous task time periods, and sequence lengths that are less than a length threshold.

[0019] Optionally, the third corresponding relationship is determined in the following manner:

[0020] At least based on the task sequence, the first corresponding relationship and the second corresponding relationship, construct a distribution model of the task undertaker to the task sequence;

[0021] Determining a fourth corresponding relationship between a plurality of task undertakers and a plurality of task sequences through the allocation model;

[0022] The third corresponding relationship is determined according to the fourth corresponding relationship.

[0023] Optionally, the construction of a distribution model of task undertakers to task sequences includes:

[0024] Get task scheduling rules;

[0025] An allocation model including constraint conditions corresponding to the task scheduling rules is constructed.

[0026] Optionally, the task scheduling rule includes at least one of the following rules:

[0027] Working days threshold, rest days threshold, upper and / or lower working hours threshold in monthly / weekly dimensions;

[0028] Upper limit of continuous working time;

[0029] The upper limit of continuous rest duration.

[0030] Optionally, the fourth corresponding relationship is determined by an optimization solver or a heuristic algorithm at least based on the first corresponding relationship and the second corresponding relationship.

[0031] Optionally, the time granularity of the task time period includes hours.

[0032] Optionally, the task volume includes the order volume;

[0033] The skills include: ordering service skills, serving service skills, cleaning service skills, cashier service skills, vegetable washing skills, vegetable cutting skills, and cooking skills.

[0034] Optionally, the task volume includes the user consultation volume for the business object;

[0035] The skills include: service skills corresponding to the business object category.

[0036] Optionally, the task volume includes sales data;

[0037] The skills include: shelf organizing skills, cleaning skills, cashiering skills, and goods shelving skills.

[0038] The present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and when the computer-readable storage medium is run on a computer, the computer executes the above-mentioned various methods.

[0039] The present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the above-mentioned various methods.

[0040] Compared with the prior art, this application has the following advantages:

[0041] The task scheduling method provided in the embodiment of the present application determines the task amount of multiple task time periods; determines a first correspondence between task skills and the number of task undertakers in multiple task time periods at least based on the task amount; determines a third correspondence between multiple task undertakers, multiple task skills and multiple task time periods at least based on the first correspondence and the second correspondence between task undertakers and task skills; this processing method can reasonably arrange working hours for hundreds or thousands of task undertakers, so that task requirements can be covered as completely as possible, and flexible employment can be achieved to meet task requirements with large dynamic changes, and realize task scheduling processing at fine time granularity level and task skill level; therefore, it can effectively take into account both lower labor costs and higher staff satisfaction, that is, it can effectively improve the rationality of task scheduling and improve labor management performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flowchart of an embodiment of a task scheduling method provided by the present application;

[0043] Figure 2 This is a skill switching schematic diagram of an embodiment of a task scheduling method provided by the present application;

[0044] Figure 3 It is a specific flow chart of an embodiment of a task scheduling method provided by the present application. DETAILED DESCRIPTION

[0045] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0046] First embodiment

[0047] Please refer to Figure 1 , which is a flow chart of an embodiment of a task scheduling method provided by the present application, wherein the execution subject of the method includes a personnel scheduling device. A task scheduling method provided by the present application includes:

[0048] Step S101: Determine the task amounts for multiple task time periods.

[0049] The task scheduling method provided in the embodiment of the present application can have a task time period of hourly level, and the scheduling granularity of the task undertaker can also be hourly level. This can adapt to business scenarios with refined operations and flexible employment needs such as fast food chains.

[0050] In one example, the business scenario is a chain fast food business scenario that performs refined operations on the order volume. In this case, the business hours of the chain fast food outlets are divided into multiple task time periods, such as one task time period per hour, and the task volume of each task time period includes the order volume, such as the number of orders from in-store customers and the total number of order orders from online users, etc.

[0051] In specific implementation, step S101 can be implemented in the following manner: according to historical ordering data and / or timeliness data (such as weather, season, promotional activities, etc.), determine the task volume forecast value of each task time period in each day within the target time period (such as the next 7 days, 1 month, etc.). For example, first, based on the user ordering log data of the last 3 months, the average order volume of each task time period is counted; then, combined with the weather conditions or seasons in the next 7 days and other factors, the order volume of each task time period in the next 7 days is predicted. For example, if the weather is hot in the next 7 days, the order volume of the stalls specializing in cold products will surge; if the weather turns cool in the next 7 days, the order volume of the restaurant operating Roujiamo will surge; or, if the next 7 days is a store anniversary discount period for a restaurant, the order volume in the next 7 days will surge, and so on. Table 1 shows the task volume forecast value of this embodiment.

[0052]

[0053]

[0054] Table 1. Task volume forecast table

[0055] As can be seen from Table 1, the time granularity of the task time period in this embodiment is 1 hour, so 1 day is divided into 24 task time periods, and the order volume in each time period is predicted respectively. The order volume in different time periods may fluctuate greatly. Since the refined operation of tasks, that is, predicting the task volume by task time period, is a relatively mature existing technology, it will not be repeated here.

[0056] In another example, the business scenario is a business scenario where an e-commerce customer service provides shopping consultation to users and finely manages the amount of user consultation. In this case, one consultation from a user is a task, and the task volume for each task time period includes the amount of user consultation.

[0057] In another example, the business scenario is a business scenario of finely operating the sales of supermarket goods. In this case, the task volume of each task time period includes sales.

[0058] After determining the task amounts in multiple task time periods, the next step may be to determine a first correspondence between task skills and the number of task undertakers in multiple task time periods at least based on the task amounts.

[0059] Step S103: Determine a first correspondence between task skills and the number of task undertakers in multiple task time periods at least based on the task volume.

[0060] In the scenario of refined operations, tasks can be broken down into various time granularities and the demand for task undertakers of various skills according to the business volume. For each time granularity of each skill in the future, there will be an estimated number of task undertakers. Task scheduling in this type of scenario is to assign task undertakers to tasks in each task time period. Unlike traditional shift-based task allocation, the demand for the number of task undertakers in different task time periods in the business scenarios of the embodiments of the present application is usually highly volatile, so there are conditions and needs for flexible employment.

[0061] The task skills refer to the work capabilities of the task undertaker required to complete a task. For example, in the fast food business scenario, the employee skills consumed by selling hamburgers include vegetable washing skills, vegetable cutting skills, cooking skills, selling skills, cleaning skills, etc. In the supermarket scenario, the employee skills consumed in each task time period include front-of-house shelf organization skills, cashier skills, product shelving skills, etc. In the customer service scenario, the employee skills consumed in each task time period include service skills for product category A, service skills for product category B, etc.

[0062] The method provided in the embodiment of the present application determines the number of task undertakers for each task skill according to the dependence of the task on different task skills. For example, in a fast food business scenario, the demand for personnel with vegetable washing skills is 2 people between 12 noon and 1 pm, the demand for personnel with vegetable cutting skills is 3 people, the demand for personnel with cooking skills is 2 people, the demand for personnel with selling skills is 5 people, the demand for personnel with cleaning skills is 3 people, and so on. In a supermarket scenario, the demand for personnel with shelf tidying skills in the front field is 3 people between 5 pm and 7 pm, the demand for personnel with cashier skills is 5 people, the demand for personnel with goods on the shelves is 6 people, and so on. In a customer service scenario, the demand for personnel with service skills for product category A is 10 people between 8 pm and 9 pm, the demand for personnel with service skills for product category B is 5 people, and so on.

[0063] In one example, the business scenario is a business scenario where an e-commerce customer service provides users with shopping consultations and fine-grained operations on the amount of user consultations. In this case, one consultation from a user is a task, and the time management granularity of the user consultation volume is hourly, such as half an hour, 1 hour or 2 hours, etc. The time granularity of the customer service staff's scheduling is also hourly, but it is not a fixed number of hours, but a flexible and variable number of hours to adapt to changes in the amount of tasks.

[0064] In this embodiment, the problem scenario is to schedule shifts for the staff of a certain restaurant. Its historical operating data can be used for analysis and mining to obtain an estimate of the operating conditions for a period of time in the future. Usually for digitally operated stores, it is relatively easy to obtain the historical transaction volume (such as by day dimension) or the historical order volume (such as by day x hour dimension) of the scenario. Combined with some common basic data, such as date, weather, activity and other indication information, basic statistical methods and machine learning methods can be used to predict the order volume in future unit time periods. Then, through a mapping relationship from order volume to job requirements, the number of people required for each position in each unit time period can be decomposed.

[0065] For example, in the catering industry, task requirements can be refined to the hour, and positions can be refined to the front office (waiters, lobby managers, cashiers, etc.) and the kitchen (vegetable washing, food preparation, chefs, etc.). By estimating the order volume per hour, the number of people required for each position and each time period can be determined. For example, from 12 noon to 1 pm, the order volume is 50, corresponding to 3 chefs, 2 vegetable washers, and 2 vegetable cutters. It should be noted that this solution can also be used in scenarios of refined operations such as new retail. For example, during the peak period from 5 pm to 7 pm in a supermarket, 5 shopping guides, 8 pickers, and 10 cashiers are required per unit time period.

[0066] Table 2 shows the predicted values ​​of the number of task undertakers with different skills in each task time period of this embodiment.

[0067]

[0068] Table 2. First correspondence table

[0069] As can be seen from Table 2, the rows are the skill dimension, and the columns are the time dimension. The value of each cell is the demand for the number of task undertakers corresponding to the skill dimension and time dimension. For example, the demand for task undertakers of skill 1 in time slot (task time period) 12 is 2, and the demand for task undertakers of skill N is 1; the demand for task undertakers of skill 1 in time slot (task time period) 15 is 2, and the demand for task undertakers of skill N is 0; the demand for task undertakers of skill 1 in time slot (task time period) 22 is 0, and the demand for task undertakers of skill N is 1, and so on. In the scenario of flexible employment, the peaks and troughs of the number of task undertakers in different periods vary greatly, and the available time and skills of each task undertaker will also be different. The practical cost (such as salary) of the task undertaker is calculated based on the actual working hours.

[0070] After determining the first corresponding relationship, you can proceed to the next step, at least based on the first corresponding relationship and the second corresponding relationship between the task undertaker and the task skills, determine the third corresponding relationship between multiple task undertakers and multiple task time periods.

[0071] Step S105: Determine a third correspondence between multiple task undertakers, multiple task skills and multiple task time periods based at least on the first correspondence and the second correspondence between the task undertakers and the task skills.

[0072] The second correspondence refers to the correspondence between the task undertaker and the task skill. Table 3 shows the second correspondence table of this embodiment.

[0073] Skill 1 Skill 2 Skill 3 Skill 4 Skill 5 Employee A yes no no no yes Employee B no yes no no no …

[0074] Table 3. Second correspondence table

[0075] As shown in Table 3, an employee can have multiple skills or only one skill. For example, in the fast food scenario, skill 1 represents cashiering skills, skill 2 represents vegetable washing skills, skill 3 represents vegetable cutting skills, skill 4 represents serving skills, and so on. Employee A can have both cashiering skills and serving skills.

[0076] Table 4 shows the task scheduling results obtained by adopting different implementation methods of step S105. The following describes the optional implementation methods of step S105 in combination with the result data of three different effects.

[0077]

[0078] Table 4. Task allocation table

[0079] In one example, step S105 may be implemented as follows: directly scheduling tasks according to the unit time granularity, that is, performing fine-grained task allocation.

[0080] The task allocation results obtained by this processing method can easily cause discontinuity of employee working hours or incoherence of work task skills. For example, the task scheduling results of employee A in Table 4 are a type of discontinuous work time allocation. The disadvantage is that there will be many gaps in the employee's attendance time that are not assigned tasks, resulting in the inability to fully utilize the employee's attendance time, reducing the effective unit time output ratio. The task scheduling results of employee B in Table 4 are a type of discontinuous work task skill allocation. The disadvantage is that the employee will switch between tasks with multiple skills, affecting work efficiency.

[0081] From the task scheduling results of employee C in Table 4, it can be seen that employee C is a relatively reasonable allocation. After working continuously for a period of time, the employee takes a short break and then performs the next continuous work. It can be seen that in order to reasonably allocate tasks in the business scenario of this embodiment, two factors need to be considered: time continuity and skill consistency.

[0082] In another example, step S105 may be implemented as follows: all work / rest time and the switching state between work skills are explicitly considered. That is, in the mathematical model, these two factors will be expressed as constraints, and a distinction is made between every two adjacent time granularities of each employee, such as Figure 2 However, this approach greatly increases the complexity of the problem, resulting in a huge scale of the problem, and it is almost impossible to produce a feasible solution within a reasonable time. Therefore, the task scheduling efficiency is extremely low and the practicality is poor.

[0083] In another example, a solution for task sequence generation is proposed, which cleverly handles these two complex factors, reduces the difficulty of the problem, improves the efficiency of task scheduling, and makes the solution more practical. Figure 3 The specific flow chart of step S105 in this embodiment is shown. In this embodiment, step S105 may include the following sub-steps:

[0084] Step S1051: Based on the first corresponding relationship, generate multiple task sequences with the same task skill requirements, number of unit task undertakers and continuous task time periods.

[0085] The task sequence includes a task sequence with time continuity and skill consistency, and is composed of a plurality of unit task undertaking units.

[0086] The generation process of the task sequence is described below in conjunction with Table 2 and Table 5. Table 5 shows the task sequence table generated in this embodiment.

[0087]

[0088]

[0089] Table 5. Sequence table of the number of task undertakers

[0090] The task sequence may be generated in the following manner: based on the first corresponding relationship, a plurality of task sequences are generated that have the same task skill requirements, number of unit task undertakers, continuous task time periods, and sequence lengths that are less than a length threshold.

[0091] In practical applications, the length threshold can be specified according to specific business requirements. The smaller the length threshold, the shorter the length of the task sequence, the more flexible the arrangement of the task undertaker, but the lower the satisfaction of the task undertaker; conversely, if the length threshold is larger, the longer the length of the task sequence, the lower the flexibility of the arrangement of the task undertaker, but the higher the satisfaction of the task undertaker.

[0092] For example, if the length threshold is 1, the length of the task sequence is 1, the number of task sequences reaches the maximum number, and the task undertaker may only arrange a 1-hour task undertaking task. In this way, for the task undertaker, the task undertaking time is very short, but the travel time is also consumed, so the satisfaction is very low. If the length threshold is 4, the length of the task sequence is 4, and the task undertaker can be arranged to undertake tasks for 4 consecutive hours. In this way, for the task undertaker, the task undertaking time is longer, but the travel time consumed does not change, so the satisfaction is higher.

[0093] By adopting this processing method, for a certain skill, the number of task undertakers can be connected in chronological order to form a continuous task sequence according to the number of task undertakers in different time slots corresponding to skill 1 in Table 2, the following task sequences can be divided: Sequence 1 corresponding to skill 1 includes: time slot 12-time slot 15; Sequence 2 corresponding to skill 1 includes: time slot 14-time slot 17; Sequence 3 corresponding to skill 1 includes: time slot 18-time slot 19, and so on.

[0094] In specific implementation, the process of generating a task sequence can be iterative, such as scanning from left to right to find a task sequence whose length does not exceed the length threshold M and the number of consecutive tasks per task undertaker. The correspondence between the sequence and each time period is represented by a vector, such as a = […0,1,1,1,1,0,…], so that the dimension of the vector corresponding to each task sequence is the same. For example, if 1 day is divided into 24 hours, the dimension of the vector of any task sequence is 24.

[0095] In the first correspondence table (Table 2), the number of unit task undertakers (such as 1) is subtracted from the number of task undertakers in the task time period corresponding to the task sequence, and the next task sequence is updated and scanned until the first correspondence list is empty, and finally the generated task sequence is output as shown in Table 5.

[0096] Step S1053: Determine the third corresponding relationship at least according to the task sequence and the second corresponding relationship.

[0097] In one example, each task undertaker may be directly assigned to a task sequence without considering the scheduling rules. Step S1053 may include the following sub-steps:

[0098] 1) Constructing a distribution model of task undertakers to task sequences based at least on the task sequence, the first corresponding relationship and the second corresponding relationship.

[0099] The mathematical model of this method can be expressed as:

[0100]

[0101]

[0102]

[0103]

[0104] Among them, x represents the task scheduling result (the third corresponding relationship), is a 0-1 variable, indicating whether the task undertaker k is assigned to task sequence s on day d; Indicates whether the task sequence s covers the time slot t on the dth day, q dt is the predicted value of the number of task undertakers of the task skill corresponding to the task sequence s at the time slot t of the dth day. Formula (1) restricts each task undertaker to be assigned to at most one task sequence s at the same time; Formula (2) indicates that the number of task undertakers assigned to a specific time slot must be greater than the number of task undertakers in that time slot. Target value It is a function of the task scheduling situation x, which can indicate the maximum coverage of task requirements, the lowest cost for the task undertaker, etc. It represents the cost of task undertaker k being assigned to task sequence s on day d, such as labor cost, skill matching degree, preference cost, etc.

[0105] 2) Determine a fourth corresponding relationship between a plurality of task undertakers and a plurality of task sequences through the allocation model.

[0106] In specific implementation, a variety of methods can be used to solve the above optimization model, such as using an optimization solver to solve it, or using a heuristic algorithm (such as a genetic algorithm), etc. Since model solving belongs to a relatively mature existing technology, it will not be described here.

[0107] 3) Determine the third corresponding relationship based on the fourth corresponding relationship.

[0108] Through the correspondence between the task undertaker and the task sequence, we can further obtain the work arrangement of the task undertaker in each task time period, as shown in Table 6.

[0109]

[0110] Table 6: Execution schedule

[0111] However, the above approach does not take into account policy and legal provisions, such as: working days, rest days, upper and lower limits of working hours in monthly / weekly dimensions, etc.; nor does it comprehensively consider factors such as humane management and employees' personalized preferences, such as: continuous working time does not exceed 4 hours, rest time does not exceed 2 hours, etc.; therefore, it will lead to problems such as low rationality of task scheduling and violation of legal provisions, resulting in low practicality.

[0112] In order to solve the above problems, the method provided in the embodiment of the present application may also include the following steps: obtaining task scheduling rules; accordingly, constructing an allocation model including restriction conditions corresponding to the task scheduling rules; using this processing method, a fourth corresponding relationship between the task undertaker and the task sequence that complies with the task scheduling rules is determined.

[0113] The task scheduling rules can be read from the task scheduling rule configuration file or obtained from the database. The task scheduling rules include but are not limited to at least one of the following rules: a threshold for the number of working days, a threshold for the number of rest days, an upper threshold and / or lower threshold for the number of working hours in the monthly / weekly dimension; an upper threshold for the continuous working time; an upper threshold for the continuous rest time, etc.

[0114] In this case, the mathematical model for assigning task undertakers to task sequence groups can be expressed as:

[0115]

[0116]

[0117]

[0118] h(x)=0 (3)

[0119]

[0120] Among them, formula (3) represents the task scheduling rule restrictions of this scenario, including all hard rules, such as the provisions of policies and regulations, and some soft rules, such as the continuous working time not exceeding 4 hours, the rest time not exceeding 2 hours, etc.

[0121] In this embodiment, a set of modeling languages ​​can be defined in advance for formula (3), and whether to add it into the mathematical model can be determined through a configuration file as shown in Table 7.

[0122] type rule Enable Hard rules Monthly working hours yes Weekly working days yes … Soft rules Continuous working time shall not exceed 4 hours no Rest time should not exceed 2 hours yes …

[0123] Table 7. Task scheduling rule configuration file

[0124] The method for solving the optimization model can be flexibly selected, such as using an optimization solver to solve it, or using a heuristic algorithm such as a genetic algorithm, etc. This plan does not make specific provisions.

[0125] In summary, this embodiment decomposes the first corresponding relationship into multiple task sequences through the above steps S1051-1053, and assigns the task undertaker to the task sequence, thereby obtaining an execution schedule from the task undertaker to the task skill to the task time period. This processing method can not only adapt to the task requirements of fine time granularity and meet the business requirements of time continuity and skill consistency, but also effectively reduce the complexity of problem solving and ensure the quality of solution as much as possible, so that work and rest time can be reasonably arranged for hundreds of task undertakers, so that the task requirements can be fully covered as much as possible to adapt to large-scale task scheduling scenarios. At the same time, various rule configurations can be taken into account to improve the satisfaction of staff.

[0126] As can be seen from the above embodiments, the task scheduling method provided by the embodiment of the present application determines the task amount of multiple task time periods; at least according to the task amount, determines the first correspondence between the task skills and the number of task undertakers in multiple task time periods; at least according to the first correspondence and the second correspondence between the task undertaker and the task skill, determines the third correspondence between multiple task undertakers, multiple task skills and multiple task time periods; this processing method can reasonably arrange work and rest time for hundreds of task undertakers, so that the task requirements can be fully covered as much as possible, and realize flexible employment to meet the task requirements with large dynamic changes, and realize the task scheduling processing of fine time granularity level and task skill level; therefore, it can effectively take into account the lower labor cost and higher staff satisfaction, that is, it can effectively improve the rationality of task scheduling and improve the performance of labor management. Among them, one aspect of flexible employment is to arrange the task acceptance time period according to the task amount according to the non-shift granularity of several hours (such as 2 hours, 4 hours, etc.), so as to match the number of personnel and the amount of tasks, avoid insufficient manpower during busy hours and rest during idle hours, realize flexible employment from the perspective of working time, and reduce labor costs. Another aspect of flexible employment is scheduling work based on the skills required by the task, matching skills with tasks, avoiding the inability to complete difficult tasks, and avoiding the waste of manpower caused by highly skilled personnel taking on tasks with low skill requirements. This is also one aspect of flexible employment, which reduces labor costs from the perspective of employee skills and ensures service quality.

[0127] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

[0128] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0129] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0130] 1. Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include non-transitory media such as modulated data signals and carrier waves.

[0131] 2. Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

Claims

1. A task scheduling method, characterized in that: include: Determine the task volume for multiple task time periods; At least according to the task amount, determining a first correspondence between task skills and the number of task undertakers in multiple task time periods, wherein the task includes multiple task skills; According to the first corresponding relationship, a plurality of task sequences having the same task skill requirements, number of unit task undertakers and continuous task time periods are generated; At least according to the task sequence, the first correspondence and the second correspondence between the task undertaker and the task skill, a distribution model of the task undertaker to the task sequence is constructed; through the distribution model, a fourth correspondence between multiple task undertakers and multiple task sequences is determined, the objective function of the distribution model is that the fourth correspondence minimizes the cost of the task undertaker, and the optimization conditions of the distribution model include: each task undertaker is assigned to at most one task sequence at the same time, and the number of task undertakers assigned to any time slot is greater than the demand value of the number of task undertakers in the time slot; According to the fourth corresponding relationship, a third corresponding relationship among multiple task undertakers, multiple task skills and multiple task time periods is determined.

2. The method according to claim 1, characterized in that The task sequence is generated in the following way: According to the first corresponding relationship, a plurality of task sequences are generated that have the same task skill requirements, number of unit task undertakers, continuous task time periods, and sequence lengths that are less than a length threshold.

3. The method according to claim 1, characterized in that The construction of the allocation model of task undertakers to task sequences includes: Get task scheduling rules; An allocation model including constraint conditions corresponding to the task scheduling rules is constructed.

4. The method according to claim 3, characterized in that The task scheduling rules include at least one of the following rules: Working days threshold, rest days threshold, upper and / or lower working hours threshold in monthly / weekly dimensions; Upper limit of continuous working time; The upper limit of continuous rest duration.

5. The method according to claim 1, characterized in that The fourth corresponding relationship is determined by an optimization solver or a heuristic algorithm at least based on the first corresponding relationship and the second corresponding relationship.

6. The method according to claim 1, characterized in that The task volume includes the order volume; The skills include: ordering service skills, serving service skills, cleaning service skills, cashier service skills, vegetable washing skills, vegetable cutting skills, and cooking skills.

7. The method according to claim 1, characterized in that The task volume includes the user consultation volume for the business object; The skills include: service skills corresponding to the business object category.

8. The method according to claim 1, characterized in that The task volume includes sales data; The skills include: shelf organizing skills, cleaning skills, cashiering skills, and goods shelving skills.

Images