Work shift automatic generation system, work shift automatic generation method, and work shift automatic generation program

Abstract

To reduce a burden on a user when creating a work shift.SOLUTION: A work shift generation system includes a shift generation unit (15) that generates a work shift, a shift presentation unit (11) that presents the work shift generated by the shift generation unit (15), a correction request acquisition unit (12) that receives a content and an importance of a correction request from a shift target person for the presented work shift, and a constraint parameter weighting unit (14) that sets a weight of a setting value of a parameter corresponding to the correction request based on the content and the importance of the correction request.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an automatic work shift generation system, an automatic work shift generation method, and an automatic work shift generation program for automatically generating work shifts. [Background technology]

[0002] In hospitals, doctors create and allocate work shifts for rotating duties such as on-call and outpatient care. Work shifts are managed by the doctor in charge. Doctors' work shifts differ from those of nurses in that they are complex and require consideration of each individual's unique requirements, such as outpatient care, field work, and specialization. Furthermore, due to a chronic shortage of doctors, local hospitals urgently need to prevent current doctors from leaving and promote the recruitment of new doctors. To address this, some hospitals are creating shifts that take into account doctors' specializations and work-life balance in order to improve doctors' work engagement. As mentioned above, shifts must be created taking into account the complex requirements of each doctor, as well as their specialty and work-life balance. To do this, the doctor in charge must manually create shifts that meet the complex requirements by checking each doctor's requests. This places a significant burden on both the person in charge of creating shifts and the other doctors, both in terms of time and mental health.

[0003] In light of these circumstances, systems disclosed in Patent Documents 1 and 2 below have been publicly known. The former is a technology that generates schedules that realize the preferences of multiple users as fairly as possible. Users input a set of preferences, and a fair work shift is generated by gradually reflecting preferences with higher priority, in order to level out the likelihood of those preferences being realized. The latter proposes a method that can create a work schedule in a relatively short time while satisfying complex evaluation rules. The user is prompted to input evaluation rules, including the user's level of skill, work schedule, work assignments for each day of the week, and the number of staff required, and the computer automatically calculates the results to satisfy the evaluation rules as much as possible. After meeting the specified termination conditions, a provisional solution is displayed on the screen, and the user is prompted to manually enter corrections to the provisional solution. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-103810 [Patent Document 2] Japanese Patent Application Laid-Open No. 2004-151790 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the former method requires the user to input the desired data set each time a work shift is created, which does not reduce the burden on the user. The latter system prompts users to input evaluation rules, creates a provisional shift schedule through automatic computer calculations, and then requests that users manually enter corrections. As a result, even from the second time onwards, users will need to enter evaluation rules and manually enter corrections every time they create a work shift, which does not reduce the burden on users. The present invention has been made in consideration of the above circumstances, and its main objective is to provide an automatic work shift generation system, an automatic work shift generation method, and an automatic work shift generation program that can reduce the burden on shift recipients (users) when creating work shifts. [Means for solving the problem]

[0006] In order to achieve the above object, the automatic work shift generation system according to the present invention comprises: a shift generation unit that generates work shifts; a shift presentation unit that presents the work shifts generated by the shift generation unit; A modification request acquisition unit that receives a modification request from a shift target person regarding the work shift presented by the shift presentation unit; a weighting unit that sets a weight for the revision request in consideration of the importance of the revision request and a weight for an existing revision request, The shift generation unit automatically generates the work shift in consideration of the modification request received by the modification request acquisition unit and the weight set by the weighting unit. [Effects of the Invention]

[0007] According to the above configuration, it is possible to reduce the burden on the shift recipient when creating work shifts. [Brief explanation of the drawings]

[0008] Advantages and features provided by embodiments of the present invention will be more fully understood from the following detailed description and accompanying drawings, which are given by way of example and are not intended to be limiting of the invention. [Figure 1] 1 is a diagram illustrating a schematic configuration of an automatic work shift generation system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a schematic diagram showing the processing content of the automatic work shift generation system according to the present embodiment. [Figure 3] FIG. 10 is a diagram showing an example of an input screen for inputting a correction request. [Figure 4] FIG. 10 is a diagram showing an example in which the reason for the correction request is converted into constraint parameters, setting values, and weights. [Figure 5] FIG. 2 is a diagram illustrating an example of processing performed by the automatic work shift generation system according to the present embodiment. [Figure 6] FIG. 10 is a diagram illustrating an example of a calculation result of the fitness of a genetic algorithm. [Figure 7] FIG. 10 is a diagram showing an example of a work shift in which a request for modification is reflected. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a schematic diagram of the automatic work shift generation system. The automatic work shift generation system 1 basically comprises a shift management server 2, a network 3, and a user terminal 4.

[0010] The shift management server 2 is connected to a user terminal 4 via a network 3 . The shift management server 2 comprises a control unit 5, a storage unit 6, and an interface unit 7. The interface unit 7 is connected to the network 3. The control unit 5 is made up of an arithmetic processing device (a processor such as a CPU, MPU, or DSP) and controls the operation of each unit of the server by performing various arithmetic processing. The processing operation of the shift management server 2 is realized when the control unit 5 executes a processing program stored in the storage unit 6. Here, the shift management server 2 is made up of one server device, but it may also be made up of multiple server devices. The shift management server 2 may also be a cloud.

[0011] The network 3 is assumed to be a wireless communication line for mobile phones, the Internet, a LAN, etc., and some user terminals are connected wirelessly via a base station 2a that performs wireless communication. The user terminal 4 is a smartphone 4 a, a tablet 4 b, a notebook personal computer 4 c, a desktop computer 4 d, or the like, and accesses the site provided by the shift management server 2. A user who is a shift target logs in to the site of the shift management server 2 via a user terminal 4. By registering in advance, the shift target is assigned a user ID and password, and can log in using that user ID and password.

[0012] The processing content of the automatic work shift generation system 1 will be described with reference to Fig. 2. Fig. 2 is a schematic diagram showing the processing content of this system. In the automatic work shift generation system 1, a processing program stored in the storage unit 6 of the shift management server 2 is executed by the control unit 5, and the processing of the functional blocks shown in FIG. 2 is realized. That is, the automatic work shift generation system 1 according to the present invention is composed of a shift presentation unit 11, a correction request acquisition unit 12, a correction request conversion unit 13, a constraint parameter weighting unit 14, and a shift generation unit 15, as shown in FIG. 2.

[0013] The shift presentation unit 11 presents the work shifts generated by the shift generation unit 15 to each doctor. Each doctor checks the shifts and submits a request for revision at the revision request acquisition unit 12.

[0014] The correction request acquisition unit 12 is provided with an input screen that allows each doctor to check the shifts presented by the shift presentation unit 11 and input correction requests, and accepts the input. The correction request input screen is, for example, as shown in Figure 3, and has boxes for entering the date and time to be corrected, the date and time to which the correction is desired, the importance, and the reason for the correction. The doctor who is the shift target (user) specifies the target date and time to be corrected from the shift, sets the date and time to which the correction is desired, sets the importance, and enters the reason for the correction. In this example, the importance is set to only two levels, "high" and "low," to reduce the burden on the shift target (user), but it may also be set to multiple levels, three or more. The reason for the change can be entered as text, and the reason for the correction is entered in natural language.

[0015] The modification request conversion unit 13 converts the desired date and time, importance, and reason entered by the shift target employee in the modification request acquisition unit 12 into constraint parameters, setting values, and weights. For example, the target date and time is set to 9:00-12:00 on Wednesday, March 13, 2024, and the desired modification date and time is set to 12:00-19:00 on Wednesday, March 13, 2024. The importance is set to "high," and the reason is entered as "Since I finish my on-call shift every Wednesday, I would like to avoid outpatient appointments in the morning." Based on the reason, the morning of Wednesday is recognized as a "prohibited work time slot." The "prohibited work time slot" is set as the constraint parameter, and the "setting value" is set to "every Wednesday morning." In this case, the constraint parameter may be selected from among pre-prepared constraint parameter candidates. For example, parameters such as "maximum continuous work hours," "prohibited work time slot," and "ratio of outpatient appointments to on-call appointments" may be prepared as constraint parameter candidates, and the parameter may be selected from among these.

[0016] This can be achieved by using a general-purpose LLM, for example, by writing a prompt like the one below. "Based on the reason for correction below, select the appropriate constraint parameter from the constraint parameter candidates below and set the setting value and weight. Please set the weight in the range of 0 to 10. Set the weight to 10 for "high" importance and 3 for "low" importance. Reason for revision: Since the on-call shift ends every Wednesday, I would like to avoid outpatient appointments in the morning. Candidates for constraint parameters: upper limit of consecutive working hours, prohibited working hours, ratio of outpatient to on-call shifts Set the value within the following range. Maximum consecutive working hours: 0-24 Work prohibition hours: Days and times Outpatient / on-call ratio X:Y

[0017] The constraint parameter weighting unit 14 weights which constraint parameters to prioritize. As shown in Figure 4, the constraint parameters are composed of multiple constraint parameters, each with a set value and weight. While the constraint parameters themselves are common to all doctors registered in the system, the set values ​​and weights of the constraint parameters can be set differently for each doctor. The weights are used as the strength of the constraints in the shift generation unit 15. Weights can be set in multiple levels. For example, in a 10-level system, the maximum value of 10 is the set value that must be achieved. In addition, values ​​1 to 9 may prioritize a higher weight of another doctor, which may not be achieved; the smaller the weight, the higher the possibility. The minimum value, 0, indicates a constraint that is not taken into account at all. These constraint parameters are always maintained and are updated when a revision request is acquired from the revision request acquisition unit 12 during shift creation.

[0018] A method for setting the weights of the constraint parameters will now be described. When the constraint parameters, setting values, and weights are received by the revision request conversion unit 13 based on the reason and importance input by the revision request acquisition unit 12, the constraint parameter weighting unit 14 modifies the weights of each constraint parameter. If the importance received by the revision request acquisition unit 12 is "high," the revision request conversion unit 13 sets the received weight to, for example, 10, and if the importance received is "low," the received weight to, for example, 3. The initial value of the weight is 0, and the previously set weight is modified by taking the received weight into account. In other words, the weight is modified by taking into account the previously set weight value and the weight set for the received importance based on the revision request.

[0019] For example, the weight is corrected by taking the average of the set weight value and the weight for the received importance. Therefore, if the set weight is 0 and the set importance is high (weight 10), the weight is changed to 5 [(0 + 10) / 2]. Also, if the set weight is 5 and the set importance is low (weight 3), the weight is changed from 5 to 4 [(5 + 3) / 2].

[0020] To ensure fairness among shift candidates and to differentiate between individual weights, it is a good idea to set a total weight limit for each individual's weight for the shift constraint parameters. For example, the total weight can be set to 30, and the results of the above calculation can be normalized by 30 to reduce the total weight. For example, if the weighting calculation results in the weights of each constraint parameter being 4, 20, 20, 12, and 4, the total weight becomes 4 + 20 + 20 + 12 + 4 = 60, so all weights are multiplied by the total weight limit value / total amount. In other words, multiply by 30 / 60 = 0.5, and the weights become 2, 10, 10, 6, and 2. This not only ensures fairness in fulfilling requests by making the total amount for all shift candidates the same, but also helps differentiate the weights by preventing all constraints from being set to a weight of 10.

[0021] The shift generation unit 15 automatically creates shifts for each doctor using the constraint parameters weighted by the constraint parameter weighting unit 14 as constraints. It obtains the setting values ​​and weights specific to each doctor and generates shifts. Conditions with high parameter weights are reflected in the shifts with higher priority than conditions with low parameter weights. An existing shift creation algorithm, such as a genetic algorithm, is used as the shift creation algorithm. However, the shift creation algorithm is not limited to this, and other algorithms may also be used.

[0022] An example of processing according to the present invention is shown in Figure 5. As shown in the figure, the automatic work shift generation system 1 first sets constraint parameter setting values ​​and initial weighting values ​​in the constraint parameter weighting unit 14 (step S1). Then, in this state, the shift generation unit 15 randomly creates shifts for each shift recipient (each doctor) (step S2). The shifts are then transmitted to the shift presentation unit 11, which then displays and presents the transmitted shifts on the screen of a terminal operable by the shift recipient (each doctor) (step S3). Each doctor confirms the presented shifts (step S4) and, if necessary, inputs a modification request via the form screen shown in FIG. 3 (step S5). The input modification request is accepted by the modification request acquisition unit 12 (step S6). The modification request conversion unit 13 converts the content of the modification request into constraint parameters, setting values, and weights based on the content of the modification request (step S7). The constraint parameter weighting unit 14 then modifies the weights of the existing constraint parameters based on the weights converted by the modification request conversion unit 13 (step S8). Thereafter, the shift generation unit 15 generates (modifies) a work shift by taking into account the setting values ​​of the constraint parameters converted by the modification request conversion unit 13 and the weights of the constraint parameters modified by the constraint parameter weighting unit 14 (step S9). This generated work shift is then presented by the shift presentation unit 11 (step S3), and similar processes are subsequently performed (steps S4 to S9).

[0023] Therefore, the work shifts presented by the shift presentation unit 11 are optimized by reflecting the latest revision request, so the shift recipient does not need to repeatedly input the same revision request each time a work shift is presented. Such work shifts are reviewed, for example, once a month and revised as necessary using the method described above.

[0024] Next, we will show an example of this idea using a genetic algorithm. A genetic algorithm is an algorithm that randomly generates multiple shift candidates, evaluates whether they satisfy certain constraints using an index called fitness, selects shift candidates based on the evaluated fitness, and regenerates shift candidates using crossover and mutation techniques, repeating this process multiple times until the fitness meets a criterion, thereby deriving the optimal shift.

[0025] In this idea, the weights of the constraint parameters mentioned above are taken into consideration when calculating the fitness of this genetic algorithm. Figure 6 shows an example of the results of fitness calculations. In this Figure 6, the fitness is calculated by narrowing it down to Person A from the overall shift results. The shift candidate is evaluated to see if it satisfies the set values ​​of each constraint parameter. If the constraint is satisfied, it is set to 1, and if not, it is set to 0. The fitness is calculated by multiplying this evaluation value by the weight. For example, the work shift shown in Figure 7, which reflects the revised request, meets the request that outpatient appointments not be accepted on Wednesday mornings, so the "No work hours: Wednesday mornings" column satisfies the constraint. Therefore, the weight of 10 becomes part of the relevance (10). If the total weight is limited to a maximum of 30 due to the total weight restriction mentioned above, the maximum fitness value will be 30 and the minimum will be 0. The closer the fitness value is to 30, the more satisfied Person A's shift request can be evaluated. This is calculated for all shift candidates. Using this fitness value, work shifts are regenerated using a genetic algorithm, generating work shifts whose fitness value meets the standard.

[0026] This regenerated work shift is presented by the next shift presentation unit 11, and any further requests for revisions to the presented work shift are accepted. As a result, the shift is optimized each time it is created, reducing the frequency of revisions. This reduces the input burden on the shift recipient and reduces the man-hours required for revisions. In addition, the complex conditions required for shift creation can be set with a few input instructions, as shown in Figure 3, reducing the burden on the person entering the conditions. Moreover, since there is no need for direct communication and coordination between the shift creator and each doctor, the mental burden on both the shift creator and each doctor is reduced, and each doctor can feel free to reflect their requests regarding shifts.

[0027] In this way, the automatic work shift generation system 1 according to this embodiment: a shift generation unit 15 that generates work shifts; a shift presentation unit 11 that presents the work shifts generated by the shift generation unit 15; an adjustment request acquisition unit 12 that receives adjustment requests and their importance from the shift target person regarding the work shift presented by the shift presentation unit 11; a weighting unit (constraint parameter weighting unit 14) that sets a weight for a parameter setting value corresponding to the modification request based on the content and importance of the modification request, The shift generation unit 15 generates the work shifts in consideration of the weights of the parameter setting values ​​set by the weighting unit (constraint parameter weighting unit 14).

[0028] Therefore, work shifts are optimized each time they are created, eliminating the need for frequent revisions. This means that there is no need to input the revision requests of the shift recipients each time a work shift is created, reducing the burden on the shift recipients (users).

[0029] Furthermore, since there is no need for direct communication and coordination between the shift planner and each doctor, the mental burden on both the shift planner and each doctor is reduced, and each doctor can feel free to reflect their requests regarding shifts. Here, the correction request acquisition unit 12 includes a correction request conversion unit 13 that acquires a correction request in natural language and converts the correction request into a setting value of one of the preset parameters. Therefore, the shift generation unit 15 generates work shifts taking into account the setting values ​​of the converted parameters and their weights. In this way, requests for corrections are entered in natural language and analyzed to correct the shift, so the complex conditions required for shift creation (shift correction) can be set with fewer input instructions, reducing the setting burden on the shift recipient (user).

[0030] Furthermore, the modification request conversion unit 13 converts the importance of the modification request into a weight for the parameter setting value. At this time, the weighting unit (constraint parameter weighting unit 14) may set the weight for the parameter setting value corresponding to the modification request in consideration of the existing weight.

[0031] Furthermore, the weight may be set using numerical values ​​set in multiple stages, and an upper limit may be set on the total weight of the shift target person, and the weighting unit (constraint parameter weighting unit 14) may set the weight so that it falls within the upper limit of the total weight of the shift target person. By imposing a total quantity limit on each person's weight in this way, it is possible to ensure fairness among those eligible for shifts and to differentiate between each person's weight.

[0032] Further, the automatic work shift generation method according to the present embodiment is a method for generating a work shift by an automatic work shift generation system, Present the work shift generated by the shift generation unit; For the proposed work shift, we accept requests for modification from the shift subject and their importance, setting a weight for a parameter setting value corresponding to the correction request based on the content and importance of the correction request; The shift generation unit generates the work shifts taking into consideration the weights of the parameter setting values. Therefore, work shifts are optimized each time they are created, eliminating the need for frequent revisions. This means that there is no need to input the revision requests of the shift recipient each time a work shift is created, reducing the burden on the shift recipient (user).

[0033] Furthermore, the automatic work shift generation program according to the present embodiment is an automatic work shift generation program that automatically generates a work shift, Present the generated work shift, For the proposed work shift, the system accepts requests for corrections from the shift target and their importance, setting a weight for a parameter setting value corresponding to the modification request based on the content and importance of the modification request; The work shifts are generated taking into consideration the weights of the set values ​​of the parameters. Therefore, work shifts are optimized each time they are created, eliminating the need for frequent revisions. This means that there is no need to input the revision requests of the shift recipients each time a work shift is created, reducing the burden on the shift recipients (users).

[0034] <Supplementary information> Although embodiments and modifications of the automatic work shift generation system, automatic work shift generation method, and automatic work shift generation program according to the present invention have been described, the present invention is not limited to the above-described embodiments and modifications. The present invention also includes forms obtained by applying various modifications to the above-described embodiments and modifications that would occur to a person skilled in the art, and forms realized by arbitrarily combining the components and functions of the embodiments and modifications within the scope of the present invention. The scope of the present invention should be interpreted by the appended claims. [Industrial Applicability]

[0035] INDUSTRIAL APPLICABILITY The present invention is useful as a technology that can reduce the input burden on shift recipients (users) when automatically generating work shifts. [Explanation of symbols]

[0036] 1. Automatic work shift generation system 11 Shift display unit 12 Correction Request Acquisition Department 13 Correction Request Conversion Unit 14 Constraint parameter weighting section 15 Shift Generation Unit

Claims

1. a shift generation unit that generates work shifts; a shift presentation unit that presents the work shifts generated by the shift generation unit; an adjustment request acquisition unit that receives adjustment requests and their importance from the shift target person regarding the work shift presented by the shift presentation unit; a weighting unit that sets a weight for a setting value of a parameter corresponding to the correction request based on the content and importance of the correction request, The shift generation unit generates the work shift taking into account the weights of the parameter setting values ​​set by the weighting unit.

2. the correction request acquisition unit acquires a correction request in natural language; The automatic work shift generation system according to claim 1, further comprising a correction request conversion unit that converts the correction request into a setting value of any one of the parameters that are set in advance.

3. 3. The system for automatically generating work shifts according to claim 2, wherein the modification request conversion unit converts the importance of the modification request into a weight for the parameter setting value.

4. The weight is set by a numerical value set in multiple stages, and an upper limit is set on the total weight of the shift target person, The automatic work shift generation system according to claim 1 , wherein the weighting unit sets the weight so that the weight falls within an upper limit of the total weight of the shift recipients.

5. An automatic work shift generation method for generating a work shift by an automatic work shift generation system, Present the work shift generated by the shift generation unit; For the proposed work shift, we accept requests for modification from the shift subject and their importance, setting a weight for a parameter setting value corresponding to the correction request based on the content and importance of the correction request; The shift generation unit generates the work shift taking into consideration the weights of the setting values ​​of the parameters. How to automatically generate work shifts.

6. An automatic work shift generation program that automatically generates work shifts, Present the generated work shift, For the proposed work shift, the system accepts requests for corrections from the shift target and their importance, setting a weight for a parameter setting value corresponding to the modification request based on the content and importance of the modification request; The work shift is generated taking into account the weight of the set value of the parameter. Automatic work shift generation program.

Images