Information processing device, information processing method, and program

Abstract

Provided are an information processing device, an information processing method, and a program capable of easily outputting a constraint condition for a work schedule of a crew. An information processing device according to the present disclosure includes: a data acquisition means that acquires electronic data including information pertaining to a work schedule of a crew; a constraint condition generation means that, by inputting the electronic data into a machine learning model and acquiring a constraint condition to be satisfied by the work schedule of the crew from the machine learning model, generates the constraint condition from the electronic data; and a constraint condition output means that outputs the constraint condition.

Description

Information processing device, information processing method, and program

[0001] The present disclosure relates to an information processing device, an information processing method, and a program.

[0002] Railway operators need to create work schedules for their crew members in accordance with train timetables, which are train operation schedules. For example, railway operators need to create work schedules for their crew members, called journeys. When creating journeys, it is necessary to consider not only the efficiency of work but also the working conditions of the crew members. Therefore, creating journeys manually is not easy. For this reason, various technologies have been proposed to automate the creation of journeys. For example, Patent Literature 1 discloses an allocation determination device that determines the optimal journeys for crew members based on a timetable.

[0003] JP 2011-34499 A

[0004] In the technology disclosed in Patent Document 1, the creator of an operation plan inputs conditions, such as upper limits on crew working hours, into a deployment determination device by operating an input device. To achieve this, the creator of the operation plan needs to find the conditions to be input into the deployment determination device from documents such as work regulations that include descriptions of working conditions, and input them as constraints into a device that creates routes. However, the task of finding and extracting constraints from paper or electronic documents that also contain information other than constraints requires a lot of effort. For this reason, there is a need for technology that can easily output constraints for crew work schedules.

[0005] Therefore, one of the objectives that the embodiments disclosed in this specification aim to achieve is to provide an information processing device, information processing method, and program that can easily output constraints on crew work schedules.

[0006] An information processing device according to a first aspect of the present disclosure includes: a data acquisition means for acquiring electronic data including information regarding crew work schedules; a constraint condition generation means for inputting the electronic data into a machine learning model and acquiring constraint conditions that the crew work schedules must satisfy from the machine learning model, thereby generating the constraint conditions from the electronic data; and a constraint condition output means for outputting the constraint conditions.

[0007] An information processing method according to a second aspect of the present disclosure includes a computer acquiring electronic data including information related to crew work schedules, inputting the electronic data into a machine learning model, and acquiring constraints that the crew work schedules must satisfy from the machine learning model, thereby generating the constraints from the electronic data and outputting the constraints.

[0008] A program according to a third aspect of the present disclosure causes a computer to execute a data acquisition step of acquiring electronic data including information regarding crew work schedules; a constraint condition generation step of generating constraint conditions from the electronic data by inputting the electronic data into a machine learning model and acquiring constraint conditions that the crew work schedules must satisfy from the machine learning model; and a constraint condition output step of outputting the constraint conditions.

[0009] According to the present disclosure, it is possible to provide an information processing device, an information processing method, and a program that can easily output constraint conditions for crew work schedules.

[0010] 1 is a block diagram showing an example of a configuration of an information processing device according to the present disclosure. FIG. 2 is a block diagram showing an example of a configuration of an information processing device according to the present disclosure. FIG. 3 is a diagram showing an example of a document in which work regulations for crew members are described. FIG. 4 is a diagram showing an example of a document in which work regulations for crew members are described. FIG. 5 is a diagram showing a specific example of constraint conditions generated by a constraint condition generation unit. FIG. 6 is a diagram showing a specific example of constraint conditions generated by a constraint condition generation unit. FIG. 7 is a diagram showing a specific example of a train timetable. FIG. 8 is a diagram showing a specific example of a route to be generated. FIG. 9 is a block diagram showing an example of a hardware configuration of an information processing device according to the present disclosure. FIG. 10 is a flowchart showing an example of the operation of an information processing device according to the present disclosure.

[0011] For clarity of explanation, the following description and drawings may be omitted or simplified as appropriate. In each drawing, identical or corresponding elements are designated by the same reference numerals, and redundant description is omitted as necessary for clarity of explanation. Furthermore, each drawing is merely an example for describing one or more embodiments. Each drawing is not related to only one particular embodiment, but may also be related to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessarily required to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0012] First Embodiment An example of the configuration of an information processing device 10 will be described below with reference to Fig. 1. Fig. 1 is a block diagram showing an example of the configuration of the information processing device 10. The information processing device 10 has a data acquisition unit 11, a constraint condition generation unit 12, and a constraint condition output unit 13. The information processing device 10 is, for example, any device equipped with computer resources.

[0013] The data acquisition unit 11 acquires electronic data including information about the crew's work schedule. The electronic data acquired by the data acquisition unit 11 may be data of an electronic document created by document creation software, or may be electronic data obtained by performing OCR (Optical Character Recognition) processing on a paper document.

[0014] The constraint condition generation unit 12 generates constraint conditions that the crew work schedule must satisfy from the electronic data acquired by the data acquisition unit 11. The constraint condition generation unit 12 generates constraint conditions by inputting the electronic data acquired by the data acquisition unit 11 into a machine learning model and acquiring constraint conditions output from the machine learning model. The machine learning model used here has been trained in advance by machine learning to extract and output only constraint conditions that the crew work schedule must satisfy from the input information, for example.

[0015] The constraint condition output unit 13 outputs the constraint conditions generated by the constraint condition generation unit 12. The constraint condition output unit 13 can output the constraint conditions to any device or processing unit. For example, the constraint condition output unit 13 may output information on the constraint conditions generated by the constraint condition generation unit 12 to a device equipped with a display in order to display the constraint conditions to a user. Furthermore, the constraint condition output unit 13 may output information on the constraint conditions to a device that executes a predetermined process using the constraint conditions, such as a device that generates a crew work schedule in accordance with the constraint conditions.

[0016] According to the information processing device 10, constraints that must be satisfied by the crew work schedule are automatically generated and output from electronic data containing information about the crew work schedule. This makes it possible to easily output the constraints for the crew work schedule. This eliminates the need for manual work to find and extract constraints from paper or electronic documents that contain information other than the constraints. Note that this effect can also be achieved by an information processing method that includes the above-described processing of the information processing device 10, a program that performs the above-described processing of the information processing device 10, or a non-transitory computer-readable medium on which the program is stored.

[0017] An embodiment that is more specific than embodiment 1 will be described below. <Embodiment 2> An example of the configuration of the information processing device 100 will be described below with reference to Fig. 2. Fig. 2 is a block diagram showing an example of the configuration of the information processing device 100. The information processing device 100 is a device corresponding to the information processing device 10 described above.

[0018] The information processing device 100 includes a data acquisition unit 101, a constraint condition generation unit 102, a constraint condition output unit 103, a diagram acquisition unit 104, a business schedule generation unit 105, and a business schedule output unit 106. The information processing device 100 is, for example, a server, but is not limited to a server and may be any device that has computer functions. For example, the information processing device 100 may be realized as a device connected via a communication network, such as a client-server system or a cloud computing system.

[0019] Like the information processing device 10, the information processing device 100 automatically generates constraint conditions that must be satisfied by the work schedule of a crew member. In this embodiment, the work schedule specifically refers to the route of a crew member who works on a vehicle. Note that in this embodiment, the vehicle on which the crew member works is specifically a train, but is not limited to a train and may also be a bus, etc. In addition, the route corresponds to the work pattern of the crew member, and is a series of schedules from when a crew member arrives at work to when he or she leaves work.

[0020] Routes must be determined taking into consideration not only the working conditions of the crew, such as working hours and rest periods, but also environmental conditions, such as which stations have various facilities, such as dining facilities, accommodations, and depots. In other words, these conditions are constraints that must be satisfied when determining a route. If there is no document or electronic file that summarizes only the constraints, it is necessary to find the conditions that should be used as constraints. Therefore, in this embodiment, first, the data acquisition unit 101 acquires electronic data including information about the crew's work schedules. The data acquisition unit 101 corresponds to the data acquisition unit 11 described above.

[0021] The data acquisition unit 101 may acquire any electronic data including information about the crew's work schedule. The electronic data acquired by the data acquisition unit 101 may be electronic document data created using document creation software, or electronic data obtained by performing OCR processing on a paper document. That is, if the medium on which the information about the crew's work schedule is written is a paper medium, the data acquisition unit 101 may acquire electronic data composed of text data by performing OCR processing on the paper medium. In this case, specifically, the data acquisition unit 101 receives a scanned image of the paper document and performs OCR processing on the scanned image to extract text written in the document and acquire electronic data of the document. This processing makes it possible to easily acquire information about the crew's work schedule even if the information is written on a paper medium. Note that the OCR processing performed by the data acquisition unit 101 may be a known AI-OCR process, which is an OCR process incorporating AI (artificial intelligence) technology, on the paper document.

[0022] The data acquisition unit 101 acquires electronic data of a document such as that shown in FIGS. 3A and 3B. FIGS. 3A and 3B show an example of a document that describes crew work regulations. As shown in FIGS. 3A and 3B, the document 40 includes not only a sentence 41 related to constraints but also sentences unrelated to constraints. Furthermore, the descriptions of conditions that should be considered constraints may be scattered across various documents, not just one document. For this reason, when a schedule creator extracts constraints from a document, it requires a great deal of effort. In the example shown in FIGS. 3A and 3B, the descriptions related to constraints are all written in sentences, but the descriptions related to constraints may also be written in a table format. In this way, the data acquisition unit 101 acquires electronic data of a document that specifies crew work, for example.

[0023] The constraint condition generation unit 102 corresponds to the constraint condition generation unit 12 described above, and generates constraint conditions that the crew work schedule must satisfy from the electronic data acquired by the data acquisition unit 101. Specifically, the constraint condition generation unit 102 inputs the electronic data acquired by the data acquisition unit 101 into a machine learning model. The constraint condition generation unit 102 then acquires the constraint conditions output from this machine learning model. As a result, the constraint condition generation unit 102 generates constraint conditions from the electronic data acquired by the data acquisition unit 101.

[0024] The constraint generation unit 102 may use any machine learning model using a known machine learning technique. For example, the machine learning model used by the constraint generation unit 102 may be a machine learning model that provides generative artificial intelligence (AI) services, such as large language models (LLMs). In this case, the constraint generation unit 102 inputs a prompt to the LLM, instructing it to extract constraints that the crew work schedule must satisfy from text included in the electronic data acquired by the data acquisition unit 101, along with the electronic data. The machine learning model used by the constraint generation unit 102 may also be a machine learning model trained as follows. For example, the machine learning model used may be a model generated by performing machine learning using a large amount of training data that is a combination of text data containing descriptions of constraints and constraints to be output from the text data. In this case, the text data containing descriptions of constraints used as training data may be text data containing only descriptions of constraints, or may be data containing text related to constraints and other text. The constraints may also be acquired through a two-stage process. For example, the constraint generation unit 102 may first use a first machine learning model to acquire only the sentences describing the constraints from text data (e.g., text data of the entire document) that includes sentences describing the constraints and other sentences, and then use a second machine learning model to extract constraints from the sentences describing the constraints output from the first machine learning model.

[0025] The data acquisition unit 101 may acquire electronic data representing a work schedule actually performed by the crew as electronic data containing information about the crew's work schedule. That is, the data acquisition unit 101 may acquire electronic data of the crew's work performance. In this case, the data acquisition unit 101 acquires electronic data of multiple work schedules as samples to discover constraints from the work performance. That is, the data acquisition unit 101 acquires a dataset of the actually performed work schedules. The dataset of the actually performed work schedules may include data indicating whether each work schedule satisfies a schedule rule that satisfies the constraints. When the data acquisition unit 101 acquires electronic data representing the work schedule actually performed by the crew, the constraint generation unit 102 generates constraints using, for example, a known machine learning model using a rule discovery inference technique. That is, the constraint generation unit 102 inputs the electronic data representing the work schedule actually performed by the crew into such a machine learning model and acquires the constraints output from the model (in other words, discovered by the model). This allows constraint conditions to be generated from the work schedules actually performed by the crew, so that constraint conditions can be acquired even when electronic data containing explicit descriptions of the conditions cannot be acquired. In this way, the electronic data containing information on the crew work schedules acquired by the data acquisition unit 101 does not necessarily need to contain explicit descriptions of the conditions.

[0026] The constraint condition generation unit 102 may generate constraint conditions expressed in natural language. FIG. 4 is a diagram showing a specific example of constraint conditions generated by the constraint condition generation unit 102. FIG. 4 shows a specific example of constraint conditions expressed in natural language generated based on the electronic data of the document shown in FIGS. 3A and 3B. In FIG. 4, the constraint conditions are listed in itemized form using natural language. As shown in FIG. 4, the constraint condition generation unit 102 may generate a list of constraint conditions expressed in natural language. In this way, the constraint condition generation unit 102 may generate text data of constraint conditions expressed in natural language. Furthermore, as shown in FIG. 4, the constraint condition generation unit 102 may generate text data of constraint conditions by classifying the constraint conditions by type. In the example shown in FIG. 4, the constraint condition group 50a is a constraint condition related to the start or end of work, the constraint condition group 50b is a constraint condition related to breaks, the constraint condition group 50c is a constraint condition related to working hours, and the constraint condition group 50d is a constraint condition related to facilities. In this way, the constraint conditions may be listed by type. It should be noted that the constraint conditions shown in FIG. 4 are merely examples, and it goes without saying that other constraint conditions may be generated based on the electronic data acquired by the data acquisition unit 101.

[0027] The constraint generating unit 102 may also generate constraints expressed in mathematical expressions. Instead of generating constraints expressed in natural language as described above, the constraint generating unit 102 may generate constraints expressed in mathematical expressions, or in addition to generating constraints expressed in natural language. Figure 5 shows a specific example of constraints generated by the constraint generating unit 102. Figure 5 shows a specific example of constraints expressed in mathematical expressions generated based on the electronic data of the document shown in Figures 3A and 3B. As shown in Figure 5, the constraint generating unit 102 may generate a list of constraints expressed using variables and mathematical symbols such as inequality signs and equality signs.

[0028] In FIG. 5 , the variables for each constraint are as follows: The variable "S_early" indicates the earliest time that can be set as the start time for work on the first day of a two-day work shift. The variable "E_late" indicates the latest time that can be set as the end time for work on the second day of a two-day work shift. The variable "T1_early" indicates the earliest time that can be set as the breakfast time slot. The variable "T1_late" indicates the latest time that can be set as the breakfast time slot. The variable "T1" indicates the length (in minutes) of the breakfast break. The variable "T2_early" indicates the earliest time that can be set as the lunch time slot. The variable "T2_late" indicates the latest time that can be set as the lunch time slot. The variable "T2" indicates the length (in minutes) of the lunch break. The variable "T3_early" indicates the earliest time that can be set as the dinner time slot. The variable "T3_late" indicates the latest time that can be set as the dinner time slot. The variable "T3" indicates the length (in minutes) of the break time for dinner. The variable "ST" indicates the length (in minutes) of the break time for sleep, i.e., the overnight stay. The variable "AT" indicates the maximum length (in minutes) of continuous train duty. The variable "W1" indicates the length (in minutes) of the work time that should be set when a train is to be departed from the depot. The variable "W2" indicates the length (in minutes) of the work time that should be set when a train is not to be departed from the depot. The variable "S1" indicates a set of identifiers for stations where depots are located. The variable "S2" indicates a set of identifiers for stations where base offices are located. The variable "S3" indicates a set of identifiers for stations where accommodation facilities are located. The variable "S4" indicates a set of identifiers for stations where dining facilities are located.

[0029] The above describes the variables shown in Figure 5, but it goes without saying that the variables shown in Figure 5 are merely examples, and other constraints may be generated based on the electronic data acquired by the data acquisition unit 101.

[0030] The constraints may be expressed in mathematical expressions by any of the machine learning models described above. That is, the machine learning model to which the electronic data acquired by the data acquisition unit 101 is input may output constraints expressed in mathematical expressions. In this case, the constraint generation unit 102 may instruct the machine learning model to output constraints using variables. For example, the constraint generation unit 102 may input, to the LLM, a prompt instructing the LLM to extract constraints from text included in the electronic data acquired by the data acquisition unit 101 and output them as mathematical expressions. Alternatively, the constraint generation unit 102 may acquire constraints expressed in natural language output from the machine learning model and convert the constraints expressed in natural language into constraints expressed in mathematical expressions. For example, the constraint generation unit 102 may convert a predetermined character string into a predetermined variable and convert a description of a numerical value into a description using mathematical symbols according to predetermined conversion rules, thereby converting the expressions.

[0031] The constraint condition output unit 103 corresponds to the constraint condition output unit 13 described above and performs processing to output the constraint conditions generated by the constraint condition generation unit 102. The constraint condition output unit 13 can output the constraint conditions to any device or processing unit. In this embodiment, the constraint condition output unit 103 outputs the constraint conditions to the work schedule generation unit 105 described later. Furthermore, the constraint condition output unit 103 outputs the constraint conditions generated by the constraint condition generation unit 102 to a device equipped with a display in order to display the constraint conditions to a user. The device equipped with a display may be a terminal device such as a smartphone, tablet, or personal computer communicatively connected to the information processing device 100 via a network or the like, or may be a display device directly connected to the information processing device 100. Performing such output processing allows the user to easily understand the constraint conditions. Furthermore, the constraint condition output unit 103 may output the constraint conditions generated by the constraint condition generation unit 102 to a device that generates a crew work schedule in accordance with the constraint conditions, i.e., another device that performs processing equivalent to the work schedule generation unit 105 described later. The other device may be a device communicably connected to the information processing device 100 via a network, etc. By performing such output processing, the information necessary for the generation processing of the business schedule can be easily provided to the device.

[0032] When the constraints are output to display them to the user, the constraint output unit 103 preferably outputs the list of constraints expressed in the above-described natural language to a device equipped with a display. This allows the user to understand the constraints more easily than when the constraints are expressed in mathematical expressions or the like. When the constraints are output to a device that performs a predetermined process, such as generating a business schedule, the constraint output unit 103 preferably outputs the list of constraints expressed in the above-described mathematical expressions. This allows the device to use the input information more easily than when the constraints are expressed in natural language or the like.

[0033] The timetable acquisition unit 104 acquires timetable data indicating travel schedules of vehicles. In this embodiment, the timetable acquisition unit 104 specifically acquires data indicating train operation timetables as the timetable data.

[0034] The work schedule generation unit 105 generates a work schedule for a crew member who works on a vehicle that moves according to a travel schedule indicated in the timetable data, the work schedule being in accordance with constraints. In the present embodiment, as an example, the work schedule generation unit 105 generates a work schedule for a crew member who works on a train. The work schedule generation unit 105 generates a work schedule using the constraints generated by the constraint condition generation unit 102 and the timetable data acquired by the timetable acquisition unit 104. Here, the work schedule specifically refers to the route of the crew member who works on the vehicle. The work schedule, i.e., the route, includes, for example, a work schedule that specifies the vehicle the crew member is responsible for, the travel section they are responsible for, and work time information, and a rest schedule that specifies rest locations and rest time information. Each crew member works according to the route assigned to them. Note that the route may indicate a schedule spanning multiple days (for example, two days).

[0035] The business schedule generating unit 105 may generate a business schedule according to any known algorithm for generating a business schedule. For example, the business schedule generating unit 105 may generate a route using the technology disclosed in Patent Document 1, may generate a route using a mixed integer programming solver, or may generate a route using any machine learning model. In this way, the business schedule generating unit 105 may generate a business schedule using a predetermined algorithm, and the specific generation method is not limited to any one method. By generating a business schedule by the business schedule generating unit 105, the user can easily obtain a business schedule that conforms to constraints.

[0036] The timetable acquisition unit 104 acquires timetable data showing a train timetable such as that shown in FIG. 6 . In the timetable diagram shown in FIG. 6 , circles indicate vehicles leaving a depot, and triangles indicate vehicles entering a depot. Taking timetable 60 as an example, how to read the timetable diagram shown in FIG. 6 will be explained. Timetable 60 shows the following movement schedule for a certain vehicle: The vehicle leaves the depot at station A, and travels from station A to station C as a train identified by train number 3M. After waiting for a while at station C, the vehicle travels from station C to station A as a train identified by train number 4M. Then, the vehicle enters the depot at station A.

[0037] The work schedule generating unit 105 generates data indicating routes, for example, as shown in FIG. 7 , using the acquired timetable data, the constraints generated by the constraint generating unit 102, and a predetermined algorithm. In the example shown here, five routes are generated to operate each train according to the timetable shown in FIG. 6 . In this case, these routes are assigned to five crew members. In this way, for example, the work schedule generating unit 105 generates routes such that the crew members' working hours and rest times satisfy the constraints, and the locations where the crew members' actions (e.g., accommodation, meals, and work) are performed satisfy the constraints related to facilities. In other words, the work schedule generating unit 105 generates routes that satisfy the working conditions, environmental conditions, and the like, which are expressed as constraints.

[0038] The work schedule output unit 106 outputs the work schedule generated by the work schedule generation unit 105. The work schedule output unit 106 may output information indicating the work schedule to any device or processing unit. For example, the work schedule output unit 106 may output the generated work schedule to a device equipped with a display to display the work schedule to a user. In the present disclosure, a device equipped with a display refers to any device equipped with a display, and may be, for example, a tablet terminal or a smartphone, or a display device using XR (Cross Reality) technology such as VR (Virtual Reality), AR (Augmented Reality), or MR (Mixed Reality). The work schedule may also be output as audio from a device equipped with a speaker. In this case, the device may interact with the user via audio when outputting information to the user or inputting information from the user. The device to which the work schedule is output may be a terminal device used by a crew member. The work schedule output unit 106 may also output the work schedule to a device that executes a predetermined process using the work schedule.

[0039] Next, a description will be given of an example of the hardware configuration of the information processing device 100. Fig. 8 is a block diagram showing an example of the hardware configuration of the information processing device 100. As shown in Fig. 8, the information processing device 100 includes an input / output interface 151, a memory 152, and a processor 153.

[0040] The input / output interface 151 is an interface for connecting to other devices or networks as needed, and may include a network interface.

[0041] The memory 152 is configured, for example, by a combination of volatile memory and non-volatile memory, and is used to store software (computer programs) including one or more instructions executed by the processor 153, as well as data used in various processes.

[0042] The processor 153 reads and executes software (computer programs) from the memory 152 to perform processing of each component shown in Fig. 2. The processor 153 may be, for example, a microprocessor, an MPU (Micro Processor Unit), or a CPU (Central Processing Unit). The processor 153 may include multiple processors. In this way, the information processing device 100 has the functionality of a computer.

[0043] The program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored on a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable media or tangible storage media include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, CD-ROM, digital versatile disc (DVD), Blu-ray disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals.

[0044] Next, the processing flow of each component shown in Fig. 2 will be described with reference to the flowchart of Fig. 9. Fig. 9 is a flowchart showing an example of the operation of the information processing device 100.

[0045] In step S100, the data acquisition unit 101 acquires electronic data including information about the crew's work schedule. Next, in step S101, the constraint condition generation unit 102 generates constraint conditions that the crew's work schedule must satisfy from the electronic data acquired in step S100.

[0046] Next, in step S102, the constraint condition output unit 103 performs processing to output the constraint conditions generated in step S101. In this embodiment, the constraint conditions are output to the business schedule generation unit 105. However, instead of outputting to the business schedule generation unit 105, the constraint conditions may be output to another device that generates a business schedule. Furthermore, the constraint conditions may be output to a device equipped with a display so that the user can see the constraint conditions. Furthermore, this device equipped with a display may be a terminal device used by the crew.

[0047] Next, in step S103, the schedule acquisition unit 104 acquires schedule data indicating the travel schedule of the vehicle. Next, in step S104, the work schedule generation unit 105 generates a work schedule for the crew using the constraints and the schedule data. Then, in step S105, the work schedule output unit 106 outputs the work schedule generated in step S104.

[0048] The second embodiment has been described above. According to the information processing device 100, constraints that crew work schedules must satisfy are automatically generated from electronic data containing information about the crew work schedules, and the constraints are output. Therefore, the constraints for the crew work schedules can be easily obtained. Furthermore, according to the information processing device 100, a work schedule is automatically generated based on the obtained constraints. Therefore, the work schedules for the crew can be easily obtained. In this embodiment, in order for the information processing device 100 to generate and output the work schedules, the information processing device 100 includes the timetable acquisition unit 104, the work schedule generation unit 105, and the work schedule output unit 106. However, if the information processing device 100 does not need to perform such processing, these components may be omitted from the information processing device 100.

[0049] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0050] Some or all of the above embodiments may also be described as, but are not limited to, the following supplementary notes. Some or all of the elements (e.g., configurations and functions) described in Supplementary Notes 2 to 9 that are dependent on Supplementary Note 1 may also be dependent on Supplementary Notes 10 and 11 in the same dependent relationship as Supplementary Notes 2 to 9. Furthermore, some or all of the elements described in any supplementary note may be applied to various hardware, software, recording means for recording software, systems, and methods.

[0051] (Supplementary Note 1) An information processing device comprising: a data acquisition means for acquiring electronic data including information on crew work schedules; a constraint generation means for generating the constraints from the electronic data by inputting the electronic data into a machine learning model and acquiring constraints that the crew work schedule must satisfy from the machine learning model; and a constraint output means for outputting the constraints. (Supplementary Note 2) The information processing device according to Supplementary Note 1, wherein the electronic data is electronic data of a document specifying crew work. (Supplementary Note 3) The information processing device according to Supplementary Note 1, wherein the electronic data is electronic data representing a work schedule actually performed by the crew. (Supplementary Note 4) The information processing device according to any one of Supplements 1 to 3, wherein the constraint output means outputs the constraints to a device equipped with a display to display the constraints to a user. (Supplementary Note 5) The information processing device according to Supplementary Note 4, wherein the constraint generation means generates a list of the constraints expressed in natural language, and the constraint output means outputs the list of constraints to the device. (Supplementary Note 6) The information processing device according to any one of Supplements 1 to 5, wherein the constraint condition output means outputs the constraint condition to a device that generates a work schedule for a crew member in accordance with the constraint conditions. (Supplementary Note 7) The information processing device according to Supplementary Note 6, wherein the constraint condition generation means generates a list of the constraint conditions expressed by a mathematical expression, and the constraint condition output means outputs the list of constraint conditions to the device. (Supplementary Note 8) The information processing device according to any one of Supplements 1 to 7, further comprising: a timetable acquisition means that acquires timetable data indicating a travel schedule of a vehicle; a work schedule generation means that uses the constraint conditions and the timetable data to generate a work schedule for a crew member who travels in accordance with the travel schedule in accordance with the constraint conditions; and a work schedule output means that outputs the work schedule. (Supplementary Note 9) The information processing device according to any one of Supplements 1 to 8, wherein the data acquisition means acquires the electronic data composed of text data by performing OCR processing on a paper medium on which the information related to the work schedule of the crew member is written.(Supplementary Note 10) An information processing method in which a computer acquires electronic data including information on crew work schedules, inputs the electronic data into a machine learning model and acquires constraint conditions that the crew work schedules should satisfy from the machine learning model, thereby generating the constraint conditions from the electronic data, and outputs the constraint conditions. (Supplementary Note 11) A program that causes a computer to execute the following steps: a data acquisition step of acquiring electronic data including information on crew work schedules, a constraint condition generation step of generating the constraint conditions from the electronic data by inputting the electronic data into a machine learning model and acquiring constraint conditions that the crew work schedules should satisfy from the machine learning model, and a constraint condition output step of outputting the constraint conditions.

[0052] This application claims priority based on Japanese Patent Application No. 2024-087618, filed May 30, 2024, the disclosure of which is incorporated herein in its entirety by reference.

[0053] REFERENCE SIGNS LIST 10 Information processing device 11 Data acquisition unit 12 Constraint condition generation unit 13 Constraint condition output unit 100 Information processing device 101 Data acquisition unit 102 Constraint condition generation unit 103 Constraint condition output unit 104 Diagram acquisition unit 105 Business schedule generation unit 106 Business schedule output unit 151 Input / output interface 152 Memory 153 Processor

Claims

1. An information processing device having: a data acquisition means for acquiring electronic data including information regarding crew work schedules; a constraint condition generation means for inputting the electronic data into a machine learning model and acquiring constraint conditions that the crew work schedules must satisfy from the machine learning model, thereby generating the constraint conditions from the electronic data; and a constraint condition output means for outputting the constraint conditions.

2. The information processing device according to claim 1, wherein the electronic data is electronic data of a document specifying the duties of a crew member.

3. The information processing device according to claim 1, wherein the electronic data is electronic data representing a work schedule actually carried out by a crew member.

4. The information processing device according to any one of claims 1 to 3, wherein the constraint output means outputs the constraint to a device equipped with a display in order to display the constraint to a user.

5. The information processing device according to claim 4, wherein the constraint generating means generates a list of the constraints expressed in natural language, and the constraint output means outputs the list of constraints to the device.

6. An information processing device according to any one of claims 1 to 5, wherein the constraint condition output means outputs the constraint condition to a device that generates a crew work schedule in accordance with the constraint conditions.

7. The information processing device according to claim 6, wherein said constraint generating means generates a list of said constraints expressed in mathematical expressions, and said constraint output means outputs said list of constraints to said device.

8. An information processing device as claimed in any one of claims 1 to 7, further comprising: a schedule acquisition means for acquiring schedule data indicating the travel schedule of a vehicle; a work schedule generation means for using the constraints and the schedule data to generate a work schedule according to the constraints for crew members on board the vehicle that travels according to the travel schedule; and a work schedule output means for outputting the work schedule.

9. An information processing device according to any one of claims 1 to 8, wherein the data acquisition means acquires the electronic data consisting of text data by performing OCR processing on a paper medium on which the information relating to the crew's work schedule is written.

10. An information processing method in which a computer acquires electronic data including information regarding crew work schedules, inputs the electronic data into a machine learning model, and acquires constraints that the crew work schedules must satisfy from the machine learning model, thereby generating the constraints from the electronic data, and outputting the constraints.

11. A program that causes a computer to execute the following steps: a data acquisition step of acquiring electronic data containing information about crew work schedules; a constraint condition generation step of generating constraint conditions from the electronic data by inputting the electronic data into a machine learning model and obtaining constraint conditions that the crew work schedules must satisfy from the machine learning model; and a constraint condition output step of outputting the constraint conditions.

Images