Information processing device, information processing method, and information processing program

The information processing device generates and combines assignment patterns based on constraint conditions to reduce the number of candidates, maintaining allocation plan quality and suppressing calculation scale increases, thus improving the final transport aircraft allocation plan.

JP2024172546APending Publication Date: 2024-12-12HITACHI LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2023090334
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-31
Publication Date
2024-12-12

AI Technical Summary

Technical Problem

Existing methods for creating transport aircraft allocation plans may exclude important allocation patterns, leading to a deterioration in the quality of the final allocation plan due to an increased number of candidates, making it difficult to determine the optimal solution.

Method used

An information processing device that generates multiple assignment patterns based on predetermined constraint conditions, combines these patterns to form a new allocation plan, and outputs a second allocation plan that adheres to these conditions, thereby reducing the number of candidates and preserving the quality of the final plan.

Benefits of technology

This approach suppresses the increase in calculation scale while ensuring that preferred allocation patterns are not excluded, resulting in a higher-quality final allocation plan.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024172546000001_ABST
    Figure 2024172546000001_ABST
Patent Text Reader

Abstract

To provide an information processing device, method, and program for suppressing an increase in calculation scale and suppressing the quality degradation of allocation plans.SOLUTION: Provided is an information processing device 10 for outputting a second allocation plan of personnel capable of carrying out a first allocation plan of a transport aircraft. The information processing device 10 generates on the basis of a first constraint condition a plurality of allocation patterns indicating whether or not personnel can be allocated to each of a plurality of sections defined by adjacent two departure and arrival points in the transport aircraft route of the first allocation plan, combines some of the application patterns to create a third allocation plan capable of executing the first allocation plan and constituting the basis of a second allocation plan, joins a plurality of allocation patterns together that are included in the third allocation plan on the basis of a second constraint condition to create a new allocation pattern, and executes a process of outputting the third allocation plan including the new allocation pattern as the second allocation plan. The new allocation pattern is such that personnel is allocated for the case where, in a plurality of sections, personnel is allocated in at least one of the joined allocation patterns.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] Implementing pre-determined aircraft allocation plans for public transportation such as buses, airplanes, and railroads To achieve this, it is necessary to determine an allocation pattern for allocating personnel to aircraft allocation plans.

[0003] In this case, first, multiple allocation pattern candidates are listed, and then the allocation of the transport aircraft is selected from among them. A quota that covers all flights included in the plan and satisfies the constraints imposed by laws, regulations, company rules, etc. The combination of this pattern is finally determined.

[0004] In this case, for example, if the number of flights included in the aircraft allocation plan increases, the allocation pattern The number of candidates can be so large that it can be difficult to determine the final allocation pattern. .

[0005] For example, in Non-Patent Document 1, the number of candidates for allocation patterns is reduced in order to reduce the scale of calculation. In order to reduce the number of allocation patterns, a technique is described in which some allocation patterns are excluded from candidates by a predetermined procedure. It is being done. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] B. Gopalakrishnan, EL Johnson, “Airline Crew Scheduling: State-of-the-Art”, Annals of Operations Research, vol.140, 2005, P.305-337 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in the technique disclosed in Non-Patent Document 1, the allocation plan that is finally created is preferably There is a possibility that allocation patterns that make an important contribution to the This may result in a poor quality allocation plan.

[0008] The present invention has been made in consideration of these problems, and is capable of suppressing an increase in the scale of calculation. To provide an information processing device capable of suppressing deterioration in the quality of the allocation plan finally obtained by the above. The purpose is to [Means for solving the problem]

[0009] In order to achieve the above object, one aspect of the present invention provides an information processing device having a computing device and a storage device. and outputting a second allocation plan of personnel capable of executing the first allocation plan of the transport aircraft. A processing device, the transportation in the first allocation plan based on a predetermined first constraint condition. The personnel are assigned to each of a number of sections determined by two adjacent departure and arrival points on the flight route. A plurality of allocation patterns indicating whether or not the target can be assigned are generated, and a part of the plurality of allocation patterns is combined. and create a third allocation plan that is the basis of the second allocation plan that can carry out the first allocation plan. and generating a plurality of the allocation patterns included in the third allocation plan based on a predetermined second constraint condition. The third allocation pattern is generated by combining the first and second allocation patterns into a new allocation pattern. A process of outputting the second allocation plan as the new allocation pattern is executed. In the process, the new allocation pattern is combined in each of the plurality of sections. When the personnel is assigned in at least one of the plurality of assignment patterns, , and the personnel are assigned to the information processing device.

[0010] Another invention for achieving the above object is an information processing device having a calculation device and a storage device. and outputting a second allocation plan of personnel capable of executing the first allocation plan of the transport aircraft. A processing device determines the operation of the transport aircraft in the first allocation plan based on a predetermined first constraint condition. The personnel are assigned to each of a plurality of sections defined by two adjacent starting and ending points on the route. A plurality of allocation patterns are generated, each of which indicates whether or not the plurality of allocation patterns can be used. and creating a third allocation plan that is the basis of the second allocation plan that can carry out the first allocation plan; Based on a predetermined second constraint condition, the plurality of allocation patterns included in the third allocation plan are combine the third allocation plan into a new allocation pattern, and include the new allocation pattern; The process of outputting the second allocation plan is executed, and the process of setting the new allocation pattern is executed. In the new allocation pattern, When the personnel is assigned in at least one of a plurality of assignment patterns, It is a method of processing information in which personnel are assigned.

[0011] Another invention for achieving the above object is an information processing device having a calculation device and a storage device. and outputting a second allocation plan of personnel capable of executing the first allocation plan of the transport aircraft. A processing device is provided for determining, based on a first predetermined constraint, the operation of the transport aircraft in the first allocation plan. The personnel are assigned to each of a plurality of sections defined by two adjacent starting and ending points on the route. A plurality of allocation patterns indicating whether or not the plurality of allocation patterns can be used are generated, and a part of the plurality of allocation patterns is combined. A third allocation plan is prepared based on the second allocation plan that can execute the first allocation plan. Based on a second predetermined constraint condition, the allocation patterns included in the third allocation plan are The third allocation plan including the new allocation pattern is then combined to form a new allocation pattern. and outputting the second allocation plan as the new allocation pattern. In the process, the new allocation pattern is a combination of When the personnel is assigned in at least one of the plurality of assignment patterns, In the case where the person is assigned to the information processing program, the person is assigned to the information processing program. The details will become clear from the description in this specification. Effect of the Invention

[0012] According to the present invention, the deterioration of the quality of the allocation plan finally obtained while suppressing the increase in the calculation scale is suppressed. It is possible to suppress the above. [Brief description of the drawings]

[0013] [Figure 1] 1 is a configuration diagram of an information processing device 10 according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating a hardware configuration of an information processing device 10 according to an embodiment. [Diagram 3] 11 is a flowchart showing an overview of how a personnel allocation plan is formulated. [Figure 4] FIG. 2 is a diagram illustrating a specific example of the transportation plan data 30a. [Diagram 5] FIG. 13 is a diagram illustrating a specific example of the candidate enumeration parameter 30b. [Figure 6] FIG. 11 is a diagram illustrating a specific example of the operating condition data 30c. [Figure 7]FIG. 13 is a diagram illustrating a specific example of an end point determination parameter 30d. [Figure 8] 11 is a flowchart illustrating an example of determining end points and classifying operating conditions. [Figure 9] 11 is a flowchart illustrating an example of determining end points and classifying operating conditions. [Figure 10] 11 is a directed graph for explaining an example of candidate enumeration by a candidate enumeration unit 60. [Figure 11] 11 is a diagram for explaining an example of candidate enumeration by a candidate enumeration unit 60. FIG. [Figure 12] FIG. 1 is a diagram illustrating an example of candidate enumeration according to a general method. [Figure 13] 13 is a flowchart illustrating details of a combining process and generation of output data. [Figure 14] 13 is a flowchart illustrating details of a combining process and generation of output data. [Figure 15] 13 is a flowchart illustrating details of a combining process and generation of output data. [Figure 16] 13 is a diagram for explaining a personnel allocation plan created by the combination processing unit 80. FIG. [Figure 17] 13 is a diagram for explaining details of a personnel allocation plan created by the combination processing unit 80. FIG. [Figure 18] 11 is a diagram for explaining an example of candidate enumeration by a candidate enumeration unit 60. FIG. [Figure 19] 1 is a directed graph illustrating an example of candidate enumeration using a general method. [Figure 20] FIG. 1 is a diagram illustrating an example of candidate enumeration according to a general method. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] == Implementation form == <<Overall device configuration>> FIG. 1 is a diagram showing an example of the configuration of an information processing device 10 according to the present embodiment. 0 is a device that outputs an allocation plan of personnel who can carry out the allocation plan of transport aircraft.

[0015] In this embodiment, an aircraft is used as an example of a transport aircraft. , transportation in other public transportation systems (bus routes, railways, etc.) that have aircraft allocation plans It is also possible to apply the present invention to a transmission (vehicle, etc.).

[0016] In the following explanation, the aircraft allocation plan corresponds to the "first allocation plan", and the personnel allocation plan corresponds to the "first allocation plan". This plan corresponds to the "Second Allocation Plan."

[0017] The information processing device 10 can be configured by one physical computer. The information processing device 10 may be implemented by a separate physical computer. good.

[0018] The information processing device 10 may be replaced by a computer system having one or more physical computers. It is also possible to run the system on other types of systems, such as on physical computing resources (e.g., cloud infrastructure). The system may be a system implemented in a cloud computing environment (e.g., a cloud computing system).

[0019] In FIG. 1, an information processing device 10 includes an input / output support unit 20, an input data storage unit 30, and an end point Determination and operation condition classification unit 40, constraint condition storage unit 50, candidate enumeration unit 60, optimization calculation unit 70 , a combination processing unit 80, and an output data storage unit 90.

[0020] The following provides an overview of each of these. For details, please refer to the following concrete examples. This will be explained later.

[0021] <Input / Output Support Unit 20> The input / output support unit 20 stores the received input data in the input data storage unit 30, which will be described later. The terminal point determination and operation condition classification unit 40 and the candidate enumeration unit 60, which will be described later in detail, are also output. To exert effort.

[0022] The input data here are transportation plan data 30a, candidate enumeration parameters 30b, These are the operating condition data 30c and the end point determination parameters 30d. For convenience, the operating condition classifying unit 40 is referred to as the "classifying unit 40."

[0023] Furthermore, the input / output support unit 20 includes an input data storage unit 30 and an output data storage unit 90. It is possible.

[0024] The input data storage unit 30 stores the import and export data as input information for creating a draft allocation plan and for creating the draft allocation plan. The transport plan data 30a, the candidate enumeration parameters 30b, the operation condition data 30c, and the end point determination parameters The input data is stored in the form of the parameter 30d. The details of the input data will be described later.

[0025] The output data storage unit 90 stores the data created based on the aircraft allocation plan and various constraints. A proposed personnel allocation plan (details will be described later) is stored as output data.

[0026] <Classification section 40> The classification unit 40 first selects the candidate enumeration parameters 30b, The operating condition data 30c and the end point determination parameters 30d are acquired.

[0027] The classification unit 40 classifies the end points based on the operation condition data 30c and the end point determination parameters 30d. The "end point" is determined and added to the candidate enumeration constraint (details will be described later). Among the departure and arrival points, these are departure and arrival points where personnel can start or finish work.

[0028] The classification unit 40 then determines whether the candidate enumeration parameter 30b and the end point determination parameter 30d are Based on this, the operation condition data 30c is classified into candidate enumeration constraint conditions or combination constraint conditions.

[0029] The details will be described later. The "candidate enumeration constraint" is a condition in which the candidate enumeration unit 60 selects candidates for the allocation pattern. The "join constraint" is a constraint used when enumerating. This is a constraint used when combining allocation patterns to generate a new allocation pattern.

[0030] In the following description, the candidate enumeration constraint condition corresponds to the "predetermined first constraint condition" and The matching constraint condition corresponds to the "predetermined second constraint condition."

[0031] The classification unit 40 outputs the candidate enumeration constraints and the join prior constraints to the constraint storage unit 50 .

[0032] <Constraint condition storage section 50> The constraint condition storage unit 50 includes a candidate enumeration constraint condition storage unit 50a and a combination constraint condition storage unit 50b. Includes.

[0033] The candidate enumeration constraint condition storage unit 50a stores the above-mentioned candidate enumeration constraint conditions. The storage unit 50b stores the above-mentioned join constraint conditions.

[0034] <Candidate Enumeration Section 60> The candidate enumeration unit 60 generates a plurality of allocation patterns based on the candidate enumeration constraint conditions. This pattern is a combination of two adjacent departure and arrival points on the route of a transport aircraft in the transport aircraft allocation plan. Personnel are assigned to each of the multiple sections (hereinafter also referred to as "nodes") defined by This is a pattern that indicates whether or not

[0035] The candidate enumeration unit 60 determines the candidate enumeration constraint conditions and the candidate enumeration data acquired from the input data storage unit 30. Based on the parameters 30b and the transportation plan data 30a, candidates for the allocation pattern are listed. Then, the candidate enumeration unit 60 outputs the enumerated allocation pattern candidates to the optimization calculation unit 70. do.

[0036] <Optimization Calculation Unit 70> The optimization calculation unit 70 selects a part of the multiple allocation patterns generated by the candidate enumeration unit 60. The allocation plan is then combined to create a basis for the allocation plan of personnel who can carry out the aircraft allocation plan. The allocation plan on which the allocation plan here is based corresponds to the "third allocation plan."

[0037] Here, the optimization calculation unit 70 creates an allocation plan by a predetermined optimization calculation. The procedure will be described later with specific examples.

[0038] The optimization calculation unit 70 is a mathematical optimization solver that runs on a CPU mounted on a conventional computer. Alternatively, a non-von Neumann-type computer such as a quantum annealing machine or a CMOS annealing machine could be used. Implemented on dedicated ASICs such as FPGAs and GPUs installed in conventional computers. Computational techniques may be used in the optimization calculations.

[0039] The optimization calculation unit 70 outputs the obtained allocation plan to the joint processing unit 80. The image is selected from the multiple allocation patterns enumerated by the candidate enumeration unit 60 by the optimization calculation unit 70. The allocation pattern includes a plurality of allocation patterns selected by

[0040] <Combination Processing Unit 80> The joining processing unit 80 calculates the allocation created by the optimization calculation unit 70 based on the joining constraint conditions. A plurality of allocation patterns included in the plan are combined to generate a new allocation pattern.

[0041] Here, "combining a plurality of allocation patterns" means covering all of the plurality of allocation patterns. This means that the allocation pattern for one person is generated based on the allocation pattern. A detailed explanation will be given later with specific examples.

[0042] <Output data storage unit 90> The output data storage unit 90 stores the new allocation pattern generated by the combination processing unit 80. The allocation plan including the above is output as the final personnel allocation plan (corresponding to the "second allocation plan").

[0043] <<Hardware configuration>> The hardware configuration of the information processing device 10 of this embodiment will be described. The form is a hardware configuration diagram of a computer that constitutes an information processing device.

[0044] The information processing device 10 includes a calculation device 100, a main memory device 101, an auxiliary memory device 102, and a transmission / reception device. The system includes a receiving device 103, an output device 104, and an input device 105.

[0045] The arithmetic device 100 includes, for example, a CPU (Central Processing Unit), an MPU (Micro Processor), processing unit), GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit), AI (Artifi It is composed of CIS (Cyber ​​Intelligence) chips, etc.

[0046] The main memory device 101 is a device that the arithmetic device 100 uses when executing a program. For example, ROM (Read Only Memory), RAM (Random Access Memory), non-volatile memory The information processing device 10 may include a non-volatile memory (NVRAM), etc. The various functions are implemented by the arithmetic device 100 by using the programs stored (in the auxiliary storage device 102). This is realized by loading the programs and data into the main memory device 101 and executing them.

[0047] The auxiliary storage device 102 is a device for storing programs and data, and is, for example, an SSD ( Solid State Drive, Hard Disk Drive, Optical Storage Device (CD (Compact Disc ), DVD (Digital Versatile Disc), storage systems, IC cards, SD cards Read / write devices for non-transitory recording media such as flash memory and optical recording media, and non-uniform The auxiliary storage device 102 can be configured as a temporary storage area, etc. via the device or the transmitting / receiving device 103, a non-transitory recording medium or another device equipped with a non-transitory storage device Programs and data can be read from the information processing device. The programs and data stored in the main memory device 101 are read as needed.

[0048] The transmitting / receiving device 103 is a device that realizes communication with other devices. is a wired or wireless communication system that realizes communication between other devices via a network. A communication interface, such as a NIC (Network Interface Card), a wireless communication module, etc. modules, USB modules, etc.

[0049] The output device 104 and the input device 105 are used for interactive processing with the user (accepting information, A user interface is constructed to realize the functions of the user (such as providing information).

[0050] <<Flow of creating a personnel allocation plan>> Figure 3 is a flow chart showing the outline of how to create a personnel allocation plan. When creating a personnel allocation plan for an aircraft route, refer to the flowchart in FIG. This article explains:

[0051] In this embodiment, it is necessary to devise a personnel allocation plan in accordance with the revision of the aircraft allocation plan, etc. It can be used when an employee is absent from work. It can also be used for simulations when revising personnel work shifts in accordance with changes in the law. This can be done.

[0052] In this embodiment, an aircraft allocation plan is set up to allocate personnel to predetermined airline routes. It is assumed that a pattern for assigning a number of characters (hereinafter, simply referred to as an "assignment pattern") is created.

[0053] <Processing Overview> First, referring to the flowchart shown in FIG. 3, the information processing device 10 calculates the personnel allocation plan. The outline of the process steps up to outputting the proposal will be explained. Each step (Step S1 to Step S Details of 6) will be explained later in the supplementary section.

[0054] First, in step S1, the information processing device 10 reads input data. In a concrete example, the input data reading is related to the original plan sent from an external device such as a PC. The processor of the input / output support unit 20 reads various pieces of information to It will be realized.

[0055] The input data is stored in the I / O support unit 20 as shown in FIG. In the input data storage unit 30 such as a DD, information is classified (divided) according to the type of information. It is recommended that the information be stored in

[0056] Next, in step S2, the classification unit 40 stores in the input data storage unit 30 shown in FIG. The stored candidate enumeration parameters 30b, operation condition data 30c and end point determination parameters 3 Get 0d.

[0057] The classification unit 40 checks the consistency between the end point determination parameters 30d and the operation condition data 30c, In step S3, the end points to be used for candidate enumeration are determined and added to the candidate enumeration constraints. The process in step S2 will be described in more detail below with reference to FIG.

[0058] The classification unit 40 performs a classification based on the obtained candidate enumeration parameters 30b and end point determination parameters 30d. Based on this, the operation conditions included in the operation condition data 30c are classified into candidate enumeration constraint conditions or combination constraint conditions. and store them in the constraint condition storage unit 50.

[0059] Next, in step S3, the candidate enumeration unit 60 selects the candidate The transport plan data 30a and the candidate enumeration parameters 30b, the candidate sequence in the constraint condition storage unit 50, The system obtains the enumeration constraints and uses this information to enumerate candidates for allocation patterns.

[0060] Next, in step S4, the optimization calculation unit 70 performs optimization calculation to obtain the step S It is possible to execute a transport aircraft allocation plan by combining some of the allocation patterns listed in 3. Develop an allocation plan that will serve as the basis for personnel allocation planning.

[0061] The optimization calculation unit 70 minimizes the cost calculated by a predetermined calculation method, as will be described later in detail. Create an allocation plan to ensure that

[0062] Next, in step S5, the combining unit 80 combines the allocation calculation data created in step S4. A new allocation pattern is generated by combining the allocation patterns included in the image based on the combining constraints. Generate a turn.

[0063] Then, the combining processing unit 80 outputs the personnel allocation plan including the new allocation pattern as output data. The output data is then output to the output data storage unit 90 of the input / output support unit 20.

[0064] Finally, in step S6, the information processing device 10 executes the process of By the above process, the allocation plan proposal in the output data storage unit 90 is output to an external device such as a PC, The process ends. The above is an overview of the process executed by the information processing device 10.

[0065] <<Flow details>> The process of each step in the flowchart shown in FIG. 3 will be described in detail below. do.

[0066] <Step S1: Reading input data> The input data read in step S1 is stored in the input data storage unit 30 as shown in FIG. The transportation plan data 30a, the candidate enumeration parameters 30b, the operation condition data 30c, and the end point The decision parameters are 30d and 30e, respectively. First, each of these will be explained with a specific example.

[0067] [Transportation planning data 30a] FIG. 4 is a diagram illustrating a specific example of the transportation plan data 30a. 30a is an aircraft for which a personnel allocation plan is proposed and output by the information processing device 10. This is information indicating an allocation plan, for example, timetable information.

[0068] As shown in FIG. 4, a specific example of the transportation plan data 30a is a train schedule of a queue type. In the example shown in Figure 4, the information of the flights assigned to a particular day is recorded. There are.

[0069] As shown in FIG. 4, the transportation plan data 30a includes the “identification ID” and “operation date” of each flight. ID", "Departure point", "Arrival point", "Departure time", "Arrival time", "Origin point and destination The "Land" item is shown.

[0070] For example, in the case of a transportation plan data with an ID of 1, the operation ID is "101" and the departure point is "B ", arrival point is "A", departure time is "11:30", arrival time is "13:00", starting point is "B" and the final destination is "A".

[0071] [Candidate enumeration parameter 30b] FIG. 5 is a diagram for explaining a specific example of the candidate enumeration parameter 30b. The parameter 30b is a parameter that controls the candidate enumeration unit 60 of the information processing device 10, and is can be set.

[0072] As shown in FIG. 5, the candidate enumeration parameters 30b include “candidate enumeration parameter ID,” The items "Parameter", "Type" and "Value" of the parameter are shown.

[0073] In this example, the candidate enumeration parameter ID=1 indicates that, for one allocation pattern candidate, There is a limit on the number of hours that employees can work, with the maximum being 6 hours. In 60, assignment patterns with working hours exceeding 6 hours are excluded from the candidates. It becomes.

[0074] Similarly, in accordance with the candidate enumeration parameter ID=2, the candidate enumeration unit 60 determines that the working hours are 1 p.m. Any allocation pattern with an interval of less than 30 minutes will be excluded from the candidates. From the enumeration parameter ID=3, the candidate enumeration unit 60 enumerates the candidates of the allocation pattern. The upper limit is 10,000.

[0075] When candidates for allocation patterns are enumerated by a general method, the candidate sequence as in this embodiment is Therefore, according to the general method, if the working time exceeds 6 hours, The list of allocation pattern candidates includes those with working hours of less than 1 hour and 30 minutes. It can be mentioned.

[0076] Furthermore, when the number of candidates for the allocation pattern to be enumerated exceeds 10,000, the general method It is possible that this could happen.

[0077] Therefore, the candidate enumeration unit 60 enumerates the candidates by using the above-described candidate enumeration parameters 30b. It is possible to suppress an increase in the number of candidates for allocation patterns to be displayed. The candidate enumeration parameters are merely examples and are not limited to these parameters.

[0078] [Operation condition data 30c] 5 is a diagram illustrating a specific example of the operating condition data 30c. 30c, when formulating a personnel allocation plan, shall consider the internal regulations and labor standards that the final allocation plan must comply with. This data contains laws and regulations.

[0079] As shown in FIG. 5, the operation condition data 30c includes "operation condition ID", "condition content", The items "Parameter" ("X" to "Z"), "Type" and "Classification Condition" are shown.

[0080] For example, operation condition ID=1 is a time-related condition (from "Type"), and is based on laws and regulations. If working hours X is 6 hours or more, break time Y must be 45 minutes or more. It should be noted that the conditions shown here are merely examples and are not intended to be limiting.

[0081] In addition, if there is a threshold value in the operation conditions, the threshold value is stored in the classification conditions. For example, in the operation condition IDs 1 and 2, the classification condition is to obtain a predetermined rest time. The threshold for the number of hours that must be worked is stored.

[0082] In addition, operation condition ID=4 is a condition related to location (from "Type"), and is a place where rest is possible. X indicates that the starting and ending point is B.

[0083] [End point determination parameter 30d] FIG. 7 is a diagram for explaining a specific example of the end point determination parameter 30d. The parameter 30d is a data item that describes the start and end points set by the user. It is data.

[0084] As shown in FIG. 7, the end point determination parameters 30d include “end point candidate ID”, “end point candidate location”, The items "Point", "Start end", "End end" and "ON / OFF" are shown.

[0085] The endpoint candidate ID is an identifier for identifying the starting and ending points that are candidates for endpoints. The points are the names of the starting and ending points that are candidates for end points.

[0086] The start point (end point) is a value that indicates whether the start point (end point) can be the start point (end point). This is a value set by the user. If it is possible to use the starting point and the ending point as the starting point (end point), set it to "1 " If not possible, "0" is stored.

[0087] ON / OFF indicates whether to use the start (end) information described above when enumerating candidates in step S3. This is a value that indicates whether the start (end) of the If it is used, "1" is stored; if not, "0" is stored.

[0088] For example, the endpoint candidate ID=3 is the endpoint candidate C, which can be the starting point, but not the end point. In addition, the endpoint candidate ID=3 is not included in the candidate enumeration in step S3. This indicates that the information is used as the aforementioned start (end) information when

[0089] <<Step S2: Determining end points and classifying operating conditions>> 8 and 9 are flowcharts illustrating an example of determining end points and classifying operating conditions. The flowchart shown in FIG. 8 corresponds to the subroutine of step S2 described above in FIG. The flowchart shown in FIG. 9 corresponds to a subroutine of the flowchart shown in FIG. Corresponds to.

[0090] After the input data is read in step S1 (FIG. 3), step S21 in FIG. In the classification unit 40, the candidate enumeration parameters 30b are stored in the input data storage unit 30 (FIG. 1). (Fig. 5), operating condition data 30c (Fig. 6) and end point determination parameters 30d (Fig. 7) are obtained. do.

[0091] The classification unit 40 classifies all the operating conditions included in the acquired operating condition data 30c (FIG. 6). If the type is "time", the "constraint condition related to time" is displayed (step S22). " (step S23), and if the type is not time, it is classified as "location constraint condition" ( Step S24).

[0092] For example, if the operation condition ID is 1, the type is "time", so "Time constraints" If the operation condition ID is 4, the type is classified as "event Since it is a "location", it is classified as a "location constraint". If there is one, the type is blank, so it is classified as a "location constraint" (step S2 4).

[0093] The classification unit 40 performs a process of confirming the types of all the operating conditions included in the operating condition data 30c. After the process is completed, all the operation conditions are time constraints or location constraints. The items will be classified as follows.

[0094] Next, the classification unit 40 acquires the time-related constraint conditions classified in step S23 ( Step S25). Then, the classification unit 40 classifies all the operations included in the time constraint. The classification unit 40 then checks whether a threshold exists in the classification condition. The threshold and the upper limit time in the candidate enumeration constraint are compared (step S26). .

[0095] In Figure 6, the classification criteria do not specify a threshold (the field is blank), or the candidate enumeration constraints If the upper limit time is equal to or greater than the upper limit time, the classification unit 40 adds the operation condition to the combination constraint condition (step S27).

[0096] For example, if the operation condition ID is 4, the result is that no threshold is specified in the classification condition. The matching constraint is added (step S27).

[0097] On the other hand, the classification unit 40 detects whether a threshold value is specified in the classification condition and whether the candidate enumeration time limit is less than the upper limit. If so, the classification unit 40 adds the operation condition to the candidate enumeration constraint conditions (step S28).

[0098] For example, if the operation condition ID is 1, the threshold value (X=6 hours) is specified in the classification condition. Since the time is less than the upper limit of candidate enumeration time (candidate enumeration parameter ID = 1 in Figure 5), the combination control The condition is added to the reservation conditions (step S28).

[0099] Furthermore, the classification unit 40 acquires constraints related to the places classified in S24 (step S2 8). Then, the classification unit 40 calculates the location-related constraints and the end point determination parameters 30d. Using this, the end point determination process (step S29) is performed.

[0100] FIG. 9 is a flowchart illustrating the details of the end point determination process (step S29) of FIG. First, in step S291, the classification unit 40 calculates the end point determination parameter 30d and the In step S24, constraints on the classified places are obtained.

[0101] As shown in FIG. 7, the end point determination parameters 30d include end point candidate IDs, end point candidate locations, and starting points. It consists of the items end, termination and ON / OFF.

[0102] The classification unit 40 checks all the end point candidates included in the end point determination parameters 30d and The classification unit 40 judges whether the ON / OFF of the signal is 1 (step S292). If it is 1, the values ​​of the start and end points are checked (step S293).

[0103] For example, for the end point candidates ID=2 and 3 in FIG. 7, ON / OFF is 1, so the In step S293, the values ​​of the start and end points are confirmed. For the end point candidate ID=1, Since ON / OFF is 0, it will not be taken into account in subsequent processing.

[0104] In step S293, if the start point is 1 and the end point is 0, the end point candidate point is available for work. If it is a possible departure / arrival point, add it to the end point ( Step S298).

[0105] For example, for the end point candidate ID=3 in Figure 7, the start point is 1 and the end point is 0, so If destination point C is a possible departure or arrival point for commuting, it is added to the end points (step S298).

[0106] In step S293, if the start point is 0 and the end point is 1, the end point candidate point is a possible exit point. If it is a departure / arrival point where work can be stopped, it is added to the end point ( Step S298).

[0107] The process in step S293 is performed when a departure / arrival point where attendance is not possible is specified only as the start point. In the case where a non-workable start / end point is specified only as the end point, This is to prevent the allocation plan that cannot be implemented due to the inability to combine. do.

[0108] In addition, in S293, if the start point and the end point are both 1, the end point candidate point is a rest possible point. If it is a departure or arrival point where a rest is possible, add it to the end point ( Step S298).

[0109] Through the above process, the candidate enumeration constraints and the join constraints are determined. The candidate enumeration constraints include a predetermined upper or lower limit of the working hours of the personnel. The list constraint includes an upper limit on the number of multiple allocation patterns. This includes the working conditions set out in the Employment Act.

[0110] <<Step S3: Candidate enumeration>> 10 to 15 are diagrams for explaining an example of candidate enumeration by the candidate enumeration unit 60. After step S2 (FIG. 3) is executed, the unit 60 generates a connection table for each service as shown in FIG. Then, data is created that converts the relationship into a directed graph (hereinafter, simply referred to as a "directed graph").

[0111] The directed graph shown in FIG. 10 is a diagram of the transportation plan data 3 stored in the input data storage unit 30. 0a (Figure 4). In this directed graph, the circles represent nodes. In addition, "- (hyphen) 1" is assigned to a specific day. This example shows the allocation plan for the assigned aircraft. It is not a cross-border allocation plan.

[0112] The candidate enumeration unit 60 enumerates the candidate enumeration parameters 30b (FIG. 5), the candidates in the constraint condition storage unit 50, Obtain the enumeration constraints, and based on this information, follow the links in the directed graph in Figure 10. Allocation pattern candidates are enumerated and added to the allocation candidates.

[0113] FIG. 11 is an example of a candidate table including a plurality of listed allocation pattern candidates. The allocation patterns in the candidate table are assigned to the end points stored in the candidate enumeration constraints. In this example, five allocation pattern candidates (j=1 to 5) are listed.

[0114] As described above, the allocation pattern is determined by the number of adjacent aircraft on the route of the aircraft allocation plan. Whether personnel are assigned to each of the multiple sections (nodes) that are determined by the two starting and ending points This is a pattern that indicates whether

[0115] In this example, the candidate allocation pattern j (1 to 5) is the number of employees assigned to node i. If the device is not assigned, the value is 1; if not, the value is 0.

[0116] For example, for candidate j=1, only nodes i=1 and 2 will be assigned personnel. Furthermore, in this case, since the "affiliation" is point B, the personnel are at point B. This indicates that work begins and ends at the designated time.

[0117] In addition, in this case, the time of arrival, time of departure, break time and working time (working hours) of the personnel will be These indicate 11:00, 14:30, 0 hours, and 3 hours and 30 minutes, respectively.

[0118] In addition, FIG. 12 shows the allocation based only on the operation condition data 30c without using the candidate enumeration constraint condition. A candidate table consisting of candidates for allocation patterns when this pattern is listed is shown.

[0119] In the candidate table of FIG. 12, all of the listed allocation patterns are In this case, nine allocation pattern candidates (j=1 to 9) are listed. In comparison with the present embodiment (FIG. 11), the number of candidates for the listed allocation patterns is greater.

[0120] For example, among the nine allocation pattern candidates (j=1 to 9) in the candidate table of FIG. For supplementary j=5, the working time (hours) is 11 hours and 30 minutes.

[0121] However, in this embodiment, the candidate enumeration constraint is the candidate enumeration parameter shown in FIG. 30b, assignment patterns with working hours exceeding 6 hours and 30 minutes are excluded from the candidates. Therefore, an allocation pattern such as candidate j=5 in the candidate table of FIG. Not at all.

[0122] Therefore, according to the process of the candidate enumeration unit 60 of this embodiment, the candidates of the allocation pattern to be enumerated are It is possible to suppress an increase in the number of

[0123] <<Step S4: Optimization calculation>> In step S4, the optimization calculation unit 70 calculates a personnel allocation plan using a set covering problem. In a general set covering problem, for example, a plan like the data in the work candidate table in Figure 12 is created. Which combination of candidate columns j is selected is determined by the binary variable x j is determined.

[0124] Specifically, when selecting a candidate for allocation pattern candidate j, x j =1, if not selected x j = 0. x j Only the candidates for allocation patterns where =1 are selected and included in the allocation plan. In the following description, the term "allocation pattern candidate" is simply used as "candidate". It may be referred to as "supplement."

[0125] A in the following formula 1 ij In the example of Figure 12, node i is a row, and the candidates for the allocation pattern are It is a binary matrix of 0 / 1 with column j being the complement.

[0126]

number

[0127] The allocation pattern is associated with a cost calculated using a predetermined calculation method. When determining the combination of j that satisfies 1, the cost C shown in Figure 12 is used as the evaluation value. j Using The cost calculated by the above is minimized (the following equation).

[0128]

number

[0129] In other words, the personnel allocation plan is j The cost is calculated so that the sum of C j is an index representing the desirability of the allocation pattern candidate j. j The user It can be set based on any calculation method. Cost C j As an example of the allocation pattern, It can be the total working hours of candidate j.

[0130] In this case, by performing optimization using the objective function of Equation 2 and the constraint condition of Equation 1, all One or more personnel are assigned to node i, and the total working time of the personnel is minimized. You can select a combination of candidates for this pattern.

[0131] Another example of cost is the desirability of the user's desired working pattern (e.g., When enumerating candidates, the degree of return flight (such as returning on a flight using the same aircraft) is calculated and C j Store in C j Maximize (or -C j It may be possible to minimize the

[0132] As another example of the cost, the labor cost of personnel for each work shift may be used as the cost. .

[0133] As another example of cost, the number of allocation patterns included in the personnel allocation plan is In this case, if you want to minimize the cost, you can use the following instead of the objective function shown in Equation 2: Then, using the objective function of Equation 3, we find candidates for allocation patterns that cover the aircraft allocation plan. It is also possible to extract only the combination of

[0134]

number

[0135] In addition, the optimization calculation unit 70 uses a quantum annealing machine or a CMOS annealing machine. When configuring hardware that operates on an interaction model, the input interaction model can be written, for example, as follows:

[0136]

number

[0137] <<Step S5: Combining process>> 13, 14 and 15 show the combining process in step S5 (FIG. 3) and the generation of output data. 10 is a flowchart for explaining the details of the combination process in step S5 of the present embodiment. The management unit 80 performs the join process using a greedy algorithm.

[0138] The flowchart shown in FIG. 13 corresponds to the subroutine of step S5 described above in FIG. 14 and 15 are subroutines of the flowchart shown in FIG. Each corresponds to a different tool.

[0139] After the optimization calculation in step S4 (FIG. 3) is performed, in step S51 shown in FIG. The processor of the information processing device 10 is included in the allocation plan created by the optimization calculation unit 70. Obtain a set of allocation patterns (Figure 11) and assign an index to the allocation patterns included in the set. Shake (step S51).

[0140] In the greedy method used in this embodiment, whether two allocation patterns can be combined is determined. To make the determination, a double loop is used.

[0141] In step S52 of the first loop, the combining unit 80 determines whether the allocation pattern candidate i is a It is determined whether the candidate has already been combined with the candidate for the allocation pattern. An allocation pattern that has already been bound is called an "initialized" allocation pattern.

[0142] If the allocation pattern candidate i has not been initialized, the combining unit 80 The process proceeds to a second loop to search for an allocation pattern that can be combined with complement i.

[0143] In step S53, the index is set to 0 to avoid the allocation pattern i being combined with itself. In the second loop, if the index is different from i, the allocation pattern It is determined whether the candidate j has been initialized (step S54). If the candidate j has not been initialized, In S55, a process for determining whether the allocation pattern candidates i and j are combined is performed.

[0144] Details of step S55 will be described with reference to FIG. 14. First, the combining processing unit 80 Allocation pattern candidates i and j are obtained (step S551).

[0145] Next, the combining unit 80 finds a candidate i of the allocation pattern whose end point coincides with the beginning point of the candidate j. It is determined whether or not the number of the items is equal to the number of the items (step S552). If the number of the items is equal to the number of the items, the process proceeds to step S56. Proceed to.

[0146] For example, in the case of the pair of candidates j=1 and 5 in the allocation pattern of FIG. 11, the terminal of candidate 1 is the departure and arrival The starting point of candidate 5 is point B, and it is determined that the starting point of candidate 5 matches point B.

[0147] If it is determined in step S552 that the end of candidate i coincides with the beginning of candidate j, The combination processing unit 80 determines whether the work end time of the candidate i is before the work start time of the candidate j. If the determination here is negative, the process proceeds to step S56.

[0148] For example, in the case of a pair of candidates j=1 and 5 in the allocation pattern of FIG. 11, at the end of the shift of candidate 1, The time (end of work time) is 14:30, and the start time (start of work time) of Candidate 5 is 15:30. Candidate 1 finishes work before Candidate 5 starts work.

[0149] In step S553, it is determined that the work end time of candidate i is before the work start time of candidate j. Then, the combination processing unit 80 calculates the sum of the working hours of the candidate i and the working hours of the candidate j. Next, the combining unit 80 calculates the sum of the rest time of the candidate i and the rest time of the candidate j (step (P S554).

[0150] For example, in the case of the pair of candidates j=1 and 5 in the allocation pattern in FIG. 11, the commitment time (working time ) is 10 hours and 30 minutes, and the sum of the break times is 1 hour.

[0151] Next, in step S555, the combining unit 80 calculates the It is judged whether the sum of the working hours is less than the upper limit of working hours. Proceed to step S56.

[0152] For example, in the case of the pair of candidates j=1 and 5 in the allocation pattern in FIG. 11, The sum of the time is 10 hours and 30 minutes. In this case, according to the flight condition ID=13 in Figure 6, the upper limit constraint The time (maximum time) is less than 15 hours.

[0153] If it is determined in step S555 that the sum of the working hours is less than the upper limit of working hours, The combination processing unit 80 calculates the break time to be obtained from the sum of the working hours based on the combination constraint conditions. (Step S556). If the result is NO, the process proceeds to Step S56.

[0154] For example, in the case of the pair of candidates j=1 and 5 in the allocation pattern in FIG. 11, The sum of the time is 10 hours and 30 minutes. In this case, according to the flight condition ID=2 in Figure 6, The break time is at least 60 minutes.

[0155] Next, in step S557, the combining unit 80 calculates the It is determined whether the sum of the rest periods is equal to or longer than the rest period that should be taken. If so, the process proceeds to step S56.

[0156] For example, in the case of the pair of candidates j=1 and 5 in the allocation pattern of FIG. 11, In this case, the total time between the flight and the rest period is 1 hour. Above.

[0157] If it is determined in step S557 that the sum of the rest periods is equal to or greater than the rest period to be taken, The combination processor 80 determines that the candidates i and j can be combined, and the process proceeds to step S56.

[0158] Returning to FIG. 13, the combining unit 80 executes a combining determination process in step S56 to determine whether combining is possible. If it is determined that the combination is possible, the process proceeds to step S57 for combination processing. If it is determined that the combination is impossible, the process proceeds to the next allocation. Proceed to the determination of pattern j+1.

[0159] The combining process unit 80 completes the combining judgment process and combining process for all pairs of allocation patterns. When the process is completed, the personnel allocation plan is outputted from the new allocation pattern combined in step S57. Create it as a data.

[0160] Details of step S57 will be described with reference to FIG. 15. First, the combining processing unit 80 Allocation pattern candidates i and j are obtained (step S571).

[0161] Next, the combination processing unit 80 combines the working hours of the allocation pattern candidate i and the working hours of the candidate j. Furthermore, the combining unit 80 calculates the sum of the rest time of the candidate i and the rest time of the candidate j. Calculate (step S572).

[0162] For example, in the case of the pair of candidates j=1 and 5 in the allocation pattern in FIG. 11, the commitment time (working time ) is 10 hours and 30 minutes, and the sum of the break times is 1 hour.

[0163] Next, the combining unit 80 determines based on the combining constraint condition that the end of the candidate i of the allocation pattern is It is determined whether or not the departure and arrival point is a place where a rest can be taken (step S573). Then, the process proceeds to step S575.

[0164] For example, if the candidate j=1 of the allocation pattern in FIG. 11 is the starting point B, the end point is the starting point B in FIG. According to the operation condition ID=5, departure and arrival point B is a departure and arrival point where rest can be taken.

[0165] If it is determined in step S573 that the departure or arrival point is a place where a rest can be performed, the calculation is performed in step S572. The time from the end of work of candidate i to the start of work of candidate j for the sum of the break times The result is used as the new rest time (step S574).

[0166] For example, in the case of the pair of candidates j=1 and 5 in the allocation pattern of FIG. 11, The sum of the time between the end of Candidate 1 and the start of Candidate 5 is 1 hour. Adding the 1 hour break, the new break time is 2 hours.

[0167] Next, the combination processing unit 80 combines the candidate i with the candidate j to obtain a candidate for a new allocation pattern. i (step S575).

[0168] New allocation pattern candidates are combined in each of multiple sections (nodes). If personnel are assigned in at least one of a plurality of assignment patterns, the personnel are assigned. This is the pattern that is assigned.

[0169] In other words, in the allocation pattern candidates (j=1 to 5) shown in FIG. Combining candidates of nodes means that at least one of them is combined at each node i. If the allocation pattern is "1", the candidate for the new allocation pattern is "1".

[0170] Also, for each node i, if all the allocation patterns are “0”, a new The candidate for the allocation pattern is "0".

[0171] For example, if the pair of candidates j=1 and 5 in the allocation pattern of FIG. 11 is combined, the new allocation The pattern is that nodes 1, 2, 8 and 9 are "1" and other nodes are 0.

[0172] In addition, the candidate i enumerated in the candidate enumeration in step S3 (FIG. 3) and the candidate i enumerated in step S575 In order to distinguish between the candidate i and the candidate j that is combined, the latter is called "new candidate i" etc. It is called.

[0173] By this combination process, the working hours of the new candidate i are calculated in step S572. The break time of the new candidate i is set as the working time in step S572 or step S The break time is calculated as 574.

[0174] Furthermore, the end of the new candidate i becomes the end of the candidate j, and the end of work time of the new candidate j becomes the end of work time of the new candidate i. The time has come.

[0175] Next, the combination processing unit 80 sets the allocation pattern candidate j as an initialized candidate (step For example, in the case of a pair of candidates j=1 and 5 in the allocation pattern of FIG. 11, candidate 5 is It will be initialized.

[0176] Returning to FIG. 13, the combining unit 80 outputs the output data to the output data storage unit 90 (FIG. 1). (step S58).

[0177] 16 and 17 illustrate a personnel allocation plan 900 created by the combined processing unit 80. FIG. 16 shows data outlining an allocation plan 900, and FIG. 17 shows data outlining the allocation plan 900. Detailed information on each of the multiple allocation patterns 900a, 900b, 900c, etc. included The data includes:

[0178] The final personnel allocation plan shown in these figures includes multiple allocation patterns. This pattern is selected from the allocation patterns selected by the optimization calculation in step S4. The allocation pattern initialized by the combining process in step S5 is eliminated, and the allocation pattern is then It is composed of new allocation patterns generated by the combining process.

[0179] For example, the allocation pattern 900a included in the allocation plan shown in FIG. The working hours are 12 hours.

[0180] The candidate enumeration constraints used in the candidate enumeration in step S3 are shown in FIG. Based on the candidate enumeration parameter 30b. Therefore, in the candidate enumeration of step S3, ,Allocation patterns with working hours exceeding 6 hours are excluded.

[0181] Therefore, the allocation pattern 900a included in the allocation plan 900 shown in FIG. , they are not enumerated in the candidate enumeration in step S3.

[0182] However, due to the combining process in step S5, the allocation pattern that does not satisfy the candidate enumeration constraint condition is The allocation pattern 900a satisfies the join constraints. Therefore, we will comply with legal regulations and internal company rules.

[0183] Through the above processing, the information processing device 10 creates and outputs a personnel allocation plan. Such processing reduces the deterioration of the quality of the final allocation plan while suppressing the increase in the computation scale. It is possible to suppress the above.

[0184] == Variations == The information processing device 10 of this embodiment can be expanded to cases where personnel are on overnight shifts. Figure 18 shows an example of a directed graph that takes overnight shifts into account. In the case where there is a separate overnight shift, the number of nodes is compared to the case where there is no overnight shift. This doubles the amount.

[0185] In this case, if all allocation patterns were to be enumerated without using the candidate enumeration constraint, the number of allocation patterns on the first day would be For the same number of allocation pattern candidates (9 patterns) as in the candidate table shown in Figure 12, For the second day only, the same number of allocation pattern candidates as in the candidate table shown in FIG. There are 9 possible combinations.

[0186] Furthermore, there are six possible allocation patterns for overnight shifts ({ {103-1,104-1}+{101-2,102-2}, {103-1,105-1} +{101-2,102-2}, {108-1,109―1}+{101-2,102- 2}, {103-1,104-1}+{106-2,107-2}, {103-1,10 5―1}+{106-2,107-2}, {108-1,109―1}+{106-2, 107-2}).

[0187] Therefore, if all allocation patterns are enumerated without using the candidate enumeration constraint, the total number of allocation patterns shown in FIG. In this section, 24 possible allocation patterns will be listed.

[0188] On the other hand, when allocation patterns are enumerated using the candidate enumeration constraint as in the above embodiment, If you list the allocation options for two days, there are 12 options. In other words, if there is an overnight shift, Even if the method is used, the number of candidates for the allocation pattern can be narrowed down compared to the general method.

[0189] <Other examples of joining methods> In the combining process in step S5 (FIG. 3), the combining process performed by the combining processor 80 An example will be described.

[0190] The combination processing unit 80 selects the allocation pattern selected by the optimization calculation unit 70 in step S4. A new candidate table is created by combining all the possible combinations of allocation patterns from Using the candidate table thus obtained, the same optimization calculation as in step S4 (optimization calculation) may be further performed. .

[0191] For example, the optimization calculation unit 70 calculates the allocation pattern candidates shown in FIG. After finding the minimum combination that covers all the routes using By combining all the complements, a new candidate table is created as shown in FIG.

[0192] In Fig. 20, the allocation pattern candidates 1, 2, 5, 6, 7, and 8 are all one-day shifts, and candidate 3, 4 and 9 are on late shift. And each assignment pattern has Assume that the weight of labor costs is set as a cost.

[0193] The late shift is because there are fewer flights per person per day compared to the full-day shift. Therefore, in this example, the cost for candidate j working one day is set to 1 , and the cost for working the late shift is set to 1.5.

[0194] Then, the cost C for each allocation pattern listed in Fig. 20 j Substituting into formula 2, we get formula 1 The optimization calculation may be performed again under the constraints.

[0195] In addition, regarding the setting of costs, for example, the links across accommodations shown by the long chain lines in Figure 18, An optimization calculation may be performed to determine whether or not to pass through a link that crosses a break. Let's say you have a day shift and a two-day shift, and you create a node for the first day and a node for the second day, respectively.

[0196] In this case, the following two equations are specific examples of constraint conditions in the optimization calculation.

[0197]

number

[0198]

number

[0199] Here, l pq is 1 if the link connecting node p on the first day and node q on the second day is passed, If the link from node p does not satisfy the constraint condition of Equation 1, the variable is set to 0. The constraint is that the number of allocations to node p must be less than or equal to the number of resources that will be assigned to node q. This is an example of a constraint that the number of nodes to be allocated to node q is equal to or less than the number determined under the constraint condition of Equation 1. .

[0200] Now, if the node p on the first day must be included in the overnight shift (for example, If personnel have no means of returning home, formula 5 must be written as an equal sign instead of an inequality sign with an equality. It won't happen.

[0201] In addition, if node p can finish work on the first day and does not need to be connected on the second day, the equation On the other hand, node q is the second day node, and as explained above, Since the node is always included in the shift that involves overnight stays, it becomes an equality constraint.

[0202] By performing optimization calculations using the constraints in Equation 5 and Equation 6 and the objective function described later, the allocation plan is Able to create sketches of artwork.

[0203] As an example of an objective function, if you want to minimize the number of overnight shifts, use the following objective function and Equation 5. And by using the constraints of Equation 6, the allocation patterns are combined to create a proposal for a personnel allocation plan. It is possible.

[0204]

number

[0205] Another example of an objective function is wt p is the working time of node p, wt q The working time of node q Then, by using the objective function shown in Equation 8 and the constraints of Equation 5 and Equation 6, The optimization may be performed by combining candidates so that their working hours are about a predetermined time (T). In this case, an allocation plan can be created that takes fairness into account.

[0206]

number

[0207] Above, we have explained about working over two days when overnight shifts are included. It can also be expanded to include cases where the work involves multiple days, three or more.

[0208] ==Summary== As described above, the information processing device 10 according to the embodiment is an information processing device having a calculation device and a storage device. and outputs a second allocation plan of personnel who can execute the first allocation plan of the transport aircraft. An information processing device 10, Personnel are assigned to each of the multiple sections defined by two adjacent departure and arrival points on the flight route. A plurality of allocation patterns are generated, each of which indicates whether or not the plurality of allocation patterns can be used. , creating a third allocation plan that is the basis of a second allocation plan that can carry out the first allocation plan, Based on the constraints, multiple allocation patterns in the third allocation plan are combined to create a new allocation. The third allocation plan including the new allocation pattern is output as the second allocation plan. In the process of executing the process and making a new allocation pattern, the new allocation pattern is In each section, personnel are allocated to at least one of the multiple allocation patterns that are combined. If so, personnel are assigned.

[0209] According to this configuration, the first constraint condition is used to suppress an increase in the number of candidates for the allocation pattern. Then, a new allocation pattern is generated by the combination process. Therefore, it is possible to suppress the elimination of a preferable allocation pattern. According to this, it is possible to suppress the deterioration of the quality of the final allocation plan while suppressing the increase in the computation scale. It becomes possible.

[0210] In the information processing device 10, the first constraint condition is that among the takeoff and landing points of the transport aircraft, personnel are working. With this configuration, the start and end points of the trip can be adjusted according to the work patterns of the personnel. A personnel allocation plan can be created.

[0211] In the information processing device 10, the first constraint condition is a predetermined upper or lower limit of the working hours of personnel. According to this configuration, it is possible to suppress an increase in candidates for the allocation pattern.

[0212] In the information processing device 10, the first constraint condition includes an upper limit on the number of multiple allocation patterns. According to this configuration, the number of candidates for the allocation pattern can be set to a desired number.

[0213] In the information processing device 10, the allocation pattern is associated with a cost calculated by a predetermined calculation method. The second allocation plan is created so that the total cost is minimized. According to this configuration, a more favorable personnel allocation plan can be created.

[0214] In the information processing device 10, the second constraint condition includes working conditions stipulated by laws and regulations. According to this configuration, it is possible to create a personnel allocation plan that complies with laws and regulations.

[0215] The information processing method according to the above embodiment is carried out by using an information processing device having a calculation device and a storage device. and an information processing device configured to output a second allocation plan of personnel capable of carrying out the first allocation plan of the transport aircraft. The location is determined based on a first constraint condition to be adjacent to the route of the transport aircraft in the first allocation plan. The allocation indicates whether personnel are assigned to each of the multiple sections that are determined by the two departure and arrival points. Generate multiple patterns and combine parts of the multiple allocation patterns to execute the first allocation plan. Generate a third allocation plan from which a possible second allocation plan is based, based on a given second constraint. A plurality of allocation patterns included in the third allocation plan are combined to form a new allocation pattern, and the new A process is executed to output the third allocation plan including the allocation pattern as the second allocation plan, and a new In the process of setting the allocation pattern, the new allocation pattern is In the case where personnel are assigned in at least one of the multiple assignment patterns to be combined, In such cases, personnel will be allocated.

[0216] According to this method, the first constraint condition suppresses the increase in the number of candidates for allocation patterns. Then, a new allocation pattern is generated by the combination process. Therefore, it is possible to suppress the elimination of a preferable allocation pattern. According to this, it is possible to suppress the deterioration of the quality of the final allocation plan while suppressing the increase in the computation scale. It becomes possible.

[0217] A first allocation plan for a transport aircraft is generated by using an information processing device having a calculation device and a storage device. The information processing device outputs a second allocation plan of personnel who can perform the task based on a predetermined first constraint condition. In the first allocation plan, multiple routes are determined by two adjacent departure and arrival points on the route of the transport aircraft. A plurality of allocation patterns are generated, each of which indicates whether personnel are assigned to each section, and the plurality of allocation patterns are generated. Combining some of the allocation patterns above will form the basis of the second allocation plan that can carry out the first allocation plan. A third allocation plan is created, and a plurality of allocations included in the third allocation plan are allocated based on a second constraint condition. The allocation patterns are combined to form a new allocation pattern, and the third allocation pattern including the new allocation pattern is This plan is output as a second allocation plan, and is made into a new allocation pattern. In the process, the new allocation pattern is a combination of a plurality of segments in each of the plurality of sections. If personnel are assigned in at least one of the assignment patterns, the personnel are assigned. An information processing program.

[0218] According to this method, the first constraint condition suppresses the increase in the number of candidates for allocation patterns. Then, a new allocation pattern is generated by the combination process. Therefore, it is possible to suppress the elimination of a preferable allocation pattern. According to this, it is possible to suppress the deterioration of the quality of the final allocation plan while suppressing the increase in the computation scale. It becomes possible. [Explanation of symbols]

[0219] 10. Information processing device 20 Input / Output Support Department 30 Input data storage unit 30a Transport planning data 30b Candidate enumeration parameters 30c Operational Conditions Data 30d End point determination parameters 40 End point determination and operation condition classification section 50 Constraint storage section 50a Candidate enumeration constraint condition storage unit 50b Join constraint storage section 60 Candidate Enumeration Section 70 Optimization Calculation Section 80 Joint processing section 90 Output data storage unit

Claims

1. A first allocation plan for a transport aircraft is generated using an information processing device having a calculation device and a storage device. An information processing device that outputs a second allocation plan of personnel who can perform the Based on a first predetermined constraint condition, in the route of the transport aircraft in the first allocation plan, Whether the personnel are assigned to each of a plurality of sections defined by two adjacent starting and ending points A plurality of allocation patterns are generated, A second allocation plan capable of executing the first allocation plan by combining a part of the plurality of allocation patterns. Create a third allocation plan on which the allocation plan is based; Based on a predetermined second constraint condition, a plurality of the allocation patterns included in the third allocation plan Combine these into a new allocation pattern, The third allocation plan including the new allocation pattern is output as the second allocation plan. Execute the process, In the process of setting a new allocation pattern, The new allocation pattern is In each of the plurality of sections, at least one of the plurality of allocation patterns to be combined If the personnel is assigned in one, the personnel is assigned; Information processing device.

2. 2. The information processing device according to claim 1, The first constraint condition is: Among the departure and arrival points of the transport aircraft, the conditions of the departure and arrival points where personnel start or end their work, Information processing device.

3. 2. The information processing device according to claim 1, The first constraint condition is: including prescribed upper or lower limits on the number of hours that personnel may work; Information processing device.

4. 2. The information processing device according to claim 1, The first constraint condition is: Including an upper limit of the number of the plurality of allocation patterns, Information processing device.

5. 2. The information processing device according to claim 1, The allocation pattern is associated with a cost calculated by a predetermined calculation method, The second allocation plan is created so as to minimize the total sum of the costs. Information processing device.

6. 2. The information processing device according to claim 1, The second constraint condition is: Including working conditions stipulated by law, Information processing device.

7. A first allocation plan for a transport aircraft is generated using an information processing device having a calculation device and a storage device. an information processing device that outputs a second allocation plan of personnel who can perform the Based on a first predetermined constraint condition, in the route of the transport aircraft in the first allocation plan, Whether the personnel are assigned to each of a plurality of sections defined by two adjacent starting and ending points A plurality of allocation patterns are generated, A second allocation plan capable of executing the first allocation plan by combining a part of the plurality of allocation patterns. Create a third allocation plan on which the allocation plan is based; Based on a predetermined second constraint condition, a plurality of the allocation patterns included in the third allocation plan Combine these into a new allocation pattern, The third allocation plan including the new allocation pattern is output as the second allocation plan. Execute the process, In the process of setting a new allocation pattern, The new allocation pattern is In each of the plurality of sections, at least one of the plurality of allocation patterns to be combined If the personnel is assigned in one, the personnel is assigned; Information processing methods.

8. A first allocation plan for a transport aircraft is generated using an information processing device having a calculation device and a storage device. an information processing device that outputs a second allocation plan of personnel who can perform the Based on a first predetermined constraint condition, in the route of the transport aircraft in the first allocation plan, Whether the personnel are assigned to each of a plurality of sections defined by two adjacent starting and ending points Generate a plurality of allocation patterns that indicate A second allocation plan capable of executing the first allocation plan by combining a part of the plurality of allocation patterns. Create a third allocation plan on which the allocation plan is based, Based on a predetermined second constraint condition, a plurality of the allocation patterns included in the third allocation plan Combine them into a new allocation pattern, The third allocation plan including the new allocation pattern is output as the second allocation plan. Execute the process, In the process of making the new allocation pattern, The new allocation pattern is In each of the plurality of sections, at least one of the plurality of allocation patterns to be combined If the personnel is assigned in one, the personnel is assigned; Information processing program.