Shift schedule creation method and system

Abstract

Provided are a method and a system for creating a shift table on the basis of objective information and prediction. The present technology is a method by which, for example, a computer creates a shift table allocating employees to processes by time slot for a workplace that includes a plurality of processes and a plurality of employees, the method including a method for creating a shift table whereby a computer: determines the necessary number of employees for each time slot for each process by using a prediction model which has been trained on the basis of data about the number of employees, the working rate of employees, and the sales data, per time slot for each process in the past; determines constraint conditions including the time constraints and skills of each employee, and the necessary skills for each process; determines desired work times, including the desired work days and the desired work time slots of each employee; and uses a genetic algorithm to create a shift table in which employees satisfying the necessary number of employees, the constraint conditions, and the desired work times are assigned to each process for each time slot.

Description

[Technical Field]

[0001] The present technology relates to a method and system for creating a shift schedule in a workplace having a plurality of processes and a plurality of employees, in which each employee is assigned to each process by time period. [Background technology]

[0002] In workplaces such as restaurants, retail stores, and factories, there are various processes such as cooking, selling, serving customers, manufacturing, packaging, and transportation, and these tasks are often shared among multiple employees. Employees are classified into various employment types, such as full-time employees, contract employees, temporary employees, part-time employees, and casual employees, and working conditions are applied to each employee according to their employment type. Furthermore, when employing minors such as high school students or foreign students, additional statutory working conditions apply, such as a ban on late-night work and a maximum weekly working hour limit.

[0003] Managers and supervisors who operate workplaces such as stores and factories may create shift schedules that assign working hours and processes to each employee so that each process can be carried out smoothly during store business hours or factory operating hours. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 11-143850 [Patent Document 2] Japanese Patent Application Laid-Open No. 2004-310530 [Patent Document 3] Japanese Patent Application Laid-Open No. 2013-30182 [Patent Document 4] Patent No. 5807980 [Patent Document 5] International Publication No. 2021 / 186747 Brochure Summary of the Invention [Problem to be solved by the invention]

[0005] The number of customers and production volume may vary depending on the season, day of the week, and time of day, and the number of employees required for each process may also vary depending on the time of day. In order to respond to such fluctuations, a method and system for creating shift schedules based on objective information and predictions is desired.

[0006] What is needed is a method and system for creating shift schedules that can reduce the time required for managers and supervisors to create shift schedules and that do not rely on personal methods.

[0007] What is needed is a method and system for creating shift schedules that appropriately meet the many conditions that must be taken into consideration and whose priorities are unclear, such as sales targets, labor cost targets, the combination of skills required to carry out each process and the skills possessed by each employee, and the working conditions and desired working hours of each employee.

[0008] There is a need for a method and system for creating shift schedules that can flexibly and quickly respond to sudden changes in sales targets or desired working hours, as well as emergency restrictions on business hours imposed by national or local governments to prevent infectious diseases.

[0009] A method and system for creating a shift schedule that can respond to multiple key performance indicators (KPIs) such as sales, labor costs, utilization rate, delivery time, customer satisfaction, and employee satisfaction according to the importance or priority in accordance with management policies is desired.

[0010] For example, in a chain store with multiple stores, a method and system for creating shift schedules that can display the shift schedules for each store at a glance and allow employees to grasp whether they are in a surplus or shortage is desirable. Also desirable is a method and system for creating shift schedules that can create and present shift schedules for each store at the headquarters of the chain store.

[0011] What is needed is a method and system for creating a shift schedule in a short time even when there are many processes, employees, or other conditions that must be met.

[0012] What is needed is a method and system for creating a shift schedule that satisfies constraints such as time limits and skills of each employee, as well as desired work hours.

[0013] What is needed is a method and system for creating a shift schedule that satisfies the highest priority of constraints such as time limits and skills of each employee, and desired working hours, even if it does not satisfy all of these. [Means for solving the problem]

[0014] The present technology includes, for example, a method for a computer to create a shift schedule in which employees are assigned to each process for each time period in a workplace having a plurality of processes and a plurality of employees, the method including: determining the number of employees required for each time period in each process using a prediction model trained by machine learning based on past data on the number of employees for each time period in each process, employee utilization rates, and sales revenue; determining constraints including the time limits and skills of each employee and the skills required for each process; determining desired working hours including desired working days and desired working hours for each employee; and using a genetic algorithm to create a shift schedule in which employees who satisfy the required number of employees, constraints, and desired working hours are assigned to each process for each time period. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a diagram illustrating a shift schedule creation system according to an embodiment of the present technology. [Figure 2] 1 is a flowchart illustrating a shift schedule creation method according to an embodiment of the present technology. DETAILED DESCRIPTION OF THE INVENTION

[0016] FIG. 1 shows a shift schedule creation system 100 according to an embodiment of the present technology.

[0017] The shift schedule creation system 100 includes an application server 120, an employee management server 130, a sales management server 140, and a client terminal 150 connected to a network 110.

[0018] The network 110 may be the Internet, a private network, a virtual private network (VPN), a local area network (LAN), a fifth generation mobile communication system (5G), or a combination thereof, and may be wired or wireless, or a combination thereof.

[0019] The application server 120, the employee management server 130, and the sales management server 140 are each computers having a processor (not shown), a memory (not shown) that stores a program, and a communication function (not shown).

[0020] The application server 120, employee management server 130, and sales management server 140 may each have a storage device (not shown) accessible directly or via the network 110. The storage device may be a network-attached storage (NAS) or cloud storage, or may be a hard disk drive (HDD) or solid-state drive (SSD) stored in the same enclosure as the server.

[0021] The application server 120 stores a program for creating a shift schedule according to an embodiment of the present technology, executes the program stored in memory in response to a request from the client 150, and returns the results to the client 150.

[0022] The employee management server 130 may include information about each employee, such as their employment status, skills, qualifications, wages, experience, processes they have undergone, other occupations, nationality, age, gender, etc. Employment status may be, for example, full-time, part-time, or casual.

[0023] Skills may be, for example, qualifications, education level, training status, or a combination thereof. Skills may also be the ability to independently perform a specific process, such as a manufacturing process or a sales process, the ability to instruct the execution of a specific process, the ability to perform a specific process under instruction, or a combination thereof. Skills may also be quantified as skill levels, such that an employee with skill level 1 can perform a process under the instruction or supervision of an employee with skill level 3, and employees with skill levels 2 and 3 can perform a process independently. Skills may also define the range of processes that can be performed. For example, an employee with skill level A can perform process A, an employee with skill level AB can perform processes A and B, and an employee with skill level C can perform process C. These may also be combined to define an employee with skill level A2 can independently perform process A, and an employee with skill level C1 can perform process C under the instruction or supervision of an employee with skill level C3. Skills may be further subdivided and associated with an employee's qualifications, wages, experience, and processes experienced.

[0024] The employee management server may also use employee information to determine the working conditions applicable to each employee and determine the corresponding time restrictions. For example, if an employee's "other occupation" is "high school student" and their "employment type" is "part-time work," the working conditions for part-time high school students, such as statutory time restrictions prohibiting late-night work, may apply, and the employee may therefore not be assigned to any processes during late-night hours. Furthermore, if an employee's "other occupation" is "student" and their "nationality" is not "Japanese," the working conditions for international students, such as statutory time restrictions limiting weekly working hours to 28 hours, may apply, and the employee may therefore need to be assigned to processes so as not to exceed the time restrictions.

[0025] Furthermore, the employee management server may determine contractual time restrictions, such as time restrictions under an employment contract, work regulations, or labor-management agreement. For example, if an employee's "other occupation" is "housewife" and her "employment type" is "part-time," a contractual time restriction that keeps working conditions within the dependent range, such as an annual salary of 1,030,000 yen or less, applies, and it may be determined that the employee must be assigned to processes so as not to exceed that time restriction.

[0026] The sales management server 140 stores sales data for each past time period (e.g., consecutive 1 minute, 5 minutes, 10 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 8 hours, 12 hours, or 1 day). The sales data for each past time period may be collected in real time, periodically (e.g., every hour), or irregularly (e.g., at any time) using, for example, a POS (Point of Sales) system. The sales management server 140 may also store data such as the number of employees and employee utilization rates for each time period in each past process, and may also store data such as labor costs, delivery dates, customer satisfaction, and employee satisfaction. Furthermore, the sales management server 140 may also store data such as the number of employees and employee utilization rates for each time period in each past process, in association with sales.

[0027] Furthermore, the sales management server 140 may predict and determine the number of employees required for each time slot in each process, either currently or in the future, using a prediction model trained by machine learning based on data on the number of employees for each time slot in each process in the past, the employee utilization rate, and sales. The prediction model may be a model trained using the correspondence between the number of employees for each time slot in each process in the past, the employee utilization rate, and sales as training data, or it may be a deep learning model trained using the number of employees for each time slot in each process in the past, the employee utilization rate, and sales.

[0028] The client terminal 150 is a computer, tablet terminal, smartphone, or the like that has the function of communicating with the application server 120, employee management server 130, and sales management server 140 via a wired or wirelessly connected network 110. There may be two or more client terminals 150, and each may be owned by a manager, supervisor, chain headquarters, or employee.

[0029] FIG. 2 shows a shift table creation method according to an embodiment of the present technology, with reference to FIG.

[0030] 2, in step 210, the shift schedule creation method 200 is started in the application server 120. Next, in step 220, the application server 120 determines the number of employees required for each time period in each process using a machine-learned prediction model based on past data on the number of employees for each time period in each process, employee utilization rates, and sales revenue stored in the application server 120. The determination of the number of employees required may be performed in advance in the sales management server 140, or may be performed by the sales management server 140 in response to a request from the application server 120.

[0031] The number of employees required may be determined so as to maximize employee utilization, to meet a target utilization rate, to maximize sales, or to meet a target sales amount.

[0032] Furthermore, the number of employees required may be determined to minimize labor costs, meet target labor costs, minimize delivery times, maximize customer satisfaction, meet target customer satisfaction, maximize employee satisfaction, or meet target employee satisfaction.

[0033] Furthermore, the required number of employees may be determined so as to maximize a value obtained by weighting two or more arbitrarily selected from the above-mentioned indicators. For example, the required number of employees may be determined so as to maximize a value obtained by weighting employee utilization rate and sales.

[0034] Next, in step 230, the application server 120 determines constraints, including the time limits and skills of each employee and the skills required for each process. The constraints may be determined in advance by the employee management server 130, or may be determined by the employee management server 130 in response to a request from the application server 120.

[0035] In addition, the constraints may be determined by receiving or adopting all of the constraints stored in the employee management server 130, by receiving or adopting only those constraints whose priority is equal to or higher than a predetermined value, or by receiving or adopting only selected constraints.

[0036] Additionally, constraint determination may be made to include cases where statutory constraints are met but contractual constraints are not met.

[0037] Next, in step 240, the application server 120 determines the desired working hours, including the desired working days and desired working hours, of each employee. The desired working hours may be determined in advance in the client terminal 150, or may be determined by the client terminal 150 in response to a request from the application server 120. More specifically, the desired working hours may be determined by receiving data entered and transmitted by the employee who owns the client terminal 150.

[0038] Desired working hours may be determined based on the time periods during which an employee actively wishes to work, the time periods during which an employee is willing to work if requested (or the time periods with the highest priority), the time periods during which an employee wishes to avoid working, the time periods during which an employee is unable to work (or the time periods with the lowest priority), or a combination of these.

[0039] Furthermore, the desired working hours may be determined periodically, such as daily, weekly, or monthly, at any time, by selecting any time period, by selecting a specific pattern (for example, 10:00 to 16:00 on weekdays), or by a combination of these.

[0040] Next, in step 250, the application server 120 uses a genetic algorithm to create a shift schedule that assigns employees who satisfy the required number of employees, constraints, and desired working hours to each process for each time period.

[0041] The shift schedule may be created so as to maximize an evaluation value including at least one of the indicators such as employee utilization rate, sales, labor costs, delivery time, customer satisfaction, and employee satisfaction.

[0042] Alternatively, the shift table may be generated by randomly generating an initial shift table, evaluating the generated shift table, leaving a plurality of highly advantageous shift tables, crossing the highly advantageous shift tables with each other, avoiding local solutions by mutation and accelerating convergence, and repeating this until convergence or the maximum number of generations is reached.

[0043] Furthermore, the shift schedule may be created by assigning employees to each process by time period in descending order of priority.

[0044] The shift schedule created in this way may be output in an editable format so that managers or supervisors can manually adjust it individually, or in a format such as an iCalendar file so that it can be easily displayed in a calendar application.

[0045] Furthermore, if the created shift schedule does not satisfy all of the constraints and desired working hours, a warning may be output to the effect that there are conditions that have not been satisfied and the details of such conditions.

[0046] Furthermore, the created shift schedule may be sent to the client terminal 150 so that it can be viewed by each employee, manager or supervisor, chain headquarters, etc.

[0047] Furthermore, the evaluation value of the KPI based on the created shift table may be transmitted to the client terminal 150.

[0048] Next, at step 260, the shift schedule creation method 200 ends. [Industrial Applicability]

[0049] The present technology makes it possible to provide a method for creating shift schedules based on objective information and predictions. [Explanation of symbols]

[0050] 100 Shift schedule creation system 110 Network 120 Application Servers 130 Employee Management Server 140 Sales Management Server 150 client terminals

Claims

1. A method for creating a shift schedule in which employees are assigned to each process by time period in a workplace having a plurality of processes and a plurality of employees, by a computer, comprising: The computer determining the number of employees required for each time period in each process using a prediction model trained by machine learning based on the number of employees and employee utilization rate for each time period in each process in the past, and sales data stored in association with the number of employees and the employee utilization rate; Determine the constraints, including the time limits and skills of each employee and the skills required for each process. Determine each employee's desired working days and desired working hours, creating a shift schedule that assigns employees who satisfy the required number of employees, the constraints, and the desired working hours to each process for each time period using a genetic algorithm; the prediction model includes a model trained using training data of a correspondence relationship between the number of employees for each time period in each process in the past, and the employee utilization rate and sales, or a deep learning model trained using the number of employees for each time period in each process in the past, the employee utilization rate and sales, Determining the required number of employees includes at least one of determining the number of employees that maximizes employee utilization, determining the number of employees that meets a target utilization rate, determining the number of employees that maximizes sales, and determining the number of employees that meets a target sales amount. How to create a shift schedule.

2. The method of claim 1 , wherein the data further comprises at least one of labor costs, delivery times, customer satisfaction, and employee satisfaction.

3. Determining the required number of employees The method includes at least one of determining the number of employees that minimizes labor costs, determining the number of employees that meets target labor costs, determining the number of employees that minimizes delivery time, determining the number of employees that meets target delivery time, determining the number of employees that maximizes customer satisfaction, determining the number of employees that meets target customer satisfaction, determining the number of employees that maximizes employee satisfaction, and determining the number of employees that meets target employee satisfaction. The method of claim 2.

4. Determining the required number of employees is determining the number of employees that maximizes a predetermined weighted evaluation of employee utilization rate and sales revenue; The method of claim 1.

5. the employee's time limits are determined by the working conditions applicable to that employee; The method of claim 1.

6. The method of claim 1 , wherein the skills include at least one of a qualification, an education level, and a training status.

7. The method of claim 1 , wherein the skills include at least one of being able to independently perform a particular process, being able to supervise the performance of a particular process, and being able to perform a particular process with supervision.

8. creating a shift schedule in which employees who satisfy the required number of employees, the constraints, and the desired working hours are assigned to each process for each time period using the genetic algorithm; and creating a shift schedule that maximizes an evaluation value including at least one of employee utilization rate, sales, labor costs, delivery time, customer satisfaction, and employee satisfaction. The method of claim 2.

9. creating a shift schedule in which employees who satisfy the required number of employees, the constraints, and the desired working hours are assigned to each process for each time period using the genetic algorithm; Randomly create an initial shift schedule, Evaluate the created shift schedule, Keeping multiple highly advantageous shift schedules, Crossing shift schedules with high advantages, Mutation avoids local minima and accelerates convergence. Repeating until convergence or a maximum number of generations is reached, The method of claim 1.

10. The method of claim 1 , wherein the process comprises at least one of a manufacturing process and a sales process.

11. 2. The method of claim 1, wherein the time period comprises at least one of 1 minute, 5 minutes, 10 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 8 hours, and 12 consecutive hours.

12. The method of claim 1 , wherein the constraints include at least one of a statutory constraint and a contractual constraint.

13. The method of claim 12 , wherein satisfying the constraints includes satisfying the statutory constraints but not the contractual constraints.

14. The method of claim 1 , wherein satisfying the required number of employees, the constraints, and the desired working hours includes satisfying only those of which priority is equal to or greater than a predetermined value.

15. In a workplace with multiple processes and multiple employees, a computer creates a shift schedule in which each employee is assigned to each process by time period, The computer Determine the number of employees required for each time period in each process using a prediction model trained by machine learning based on the number of employees and employee utilization rates for each time period in each process in the past, as well as sales data stored in association with the number of employees and the employee utilization rates; Determine the constraints, including the time limits and skills of each employee and the skills required for each process. Determine each employee's desired working days and desired working hours, a genetic algorithm is used to create a shift schedule in which employees who satisfy the required number of employees, the constraints, and the desired working hours are assigned to each process for each time period; the prediction model includes a model trained using training data of a correspondence relationship between the number of employees for each time period in each process in the past, and the employee utilization rate and sales, or a deep learning model trained using the number of employees for each time period in each process in the past, the employee utilization rate and sales, Determining the required number of employees includes at least one of determining the number of employees that maximizes employee utilization, determining the number of employees that meets a target utilization rate, determining the number of employees that maximizes sales, and determining the number of employees that meets a target sales amount. Shift schedule creation system.

16. A computer-readable medium storing a program for a computer to create a shift schedule in which each employee is assigned to each process by time period in a workplace having a plurality of processes and a plurality of employees, The computer, Determine the number of employees required for each time period in each process using a prediction model trained by machine learning based on the number of employees and the employee utilization rate for each time period in each process in the past, and the sales data stored in association with the number of employees and the employee utilization rate; Determine the constraints, including the time limits and skills of each employee and the skills required for each process; Determining each employee's preferred working days and preferred working hours, including preferred working hours; using a genetic algorithm to create a shift schedule in which employees who satisfy the required number of employees, the constraints, and the desired working hours are assigned to each process for each time period; the prediction model includes a model trained using training data of a correspondence relationship between the number of employees for each time period in each process in the past, and the employee utilization rate and sales, or a deep learning model trained using the number of employees for each time period in each process in the past, the employee utilization rate and sales, Determining the required number of employees includes at least one of determining the number of employees that maximizes employee utilization, determining the number of employees that meets a target utilization rate, determining the number of employees that maximizes sales, and determining the number of employees that meets a target sales amount. Computer-readable medium.

Images