Information providing device, information providing system, information providing method and information providing program

Abstract

To provide an information providing device capable of providing, to a user, highly precise information generated based on a train operation situation.SOLUTION: An information providing device 10 includes: a data collecting unit 11 that collects, from a plurality of railroad companies, operation relevant data relating to the operations of trains; an accumulated data storing unit 12 that accumulates the operation relevant data collected by the data collecting unit 11; and a predicting unit 13 that predicts future operations of the trains based on the operation relevant data collected from the plurality of railroad companies.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to an information providing apparatus, an information providing system, an information providing method, and an information providing program that create information based on information held by a railway operator and provide the information to users.

Background Art

[0002] In recent years, in order to improve the convenience of users who use railways, station facilities, and facilities around stations, the efficiency of the operations of railway operators, and the efficiency of the operations of operators who manage facilities around stations, it has been required to provide information generated based on the operating status of railways to users, railway operators, or operators of facilities around stations.

[0003] For example, Patent Document 1 discloses a passenger guidance system that centrally manages train information for multiple line sections and provides the operating status of the entire line in real time. The passenger guidance system disclosed in Patent Document 1 integrates the train schedule information obtained from the operation management system to generate an overall line schedule, performs a train delay prediction based on the overall line schedule and the delay information of each line section, generates a predicted schedule including the delay prediction, and presents it to the user. According to the passenger guidance system disclosed in Patent Document 1, railway users can know the predicted schedule including the train delay prediction of the railway, and thus can make a judgment on whether to change the travel route based on the predicted delay time.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] With the development of the railway network, the number of interchangeable stations where multiple railway operators can enter has increased, and the number of cases where multiple railway operators mutually enter each other has also increased. In addition, large-scale passenger attraction facilities may be constructed around stations due to redevelopment projects or the like. For this reason, due to delays on other railway lines, the number of incoming passengers due to transfers may suddenly increase, delays in a section of a certain railway operator may spread to the railway operator of the mutual entry destination, or the number of railway users may increase due to an event being held at a passenger attraction facility around the station. Thus, the number of events originating outside a railway operator that affect railway operations is increasing.

[0006] The passenger guidance system disclosed in Patent Document 1 collects information from multiple sections of the same railway operator to generate a predicted timetable. Therefore, when railway operations are affected by events originating outside the railway operator, the accuracy of the information generated based on the railway operation status may be reduced.

[0007] The present disclosure has been made in view of the above, and an object thereof is to obtain an information providing device that can provide a user with highly accurate information generated based on the railway operation status.

Means for Solving the Problems

[0008] In order to solve the above-described problems and achieve the object, an information providing device according to the present disclosure includes a data collection unit that collects operation-related data regarding railway operations from a plurality of railway operators, and a storage data storage unit that stores the operation-related data collected by the data collection unit. The information providing device includes a prediction unit that makes a future prediction regarding railway operations based on the operation-related data collected from a plurality of railway operators.

Effects of the Invention

[0009] According to the present disclosure, there is an effect that an information providing device that can provide a user with highly accurate information generated based on the railway operation status can be obtained.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Mode for Carrying Out the Invention

[0011] Hereinafter, an information providing apparatus, an information providing system, an information providing method, and an information providing program according to an embodiment will be described in detail with reference to the drawings.

[0012] Embodiment 1. FIG. 1 is a diagram showing the configuration of the information providing system according to Embodiment 1. The information providing system 100 according to Embodiment 1 includes an information providing apparatus 10, an operation management apparatus 20 operated by each of a plurality of railway operators, and an external server 30.

[0013] The information providing device 10 includes a data collection unit 11 that collects data from the operation management device 20 and the external server 30, and a storage data storage unit 12 that stores the data collected by the data collection unit 11 as storage data. Further, the information providing device 10 includes a prediction unit 13 that makes future predictions regarding railway and railway operations, and a prediction result data storage unit 14 that stores prediction result data indicating the results of the predictions by the prediction unit 13. Further, the information providing device 10 includes a data transmission unit 15 that transmits the prediction result data stored in the prediction result data storage unit 14 to the user terminal 40 of users who use railway users, station facilities, and facilities around the station, the service provider terminal 50 of the service provider, and the operation management device 20. The prediction result data storage unit 14 stores the prediction result data without overwriting it until a preset number, and when newly storing the prediction result data while storing the preset number of prediction result data, the oldest prediction result data is overwritten to store the latest prediction result data. Generally, the information providing device 10 is realized by a cloud server connected to the Internet, but in the case where the number of railway operators is small, an on-premises server may be used.

[0014] The data collection unit 11 collects operation-related data, which is data regarding the operation of the railway of each railway operator, from the operation management device 20 operated by each of a plurality of railway operators related to each other. Here, the plurality of railway operators related to each other are railway operators where the inflow and outflow of users due to transfer occur, or railway operators that perform mutual entry.

[0015] The operation-related data is schedule diagram data, ridership data, number of passengers getting on and off data, origin and destination data, power consumption information, staff allocation information, vehicle information, accident information, and substitute transport information.

[0016] The planned schedule data is data indicating which vehicles will operate as what types of premium trains at what times by railway operators. The ridership data is data indicating the ratio of the number of passengers actually on board to the seating capacity of the trains operated. The ridership can be calculated based on the load on the air springs installed in the vehicle or based on the images of the cameras installed inside the vehicle. The passenger flow data includes the number of passengers boarding and alighting at each station of the trains operated and the number of people entering and leaving each station. The origin-destination data is data indicating the entry station and the exit station of railway users, and is also referred to as OD (Origin Destination) data. The power consumption information is information on the power consumed by railway operators. The staff allocation information is information indicating the allocation status of the staff of railway operators. The staff allocation information includes crew allocation information indicating the allocation of crew members on board the train, station staff allocation information indicating the allocation of station staff involved in station operations, and maintenance staff allocation information indicating the allocation of maintenance staff involved in track maintenance work. The vehicle information is information indicating the condition of the vehicle, such as the degree of wear of the vehicle used for the train. The accident information is information indicating where and what kind of accident occurred on the track. The transfer transportation information is information indicating in which section of the track the transfer transportation was carried out.

[0017] In addition, the data collection unit 11 collects operation impact cause data, which is data on events that cause an impact on railway operations, from the external server 30. The operation impact cause data is weather data, weather forecast data, and event information.

[0018] The weather data is data indicating the actual changes in weather, such as weather, temperature, wind speed, etc. The weather forecast data is data indicating the prediction of changes in weather, such as weather, temperature, wind speed, etc. By combining the weather forecast data and the weather data at past points in time, it is shown whether the weather forecast at past points in time was accurate or inaccurate.

[0019] Event information is data indicating events related to the railway business. Events related to the railway business are, for example, events held at stations or facilities around stations. Event information is information including, for example, the type of event and the number of visitors, such as a concert held in a venue with a capacity of 10,000 people. Examples of events held at stations include the opening ceremony of a new line and the retirement ceremony of old trains. Examples of events held at facilities around stations include sports games and concerts.

[0020] The accumulated data storage unit 12 stores the operation-related data and operation impact cause data collected by the data collection unit 11 as accumulated data.

[0021] By combining the planned timetable data of each of a plurality of railway operators with at least one of the origin-destination data of the railway users of each of the plurality of railway operators, the boarding rate data of the trains operated by each of the plurality of railway operators, and the number-of-passengers data of each of the plurality of railway operators, it is possible to estimate from which station a railway user boards, which train the user boards, which route the user takes, and at which station the user gets off. That is, by combining the planned timetable data of each of the plurality of railway operators with at least one of the origin-destination data of the railway users of each of the plurality of railway operators, the boarding rate data of the trains operated by each of the plurality of railway operators, and the number-of-passengers data of each of the plurality of railway operators, it is possible to create congestion data, count data, stay data, and movement trajectory data. The congestion data is data indicating the congestion level of each station owned by the railway operator, for example, data indicating the number of passengers in the station precincts. The count data is data indicating the number of people passing through a certain point or the number of people within a specific range. The stay data is data indicating the number of people staying in a certain point or space for a certain period of time. The movement trajectory data is data indicating the movement trajectory of a person. The four types of data, namely the count data, the stay data, the origin-destination data, and the movement trajectory data, are collectively referred to as the pedestrian flow data. Therefore, by combining the planned timetable data of each of the plurality of railway operators with at least one of the origin-destination data of the railway users of each of the plurality of railway operators, the boarding rate data of the trains operated by each of the plurality of railway operators, and the number-of-passengers data of each of the plurality of railway operators, it is possible to create congestion data and pedestrian flow data. Hereinafter, the number of people passing through a certain point within the area under the jurisdiction of the railway operator or the number of people within a specific range, the number of people staying in a certain point or space within the area under the jurisdiction of the railway operator for a certain period of time, the entry and exit stations of the railway users, and the movement trajectory of the railway users are collectively referred to as "pedestrian flow".

[0022] By combining the planned timetable data with at least one of the number-of-passengers data, the boarding rate data, and the origin-destination data, it is possible to create delay information indicating how much time each train has been delayed in operation with respect to the planned timetable.

[0023] Based on the operation-related data collected from the operation management devices 20 of each of the plurality of railway operators, or based on the operation-related data collected from the operation management devices 20 of each of the plurality of railway operators and the operation impact cause data collected from an external server, the prediction unit 13 makes a future prediction regarding the operation of the railway. For example, the prediction unit 13 uses the accumulated data stored in the accumulated data storage unit 12, the operation-related data and the operation impact cause data collected by the data collection unit 11 to make a future prediction regarding the operation of the railway.

[0024] FIG. 2 is a diagram showing the configuration of the prediction unit of the information providing apparatus according to Embodiment 1. The prediction unit 13 includes a learning device 131, a learned model storage unit 132 that stores learned data learned by the learning device 131, and an inference device 133.

[0025] The learning device 131 learns congestion and the flow of people using, as learning data, at least one of the planned schedule data among the accumulated data, and at least one of the origin-destination data, the boarding rate data, and the number of boarding and alighting passengers data, and at least one of the weather data, the weather forecast data, the accident information, and the substitute transportation information. Further, the learning device 131 learns the delay in the section of the railway operator using, as learning data, at least one of the planned schedule data among the accumulated data, and at least one of the origin-destination data, the boarding rate data, and the number of boarding and alighting passengers data, and at least one of the weather data, the weather forecast data, the event information, and the substitute transportation information. Further, the learning device 131 learns the amount of power consumed by the railway operator using, as learning data, the planned schedule data and the power amount information among the accumulated data. Note that the learning data for learning the amount of power consumed by the railway operator may include the origin-destination data. Further, the learning device 131 learns the staff allocation status of the railway operator based on the planned schedule data and the staff allocation information. Further, the learning device 131 learns the state of the vehicles owned by the railway operator using, as learning data, the planned schedule data and the vehicle information. Note that the learning data for learning the state of the vehicles owned by the railway operator may include the weather data.

[0026] FIG. 3 is a diagram showing the configuration of the learning device of the information providing device according to Embodiment 1. The learning device 131 includes a learning data acquisition unit 1311 and a model generation unit 1312.

[0027] When learning congestion and the flow of people using as learning data the schedule diagram data, at least one of the origin-destination data, boarding rate data, and number of passengers getting on and off data, and at least one of the weather data, weather forecast data, accident information, and substitute transportation information, the learning data acquisition unit 1311 acquires as learning data the schedule diagram data, at least one of the origin-destination data, boarding rate data, and number of passengers getting on and off data, and at least one of the weather data, weather forecast data, accident information, and substitute transportation information. As described above, by combining the schedule diagram data with at least one of the origin-destination data, boarding rate data, and number of passengers getting on and off data, it is possible to create congestion data and people flow data. Therefore, the learning data acquisition unit 1311 can be regarded as acquiring as learning data the congestion data and people flow data, and at least one of the weather data, weather forecast data, accident information, and substitute transportation information. The model generation unit 1312 learns the relationship between the congestion data and people flow data generated from the schedule diagram data and at least one of the origin-destination data, boarding rate data, and number of passengers getting on and off data, and at least one of the weather data, weather forecast data, accident information, and substitute transportation information.

[0028] When learning about delays in a railway operator's section using, as learning data, schedule diagram data, at least one of origin-destination data, boarding rate data, and passenger boarding / alighting data, and at least one of weather data, weather forecast data, event information, and diversion transport information, the learning data acquisition unit 1311 acquires, as learning data, schedule diagram data, at least one of origin-destination data, boarding rate data, and passenger boarding / alighting data, and at least one of weather data, weather forecast data, event information, and diversion transport information. As described above, since delay information can be created by combining schedule diagram data with at least one of origin-destination data, boarding rate data, and passenger boarding / alighting data, the learning data acquisition unit 1311 can be regarded as acquiring, as learning data, delay information and at least one of weather data, weather forecast data, event information, and diversion transport information. The model generation unit 1312 learns the relationship between delay information generated from schedule diagram data and at least one of origin-destination data, boarding rate data, and passenger boarding / alighting data, and at least one of weather data, weather forecast data, event information, and diversion transport information.

[0029] Also, when learning about the amount of power consumed by a railway operator using schedule diagram data and power amount information as learning data, the learning data acquisition unit 1311 acquires schedule diagram data and power amount information as learning data. The model generation unit 1312 learns the relationship between schedule diagram data and power amount information.

[0030] Also, when learning about the staff allocation situation of a railway operator based on schedule diagram data and staff allocation information, the learning data acquisition unit 1311 acquires schedule diagram data and staff allocation information as learning data. The model generation unit 1312 learns the relationship between schedule diagram data and staff allocation information.

[0031] Also, when the railway operator learns the state of the vehicles owned by the operator using the planned train schedule data and vehicle information as learning data, the learning data acquisition unit 1311 acquires the planned train schedule data and vehicle information as learning data. The model generation unit 1312 learns the relationship between the planned train schedule data and the vehicle information.

[0032] As the learning algorithm used by the model generation unit 1312, known algorithms such as supervised learning, unsupervised learning, and reinforcement learning can be used. As an example, the case of applying a neural network will be described.

[0033] The model generation unit 1312 learns, for example, the congestion level and the flow of people by so-called supervised learning according to a neural network model. Here, supervised learning refers to a method of giving a learning device 131 a set of data consisting of "inputs" and the "labels" that are the results, learning the characteristics in those learning data, and inferring the results from the inputs.

[0034] A neural network is composed of an input layer consisting of a plurality of neurons, an intermediate layer consisting of a plurality of neurons, and an output layer consisting of a plurality of neurons. The intermediate layer is also referred to as a hidden layer. The intermediate layer may be one layer or two or more layers.

[0035] When learning congestion levels and pedestrian flow using planned schedule data, at least one of origin-destination data, boarding rate data, and passenger volume data, and at least one of weather data, weather forecast data, accident information, and substitute transportation information as learning data, the model generation unit 1312 generates congestion level data and pedestrian flow data from the planned schedule data output from the learning data acquisition unit 1311 and at least one of origin-destination data, boarding rate data, and passenger volume data, and learns congestion levels and pedestrian flow by performing learning with at least one of weather data, weather forecast data, accident information, and substitute transportation information as the "input" based on the combined learning data. That is, by learning the relationship between at least one of weather data, weather forecast data, accident information, and substitute transportation information with the congestion level data and pedestrian flow data created by the planned schedule data of multiple railway operators and at least one of origin-destination data, boarding rate data, and passenger volume data as the correct answer, a learned model for inferring congestion levels and pedestrian flow from at least one of weather data, weather forecast data, accident information, and substitute transportation information is generated.

[0036] When learning the delays in the line sections of railway operators using the planned schedule data and at least one of the origin-destination data, boarding rate data, and number of boarding and alighting passengers data, and at least one of the weather data, weather forecast data, event information, and transfer transportation information as learning data, the model generation unit 1312 combines the planned schedule data output from the learning data acquisition unit 1311, the delay information generated from at least one of the origin-destination data, boarding rate data, and number of boarding and alighting passengers data, and at least one of the weather data, weather forecast data, event information, and transfer transportation information. Based on the combined learning data, the model generation unit 1312 performs learning with the delay information as the "label" and at least one of the weather data, weather forecast data, event information, and transfer transportation information as the "input" to learn the delays in the line sections of railway operators. That is, using the delay information created by the planned schedule data of multiple railway operators and at least one of the origin-destination data, boarding rate data, and number of boarding and alighting passengers data as the correct answer, the model generation unit 1312 learns the relationship with at least one of the weather data, weather forecast data, event information, and transfer transportation information, thereby generating a learned model for inferring the delays in the line sections of railway operators from at least one of the weather data, weather forecast data, event information, and transfer transportation information.

[0037] When learning the amount of power consumed by railway operators using the planned schedule data and power consumption information as learning data, the model generation unit 1312 performs learning with the power consumption information as the "label" and the planned schedule data as the "input" based on the learning data obtained by combining the planned schedule data output from the learning data acquisition unit 1311 and the power consumption information. That is, using the power consumption information of multiple railway operators as the correct answer, the model generation unit 1312 learns the relationship with the planned schedule data, thereby generating a learned model for inferring the amount of power consumed by railway operators from the planned schedule data.

[0038] When learning the staff allocation situation of a railway operator based on the planned train schedule data and the staff allocation information, the model generation unit 1312 learns the staff allocation situation of the railway operator by performing learning with the staff allocation information as the "label" and the planned train schedule data as the "input" based on the learning data obtained by combining the planned train schedule data output from the learning data acquisition unit 1311 and the staff allocation information. That is, a learning model for inferring the staff allocation situation of the railway operator from the planned train schedule data is generated by learning the relationship with the planned train schedule data using the staff allocation information of a plurality of railway operators as the correct answer.

[0039] When learning the state of the vehicles owned by a railway operator using the planned train schedule data and the vehicle information as learning data, the model generation unit 1312 learns the state of the vehicles owned by the railway operator by performing learning with the vehicle information as the "label" and the planned train schedule data as the "input" based on the learning data obtained by combining the planned train schedule data and the weather data output from the learning data acquisition unit 1311. That is, a learning model for inferring the state of the vehicles owned by the railway operator from the planned train schedule data is generated by learning the relationship with the planned train schedule data using the vehicle information of a plurality of railway operators as the correct answer.

[0040] FIG. 4 is a diagram showing an example of the neural network of the learning device of the information providing device according to the first embodiment. The neural network 70 shown in FIG. 4 has a three-layer structure including an input layer 71 composed of neurons X1, X2, X3, an intermediate layer 72 composed of neurons Y1, Y2, and an output layer 73 composed of neurons Z1, Z2, Z3. When a plurality of inputs are input to the input layer 71 of the neural network 70, their values are multiplied by weights w11, w12, w13, w14, w15, w16 and input to the intermediate layer 72, and the result is further multiplied by weights w21, w22, w23, w24, w25, w26 and output from the output layer 73. This output result varies depending on the values of the weights w11, w12, w13, w14, w15, w16 and the values of the weights w21, w22, w23, w24, w25, w26.

[0041] When learning congestion levels and pedestrian flows using, as learning data, congestion data and pedestrian flow data generated from planned schedule data and at least one of origin-destination data, boarding rate data, and passenger volume data, and at least one of weather data, weather forecast data, accident information, and diversion transportation information, neural network 70 learns congestion levels and pedestrian flows through so-called supervised learning according to learning data created based on a combination of the planned schedule data acquired by learning data acquisition unit 1311, at least one of origin-destination data, boarding rate data, and passenger volume data, and at least one of weather data, weather forecast data, accident information, and diversion transportation information.

[0042] That is, neural network 70 inputs at least one of weather data, weather forecast data, accident information, and diversion transportation information into input layer 71 and adjusts weights w11, w12, w13, w14, w15, w16 and weights w21, w22, w23, w24, w25, w26 so that the result output from output layer 73 approaches at least one of the planned schedule data, origin-destination data, boarding rate data, and passenger volume data, thereby performing learning.

[0043] When learning delays in a railway operator's section using, as learning data, planned schedule data, delay information generated from at least one of origin-destination data, boarding rate data, and passenger volume data, and at least one of weather data, weather forecast data, event information, and diversion transportation information, neural network 70 learns delays in the railway operator's section through so-called supervised learning according to learning data created based on a combination of the planned schedule data acquired by learning data acquisition unit 1311, delay information generated from at least one of origin-destination data, boarding rate data, and passenger volume data, and at least one of weather data, weather forecast data, accident information, and diversion transportation information.

[0044] That is, the neural network 70 learns by adjusting the weights w11, w12, w13, w14, w15, w16 and the weights w21, w22, w23, w24, w25, w26 so that the result output from the output layer 73 approaches the delay information when at least one of weather data, weather forecast data, event information, and rescheduling transport information is input to the input layer 71.

[0045] When learning the amount of power consumed by a railway operator using the train schedule data and power consumption information as learning data, the neural network 70 learns the amount of power consumed by the railway operator by so-called supervised learning according to the learning data created based on the combination of the train schedule data and power consumption information acquired by the learning data acquisition unit 1311.

[0046] That is, the neural network 70 learns by adjusting the weights w11, w12, w13, w14, w15, w16 and the weights w21, w22, w23, w24, w25, w26 so that the result output from the output layer 73 approaches the power consumption information when the train schedule data is input to the input layer 71.

[0047] When learning the staffing situation of a railway operator based on the train schedule data and the staffing information, the neural network 70 learns the amount of power consumed by the railway operator by so-called supervised learning according to the learning data created based on the combination of the train schedule data and the staffing information acquired by the learning data acquisition unit 1311.

[0048] That is, the neural network 70 learns by adjusting the weights w11, w12, w13, w14, w15, w16 and the weights w21, w22, w23, w24, w25, w26 so that the result output from the output layer 73 approaches the staffing information when the train schedule data is input to the input layer 71.

[0049] When the neural network 70 learns the state of the vehicles owned by the railway operator using the planned train schedule data and vehicle information as learning data, it learns the state of the vehicles owned by the railway operator by so-called supervised learning according to the learning data created based on the combination of the planned train schedule data and vehicle information acquired by the learning data acquisition unit 1311.

[0050] That is, the neural network 70 learns by adjusting the weights w11, w12, w13, w14, w15, w16 and the weights w21, w22, w23, w24, w25, w26 so that the result output from the output layer 73 when the planned train schedule data is input to the input layer 71 approaches the vehicle information.

[0051] The model generation unit 1312 generates and outputs a learned model by executing the learning as described above.

[0052] The learned model storage unit 132 stores the learned model output from the model generation unit 1312.

[0053] Hereinafter, the learning process will be described by taking as an example the case where the congestion level and the flow of people are learned using, as learning data, the planned train schedule data, and at least one of the origin-destination data, the boarding rate data, and the number of passengers getting on and off, and the congestion level data and the flow of people data generated therefrom, and at least one of the weather data, the weather forecast data, the accident information, and the substitute transportation information.

[0054] FIG. 5 is a flowchart relating to the learning process by the learning device of the information providing device according to Embodiment 1. The learning process by the learning device 131 will be described with reference to FIG. 5.

[0055] In step S11, the learning data acquisition unit 1311 acquires, as learning data, at least one of schedule diagram data, and at least one of origin-destination data, boarding rate data, and number of boarding and alighting passengers data, and at least one of weather data, weather forecast data, accident information, and substitute transportation information. Although it was assumed that at least one of schedule diagram data and at least one of weather data, weather forecast data, accident information, and substitute transportation information were acquired simultaneously, it is sufficient that at least one of schedule diagram data, and at least one of origin-destination data, boarding rate data, and number of boarding and alighting passengers data, and at least one of weather data, weather forecast data, accident information, and substitute transportation information can be input in association with each other. They may be acquired at different timings as long as at least one of schedule diagram data, and at least one of origin-destination data, boarding rate data, and number of boarding and alighting passengers data, and at least one of weather data, weather forecast data, accident information, and substitute transportation information are associated with each other.

[0056] In step S12, the model generation unit 1312 learns congestion and pedestrian flow by so-called supervised learning according to the learning data created based on a combination of at least one of the schedule diagram data, and at least one of origin-destination data, boarding rate data, and number of boarding and alighting passengers data, and at least one of weather data, weather forecast data, accident information, and substitute transportation information acquired by the learning data acquisition unit 1311, and generates a learned model.

[0057] In step S13, the learned model storage unit 132 stores the learned model generated by the model generation unit 1312.

[0058] In the above example, the learning process of learning congestion and pedestrian flow using, as learning data, at least one of schedule diagram data, and at least one of origin-destination data, boarding rate data, and number of boarding and alighting passengers data, and at least one of weather data, weather forecast data, accident information, and substitute transportation information has been described. However, the learning process in the case of performing learning other than congestion and pedestrian flow is also performed by the same procedure.

[0059] The same processing is performed whether the learning data includes all of the exemplified data and information or only a part of the exemplified data and information.

[0060] FIG. 6 is a diagram showing the configuration of the inference device of the information providing apparatus according to Embodiment 1. The inference device 133 includes an inference data acquisition unit 1331 and an inference unit 1332.

[0061] The inference data acquisition unit 1331 acquires, as inference data, the operation-related data and operation influence cause data newly collected by the data collection unit 11.

[0062] When inferring congestion and pedestrian flow from at least one of weather data, weather forecast data, accident information, and transfer transportation information, the inference data acquisition unit 1331 acquires at least one of weather data, weather forecast data, accident information, and transfer transportation information.

[0063] The inference unit 1332 infers congestion and pedestrian flow using the learned model. That is, the inference unit 1332 inputs at least one of the weather data, weather forecast data, accident information, and transfer transportation information acquired by the inference data acquisition unit 1331 to the learned model, and outputs, as inference results, the congestion and pedestrian flow inferred from at least one of the weather data, weather forecast data, accident information, and transfer transportation information.

[0064] Although the description has been made on the assumption that the congestion and pedestrian flow are output using the learned model learned by the model generation unit 1312 of the information providing apparatus 10, a learned model may be acquired from outside the information providing apparatus 10, and the congestion and pedestrian flow may be output based on this learned model.

[0065] When inferring a delay in a section of a railway operator from at least one of weather data, weather forecast data, event information, and transfer transportation information, the inference data acquisition unit 1331 acquires at least one of weather data, weather forecast data, event information, and transfer transportation information.

[0066] The inference unit 1332 infers delays in the railway operator's section using the learned model. That is, by inputting at least one of the weather data, weather forecast data, event information, and diversion transport information acquired by the inference data acquisition unit 1331 into the learned model, the delay in the railway operator's section inferred from at least one of the weather data, weather forecast data, event information, and diversion transport information is output as an inference result.

[0067] Although it has been described that the learned model learned by the model generation unit 1312 of the information providing device 10 is used to output the delay in the railway operator's section, a learned model may be acquired from outside the information providing device 10, and the delay in the railway operator's section may be output based on this learned model.

[0068] When inferring the amount of power consumed by the railway operator from the planned train schedule data, the inference data acquisition unit 1331 acquires the planned train schedule data.

[0069] The inference unit 1332 infers the amount of power consumed by the railway operator using the learned model. That is, by inputting the planned train schedule data acquired by the inference data acquisition unit 1331 into the learned model, the amount of power consumed by the railway operator inferred from the planned train schedule data is output as an inference result.

[0070] Although it has been described that the learned model learned by the model generation unit 1312 of the information providing device 10 is used to output the amount of power consumed by the railway operator, a learned model may be acquired from outside the information providing device 10, and the amount of power consumed by the railway operator may be output based on this learned model.

[0071] When inferring the staff allocation situation of the railway operator from the planned train schedule data, the inference data acquisition unit 1331 acquires the planned train schedule data.

[0072] The inference unit 1332 infers the staffing situation of railway operators using the learned model. That is, by inputting the planned train schedule data acquired by the inference data acquisition unit 1331 into the learned model, the staffing situation of railway operators inferred from the planned train schedule data is output as an inference result.

[0073] Although the description has been given of outputting the staffing situation of railway operators using the learned model learned by the model generation unit 1312 of the information providing apparatus 10, it is also possible to acquire a learned model from outside the information providing apparatus 10 and output the staffing situation of railway operators based on this learned model.

[0074] When inferring the state of the vehicles owned by a railway operator from the planned train schedule data, the inference data acquisition unit 1331 acquires the planned train schedule data.

[0075] The inference unit 1332 infers the state of the vehicles owned by a railway operator using the learned model. That is, by inputting the planned train schedule data acquired by the inference data acquisition unit 1331 into the learned model, the state of the vehicles owned by a railway operator inferred from the planned train schedule data is output as an inference result.

[0076] Although the description has been given of outputting the state of the vehicles owned by a railway operator using the learned model learned by the model generation unit 1312 of the information providing apparatus 10, it is also possible to acquire a learned model from outside the information providing apparatus 10 and output the state of the vehicles owned by a railway operator based on this learned model.

[0077] Hereinafter, the inference process will be described by taking as an example the case of inferring the congestion level and the flow of people from at least one of the weather data, weather forecast data, accident information, and substitute transportation information.

[0078] FIG. 7 is a flowchart showing the flow of the inference process by the inference device of the information providing apparatus according to Embodiment 1.

[0079] In step S21, the inference data acquisition unit 1331 acquires at least one of weather data, weather forecast data, accident information, and substitute transportation information as inference data.

[0080] In step S22, the inference unit 1332 inputs at least one of the weather data, weather forecast data, accident information, and substitute transportation information acquired by the inference data acquisition unit 1331 into the learned model stored in the learned model storage unit 132, and infers the congestion level and the flow of people.

[0081] In step S23, the inference unit 1332 outputs the congestion level and the flow of people obtained by the learned model to the prediction result data storage unit 14.

[0082] In step S24, the prediction result data storage unit 14 stores the congestion level and the flow of people as prediction result data.

[0083] The data transmission unit 15 transmits the congestion level and the flow of people stored in the prediction result data storage unit 14 in response to a request from at least one of the operation management device 20, the user terminal 40, and the service provider terminal 50. Thereby, the information providing device 10 can provide highly accurate data on the congestion level and the flow of people accumulated in the prediction result data storage unit 14 to railway operators and the like.

[0084] In the above example, the inference process of inferring the congestion level and the flow of people from at least one of the weather data, weather forecast data, accident information, and substitute transportation information has been described. However, the inference process in the case of performing inferences other than the congestion level and the flow of people is also performed by the same procedure.

[0085] The same processing is performed whether the inference data includes all of the exemplified data and information or only a part of the exemplified data and information.

[0086] Although the case where supervised learning is applied to the learning algorithm used by the model generation unit 1312 has been described, it is not limited to this. Regarding the learning algorithm, in addition to supervised learning, semi-supervised learning or the like can also be applied.

[0087] In addition, the model generation unit 1312 can also add or remove railway operators that collect learning data as targets midway. Furthermore, the learning device 131 that has learned the congestion level and the flow of people for a certain railway operator may be applied to another information providing device 10, and the congestion level and the flow of people may be relearned and updated for the other information providing device.

[0088] Also, as the learning algorithm used in the model generation unit 1312, deep learning that learns the extraction of the feature quantities themselves such as origin-destination data, boarding rate data, number of passengers getting on and off, planned train schedule data, weather data, weather forecast data, accident information, event information, and transfer transportation information can be used, or machine learning may be executed according to other known methods, for example, genetic programming, functional logic programming, support vector machine, etc.

[0089] Note that based on the comparison result between the prediction result data at a certain past time point stored in the prediction result data storage unit 14 and the prediction result data at a time point after the time when the inference was made, the values of the weights w11, w12, w13, w14, w15, w16 and the values of the weights w21, w22, w23, w24, w25, w26 are adjusted, and a process of correcting the learned model is performed, whereby the learning accuracy can be improved.

[0090] Note that the learning device 131 and the inference device 133 may be devices separate from the information providing device 10 connected to the information providing device 10 via a network. Also, the learning device 131 and the inference device 133 may exist on a cloud server different from the information providing device 10.

[0091] The information providing system 100 according to Embodiment 1 can make a more accurate prediction regarding the future operation of a railway than when making a prediction regarding the future operation of a railway based on operation-related data collected from a single railway operator, because it makes a prediction regarding the future operation of a railway based on operation-related data collected from a plurality of railway operators. Therefore, the information providing system 100 according to Embodiment 1 can provide highly accurate prediction result data to railway users, station facility users, users of facilities around stations, service providers, and railway operators. For example, when making a prediction regarding the future operation of a railway based on operation-related data collected from a single railway operator, if an operation is affected by an external factor such as an accident in a section of a railway operator that is a reciprocal access partner, the actual operation situation may deviate from the predicted operation situation. Since the information providing system 100 according to Embodiment 1 makes a prediction regarding the future operation of a railway based on operation-related data collected from a plurality of railway operators, even when an operation is affected by an external factor, it can make a prediction close to the actual operation situation, and thus can provide highly accurate prediction result data to railway users, station facility users, users of facilities around stations, service providers, and railway operators.

[0092] In addition, railway users, station facility users, users of surrounding facilities, railway operators, and service providers that have received the prediction result data transmitted from the data transmission unit 15 can act based on the received prediction result data and based on the future prediction of the railway operation. For example, railway users, station facility users, and users of facilities around stations can select and use a route predicted to have a low congestion level, or can select and use a route predicted to have a small delay. In addition, railway operators can optimize the staffing arrangement of employees or take measures to suppress the power consumption to a certain level or below. In addition, service providers can provide highly accurate route guidance services to railway users or provide highly accurate marketing information to operators who operate stores at stations or around stations.

[0093] The hardware configuration of the prediction unit 13 of the information providing apparatus 10 according to Embodiment 1 will be described.

[0094] FIG. 8 is a diagram showing the hardware configuration of the prediction unit of the information providing apparatus according to Embodiment 1. The prediction unit 13 is realized by a computer system including a processor 91 that executes various processes, a memory 92 that is a main memory, and a storage device 93 that stores information.

[0095] The processor 91 may be an arithmetic means such as an arithmetic unit, a microprocessor, a microcomputer, a CPU (Central Processing Unit), or a DSP (Digital Signal Processor). Also, as the memory 92, a non-volatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), or an EEPROM (registered trademark) (Electrically Erasable Programmable Read Only Memory) can be used. The storage device 93 stores a program for executing a learning process and an inference process for making a future prediction regarding the operation of a railway.

[0096] The above computer system realizes the functions of the prediction unit 13 by the processor 91 reading out the programs corresponding to the processes of the respective components stored in the storage device 93 into the memory 92 and executing them. Also, the memory 92 is also used as a temporary memory in each process executed by the processor 91. The programs executed by the processor 91 may be provided in a state stored in a storage medium or may be provided via a network.

[0097] The configurations shown in the above embodiments are examples of the content, and it is possible to combine them with other known techniques, and it is also possible to omit or change a part of the configuration without departing from the gist.

[0098] Hereinafter, aspects of the present disclosure will be summarized in the appendices.

[0099] (Appendix 1) A data collection unit that collects operation-related data regarding the operation of railways from a plurality of railway operators, An accumulated data storage unit that stores the operation-related data collected by the data collection unit, A prediction unit that makes a future prediction regarding the operation of the railway based on the operation-related data collected from the plurality of railway operators, An information providing apparatus, characterized by comprising the above. (Appendix 2) The operation-related data is the planned schedule data of each of the plurality of railway operators and the amount of power consumed by each of the plurality of railway operators, The prediction unit predicts the amount of power consumed by each of the plurality of railway operators. The information providing apparatus according to Appendix 1, characterized by this. (Appendix 3) The operation-related data is the planned schedule data of each of the plurality of railway operators and the staff allocation information of each of the plurality of railway operators, The prediction unit predicts the staff allocation situation of each of the plurality of railway operators. The information providing apparatus according to Appendix 1, characterized by this. (Appendix 4) The operation-related data is the planned schedule data of each of the plurality of railway operators, the start and end point data indicating the entry and exit stations of railway users of each of the plurality of railway operators, the boarding rate data of trains operated by each of the plurality of railway operators, at least one of the boarding and alighting passenger numbers data of each of the plurality of railway operators, and at least one of the accident information and the substitute transportation information of each of the plurality of railway operators, The prediction unit predicts the congestion level and the flow of people. The information providing apparatus according to Appendix 1, characterized by this. (Appendix 5) The operation-related data includes at least one of the planned timetable data of each of the plurality of railway operators, the origin and destination data indicating the entry and exit stations of railway users of each of the plurality of railway operators, the boarding rate data of trains operated by each of the plurality of railway operators, and the number of boarding and alighting passengers data of each of the plurality of railway operators, and the transfer transportation information of each of the plurality of railway operators. The prediction unit is the information providing device according to appended note 1, characterized by predicting railway delays. (Appended note 6) The operation-related data is the planned timetable data of each of the plurality of railway operators and the vehicle information indicating the state of the vehicles operated by each of the plurality of railway operators. The prediction unit is the information providing device according to appended note 1, characterized by predicting the state of the vehicles operated by each of the plurality of railway operators. (Appended note 7) The prediction unit The information providing device according to any one of appended notes 1 to 6, characterized by making a future prediction regarding the operation of the railway using the operation-related data and a learned model that makes a future prediction based on the operation-related data. (Appended note 8) The prediction unit is the information providing device according to appended note 7, characterized by creating the learned model based on the operation-related data including the planned timetable data of each of the plurality of railway operators. (Appended note 9) The prediction unit is the information providing device according to appended note 7, characterized by creating the learned model based on the operation-related data including the planned timetable data of each of the plurality of railway operators accumulated in the accumulated data storage unit. (Appended note 10) The data collection unit further collects operation influence cause data that causes an impact on the operation of the railway. The accumulated data storage unit further accumulates the operation influence cause data collected by the data collection unit. The prediction unit performs a future prediction regarding the operation of the railway based on the operation-related data and the operation influence cause data collected from the plurality of railway operators, which is the information providing apparatus according to appended note 1. (Appended note 11) The operation-related data is at least one of the planned schedule data of each of the plurality of railway operators, the start and end point data indicating the entry station and the exit station of the railway users of each of the plurality of railway operators, the boarding rate data of the trains operated by each of the plurality of railway operators, and the number of boarding and alighting passengers data of each of the plurality of railway operators. The operation influence cause data is at least one of weather data and weather forecast data. The prediction unit predicts the congestion level and the flow of people, which is the information providing apparatus according to appended note 10. (Appended note 12) The operation-related data is at least one of the planned schedule data of each of the plurality of railway operators, the start and end point data indicating the entry station and the exit station of the railway users of each of the plurality of railway operators, the boarding rate data of the trains operated by each of the plurality of railway operators, and the number of boarding and alighting passengers data of each of the plurality of railway operators. The operation influence cause data is at least one of weather data, weather forecast data, and event information indicating events related to the railway business. The prediction unit predicts the delay of the railway, which is the information providing apparatus according to appended note 10. (Appended note 13) The prediction unit uses the operation-related data and the operation influence cause data, and a learned model that makes a future prediction based on the operation-related data and the operation influence cause data, to perform a future prediction regarding the operation of the railway, which is the information providing apparatus according to any one of appended notes 10 to 12. (Appended note 14) The prediction unit creates the learned model based on the operation-related data including the planned schedule data of each of the plurality of railway operators and the operation influence cause data, which is the information providing apparatus according to appended note 13. (Appended note 15) The prediction unit creates the learned model based on the operation-related data including the planned train schedule data of each of the plurality of railway operators stored in the accumulated data storage unit and the operation influence cause data, the information providing apparatus according to appended note 13. (Appended note 16) The information providing apparatus according to any one of appended notes 7 to 9 and appended notes 13 to 15, further comprising a prediction result data storage unit that stores prediction result data indicating the result of prediction by the prediction unit. The information providing apparatus according to any one of appended notes 7 to 9 and appended notes 13 to 15, further comprising a prediction result data storage unit that stores prediction result data indicating the result of prediction by the prediction unit, and correcting the learned model using the prediction result data stored in the prediction result data storage unit. (Appended note 17) The information providing apparatus according to any one of appended notes 1 to 16, further comprising a data transmission unit that transmits prediction result data indicating the result of prediction by the prediction unit to the outside. (Appended note 18) The information providing apparatus according to appended note 17, wherein the data transmission unit transmits the prediction result data to at least any one of an operation management apparatus used by each of the plurality of railway operators, a user terminal of a user who uses the railway, a station, or a peripheral facility, and a service provider terminal of a service provider.

Explanation of reference numerals

[0100] 10 Information providing apparatus, 11 Data collection unit, 12 Accumulated data storage unit, 13 Prediction unit, 14 Prediction result data storage unit, 15 Data transmission unit, 20 Operation management apparatus, 30 External server, 40 User terminal, 50 Service provider terminal, 70 Neural network, 71 Input layer, 72 Intermediate layer, 73 Output layer, 91 Processor, 92 Memory, 93 Storage device, 100 Information providing system, 131 Learning device, 132 Learned model storage unit, 133 Inference device, 1311 Learning data acquisition unit, 1312 Model generation unit, 1331 Inference data acquisition unit, 1332 Inference unit.

Claims

1. A data collection unit that collects operation-related data regarding railway operations from a plurality of railway operators, An accumulated data storage unit that stores the operation-related data collected by the data collection unit, A prediction unit that makes a future prediction regarding the operation of the railway based on the operation-related data collected from the plurality of railway operators, An information providing device, characterized by comprising the above.

2. The operation-related data is the planned schedule data of each of the plurality of railway operators and the amount of power consumed by each of the plurality of railway operators, The prediction unit predicts the amount of power consumed by each of the plurality of railway operators. The information providing device according to claim 1, characterized by this.

3. The operation-related data is the planned schedule data of each of the plurality of railway operators and the staff allocation information of each of the plurality of railway operators, The prediction unit predicts the staff allocation situation of each of the plurality of railway operators. The information providing device according to claim 1, characterized by this.

4. The operation-related data is the planned schedule data of each of the plurality of railway operators, the start and end point data indicating the entry and exit stations of railway users of each of the plurality of railway operators, the boarding rate data of the trains operated by each of the plurality of railway operators, at least one of the boarding and alighting passenger count data of each of the plurality of railway operators, and at least one of the accident information and substitute transportation information of each of the plurality of railway operators, The prediction unit predicts congestion and the flow of people. The information providing device according to claim 1, characterized by this.

5. The operation-related data is the planned schedule data of each of the plurality of railway operators, the start and end point data indicating the entry and exit stations of railway users of each of the plurality of railway operators, the boarding rate data of the trains operated by each of the plurality of railway operators, at least one of the boarding and alighting passenger count data of each of the plurality of railway operators, and the substitute transportation information of each of the plurality of railway operators, The prediction unit predicts delays of the railway. The information providing device according to claim 1, characterized by this.

6. The operation-related data is the planned schedule data of each of the plurality of railway operators and vehicle information indicating the state of the vehicles operated by each of the plurality of railway operators, The prediction unit predicts the state of the vehicles operated by each of the plurality of railway operators. The information providing device according to claim 1, characterized by this.

7. The prediction unit The information providing apparatus according to claim 1, wherein future prediction regarding railway operation is performed using the operation-related data and a learned model that makes future predictions based on the operation-related data.

8. The information providing apparatus according to claim 7, wherein the prediction unit creates the learned model based on operation-related data including planned schedule data of each of the plurality of railway operators.

9. The information providing apparatus according to claim 7, wherein the prediction unit creates the learned model based on operation-related data including planned schedule data of each of the plurality of railway operators stored in the accumulated data storage unit.

10. The data collection unit further collects operation influence cause data that is a cause affecting the operation of the railway, The accumulated data storage unit further accumulates the operation influence cause data collected by the data collection unit, The information providing apparatus according to claim 1, wherein the prediction unit makes a future prediction regarding the operation of the railway based on the operation-related data and the operation influence cause data collected from the plurality of railway operators.

11. The operation-related data is at least one of planned schedule data of each of the plurality of railway operators, start and end point data indicating the entry and exit stations of railway users of each of the plurality of railway operators, boarding rate data of trains operated by each of the plurality of railway operators, and boarding and alighting passenger count data of each of the plurality of railway operators, The operation influence cause data is at least one of weather data and weather forecast data, The information providing apparatus according to claim 10, wherein the prediction unit predicts congestion and passenger flow.

12. The operation-related data is at least one of planned schedule data of each of the plurality of railway operators, start and end point data indicating the entry and exit stations of railway users of each of the plurality of railway operators, boarding rate data of trains operated by each of the plurality of railway operators, and boarding and alighting passenger count data of each of the plurality of railway operators, The operation influence cause data is at least one of weather data, weather forecast data, and event information indicating events related to the railway business, The information providing apparatus according to claim 10, wherein the prediction unit predicts delays of the railway.

13. The prediction unit An information providing apparatus according to claim 10, wherein future prediction regarding the operation of the railway is performed using the operation-related data and the operation influence cause data, and a learned model that performs future prediction based on the operation-related data and the operation influence cause data.

14. The information providing apparatus according to claim 13, wherein the prediction unit creates the learned model based on the operation-related data and the operation influence cause data including the planned train diagram data of each of the plurality of railway operators.

15. The information providing apparatus according to claim 13, wherein the prediction unit creates the learned model based on the operation-related data and the operation influence cause data including the planned train diagram data of each of the plurality of railway operators stored in the accumulated data storage unit.

16. A prediction result data storage unit that stores prediction result data indicating the result of prediction by the prediction unit, The information providing apparatus according to claim 7, wherein the learned model is corrected using the prediction result data stored in the prediction result data storage unit.

17. The information providing apparatus according to any one of claims 1 to 16, further comprising a data transmission unit that transmits prediction result data indicating the result of prediction by the prediction unit to the outside.

18. The data transmission unit according to claim 17, wherein the prediction result data is transmitted to at least one of an operation management device used by each of the plurality of railway operators, a user terminal owned by a user who uses the railway, a station, or a surrounding facility, and a service provider terminal owned by a service provider.

19. An operation management device owned by each of a plurality of railway operators, An information providing apparatus having a data collection unit that collects operation-related data regarding the operation of the railway from the operation management device, an accumulated data storage unit that stores the operation-related data collected by the data collection unit, and a prediction unit that performs future prediction regarding the operation of the railway based on the operation-related data collected from the plurality of railway operators. An information providing system, characterized by comprising:

20. A data collection step of collecting operation-related data regarding the operation of the railway from a plurality of railway operators, An accumulated data storage step of storing the operation-related data collected in the data collection step, A prediction step of making a future prediction regarding the operation of the railway based on the operation-related data collected from the plurality of railway operators; An information providing method, characterized by comprising the above.

21. A data collection step of collecting operation-related data regarding the operation of the railway from a plurality of railway operators; An accumulated data storage step of accumulating the operation-related data collected in the data collection step; A prediction step of making a future prediction regarding the operation of the railway based on the operation-related data collected from the plurality of railway operators; An information providing program, characterized by causing a computer to execute the above.

Images