JP2018010454A - Congestion degree prediction device, congestion degree prediction information delivery system, and congestion degree prediction method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide technology that calculates information about the congestion state of a station and a train providing a high level of convenience to railroad users.SOLUTION: A congestion degree prediction device of the present invention comprises: a boarding-alighting station prediction computation unit for generating boarding-alighting station prediction information that indicates a time of day and a station at which users have entered and a time of day and a station at which the users are predicted to exit, on the basis of seat reservation information that indicates previously made seat reservations and ticket information read from tickets by a ticket examination machine; a station congestion degree prediction computation unit for generating station congestion degree prediction information in which the number of users in premises at each future time of day at each station is predicted, on the basis of the boarding-alighting station prediction information and a prediction diagram that indicates an arrival time and a departure time at each station of each future train; and a train congestion degree prediction computation unit for generating train congestion degree prediction information in which the number of users in cars at each future time of each train is predicted, on the basis of the station congestion degree prediction information and the number of people in car information that indicates the present boarding state of each train.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for providing railway users with information on the congestion status of trains and stations in a railway.

  An information distribution system that calculates the congestion situation of stations and trains in urban areas where the railway network is large and complicated, and distributes it to railway users using WEB contents etc. is attracting attention, and research and development are actively conducted Yes. Aggregating and accumulating the ticket information of railway users obtained from the ticket gate at the station and the schedule information obtained from the train operation management system, and analyzing the information using methods such as data mining Thus, the behavior pattern of the railway user can be calculated. And the congestion situation of a station or a train can be calculated based on the information of a railway user's action pattern. Information on the congestion status is distributed to railway users and the like using WEB content and the like.

  Moreover, in the field of railway operation management, it is possible to provide information for supporting the operation arrangement work to the transport commander. In the operation management system, it is possible to obtain information on a prediction diagram by simulating the operation of a train based on constant values such as the time at which the station stops and the minimum operation interval with the following train, and the train running record. By visually displaying the prediction diagram on the transport commander before the train travels, the operation arrangement work by the transport commander is made more efficient.

  In Patent Document 1, by using the ticket gate time and commuter pass information obtained by the ticket gate when a railway user passes the ticket gate, the departure or arrival station having a prominent frequency is estimated as the predicted destination station. Furthermore, the congestion level is totaled.

JP 2014-213697 A

  The congestion degree information obtained by the method of Patent Document 1 is a total of human flows between two points of a departure station and an arrival station from a macro viewpoint. Therefore, the expected congestion level may deviate from the congestion level at the micro level of the station and train that the user actually encounters and experiences, and is highly convenient for railway users. It's not information.

  An object of the present invention is to provide a technique for calculating information on a congestion state of a station and a train that is highly convenient for railway users.

  In order to achieve the above object, the congestion degree predicting device of the present invention is based on seat reservation information indicating a seat reservation made in advance and ticket information read from a ticket by a ticket gate. Boarding station prediction calculation unit for generating boarding / alighting station prediction information indicating the time and station where the user entered and the time and station where the user is expected to enter, the boarding station prediction information, and arrival of each future train at each station A station congestion degree prediction calculation unit that generates station congestion degree prediction information that predicts the number of users on the premises at each future time of each station based on a prediction diagram that indicates a time and a departure time; and the station congestion degree prediction Train congestion degree prediction calculation that generates train congestion degree prediction information that predicts the number of passengers in each vehicle at each future time based on the information and the number of passengers in the train that indicates the current riding situation of each train And having .

  According to the present invention, the boarding behavior of each user is predicted as boarding station prediction information, the future congestion of each station is predicted as station congestion degree prediction information based on the boarding station prediction information and the prediction diagram, and the station congestion degree The future congestion of each train is predicted as train congestion degree prediction information based on the prediction information and the number of passengers in each train, so the future congestion of each station and each train can be determined based on the behavior of individual users and the individual trains. Based on the prediction of operation, it can be predicted with high accuracy. As a result, it is possible to generate information that is highly convenient for railway users.

It is a block diagram of the congestion degree prediction information delivery system by this embodiment. It is a figure which shows an example of a seat reservation information table. It is a figure which shows an example of a boarding information table. It is a figure which shows an example of a commuter pass information table. It is a figure which shows an example of a utilization performance table. It is a flowchart which shows the process of the boarding / alighting station prediction calculating part 111. It is a figure which shows an example of the boarding / alighting station prediction information table 510 in which the boarding / alighting station prediction calculating part 111 stores information. It is a figure which shows an example of an event information table. It is a figure which shows an example of a station congestion information table. It is a figure which shows an example of a prediction diamond information table. It is a flowchart which shows the process of the station congestion degree prediction calculating part 112. FIG. It is a figure which shows an example of a station congestion degree prediction information table. It is a figure which shows an example of a passenger number information table in a vehicle. It is a flowchart which shows the process of a train congestion degree prediction calculating part. It is a figure which shows an example of a train congestion degree prediction information table.

  FIG. 1 is a block diagram of a congestion degree prediction information distribution system according to this embodiment. Referring to FIG. 1, a congestion degree prediction information distribution system 100 is an information collection and distribution system in which a plurality of computers are connected to a network as an example. The congestion degree prediction server 101, a seat reservation system 102, an IC ticket system 103, an event An input terminal 104, a passenger count server 105, an operation management system 106, a vehicle information system 107, and an information distribution system 108 are included.

  The congestion level prediction server 101 includes a boarding / alighting station prediction calculation unit 111, a station congestion level prediction calculation unit 112, and a train congestion level prediction calculation unit 113. Further, the congestion degree prediction server 101 includes a management database (not shown), and seat reservation information 114, ticket information 115, boarding / departure station prediction information 116, event information 117, station congestion information 118, prediction diagram 119, and station congestion degree prediction. Information 120, in-car number information 121, and train congestion degree prediction information 122 are stored and managed in a management database.

  The event input terminal 104 includes a CPU, a disk, and an input / output device, and inputs event information 117. As the event information 117, it is possible for the user to register information such as the event execution date, the station name, the entrance time (plan), and the number of visitors (plan). The number of visitors is a predicted value of the number of users who enter the station at the end of the event. The input information is registered and managed in the event information table.

  FIG. 2 is a diagram illustrating an example of a seat reservation information table. The seat reservation information 114 is information related to the seat reservation acquired from the seat reservation system 102, and is stored in the seat reservation information table 210 that can be referred to from the boarding / alighting station prediction calculation unit 111 illustrated in FIG. When the seat reservation is made, the boarding station and the getting-off station for the ticket including the section where the seat is not reserved may be different from the boarding station and the getting-off station for boarding the train for which the seat is reserved. Therefore, in the seat reservation information 114, for each entry, the ID 211 given in the order stored in the seat reservation information table 210, the enforcement date 212 of the train that reserved the seat, the boarding station 213 of the ticket, the ticket An exit station 214, train information 215 including a schedule of the train, the train boarding station 216 that reserves a seat, and the train exit station 217 are included.

  FIG. 3A is a diagram illustrating an example of a boarding information table. FIG. 3B is a diagram illustrating an example of a commuter pass information table. FIG. 3C is a diagram illustrating an example of a usage record table.

  The boarding ticket information 115 includes boarding ticket information and commuter pass information distributed from the IC boarding ticket system 103 and information on the use results of the IC boarding ticket. The ticket information is information related to the entrance of a user who has entered the station through an automatic ticket gate using an IC ticket. The usage record is information related to the record of entering the entrance station with the IC ticket and entering from the entry station. The commuter pass information is information related to the commuter pass purchased by the user specifying the start station, end station and period of use.

  The ticket information is stored in the ticket information table 310 illustrated in FIG. 3A, the commuter ticket information is stored in the commuter ticket information table 320 illustrated in FIG. 3B, and the use record of the IC ticket is illustrated in FIG. 3C. Stored in the actual usage information table 330. These tables can be referred to from the boarding / alighting station prediction calculation unit 111.

  The ticket information includes a search ID 311 that combines an ID unique to an IC ticket and an ID given in the order stored in the ticket information table 310, and an enforcement date recorded when the user passes the automatic ticket gate 312, an entrance station 313, an entrance time 314, and an entrance flag 315. The entrance flag 315 is changed from 0 to 1 when the user passes through the automatic ticket gate and enters the station, and is changed from 1 to 0 when the user enters.

  The commuter pass information includes a search ID 321 combining an ID unique to an IC ticket and an ID given in the order stored in the commuter pass information table, a use start date (commuter pass start date) 322, an end date ( Commuter pass end date) 323, commuter pass start station (1) 324, end station (1) 325, start station (2) 326, end station (2) 327, and flag 328 are included. Here, two sections (1) and (2) can be registered in one commuter pass.

  The usage record information includes a search ID 331 combining an ID specific to an IC ticket and an ID given in the order stored in the usage record table, an enforcement date 332 for entering and participating using an IC ticket, and an entrance station (Actual result) 333, entrance time (actual) 334, participating station (actual) 335, participating time (actual) 336, and estimated transfer station 337 are included. In the case where a plurality of sections are used for one entry and participation, an estimated transfer station 337 registered in advance is recorded.

  FIG. 4 is a flowchart showing processing of the boarding / alighting station prediction calculation unit 111. FIG. 5 is a diagram illustrating an example of the boarding / alighting station prediction information table 510 in which the boarding / alighting station prediction calculation unit 111 stores information.

  The boarding / alighting station prediction information table 510 records a participating station and a participating time at which a user who has entered the entering station at the entering time will participate. Referring to FIG. 5, the boarding station prediction information table 510 includes a search ID 511 that combines an ID specific to an IC ticket and an ID given in the order stored in the boarding station prediction information table 510, and a railway use enforcement date 512. The entry station 513, the entry time 514, the entry station (plan) 515, the entry time (plan) 516, and the estimated transfer station 517 of the IC ticket are stored as the exit station prediction information 116. At the admission station 513 and the admission time 514, not only the results of the IC ticket actually passing through the automatic ticket gate included in the ticket information table 310, but also the boarding of the seat reservation information table 210 where the seat reservation was made The entry time to the station estimated to be the station 213 is also registered. The boarding / alighting station prediction information table 510 can be referred to from the boarding / alighting station prediction calculation unit 111, the station congestion degree prediction calculation unit 112, and the train congestion degree prediction calculation unit 113.

  The boarding / alighting station prediction calculation unit 111 accesses the seat reservation system 102 to acquire the train information 215, acquires a train diagram (not shown) indicated by the train information 215, and arrives at the departure time and departure of the train indicated in the diagram By adding or subtracting a predetermined time to or from the time, the entry time when the user enters the boarding station, that is, the entrance station, to get on the train at the boarding station, and the exit station, that is, the exit, after getting off the train at the getting off station Estimate the participation time from the station. Then, the boarding / alighting station prediction calculation unit 111 registers the boarding / alighting station prediction information in the boarding / alighting station prediction information table 510.

  Moreover, the boarding / alighting station prediction calculation part 111 is based on the information of the boarding ticket information table 310, the commuter pass information table 320, and the utilization track record information table 330, as shown in the processing flow 400 of FIG. Is registered in the boarding / alighting station prediction information table 510. Hereinafter, the processing flow 400 of FIG. 4 will be described in detail.

  The boarding / alighting station prediction calculation unit 111 first acquires the boarding ticket information table 310 in step 401. Subsequently, the boarding / alighting station prediction calculation unit 111 acquires the commuter pass information table 320 in step 402. Furthermore, the boarding / alighting station prediction calculation unit 111 acquires the usage record information table 330 in step 403.

  In step 404, the boarding / alighting station prediction calculation unit 111 searches and acquires the ticket information table 310 in units of one record. Furthermore, if the entrance flag 315 of the acquired record is 1 in step 405, the boarding / alighting station prediction calculation unit 111 performs the processing from step 406 onward, and if the entrance flag 315 is 0, returns to step 404 and returns to the next boarding Search ticket information.

  In step 406, the boarding / alighting station prediction calculation unit 111 searches the commuter pass information 320 for the commuter pass information of the IC ticket of the acquired record, and acquires it. Subsequently, in step 407, the boarding station prediction calculation unit 111 determines that the entrance station 313 in the record acquired in the ticket information table 310 is the commuter ticket start station (1) 324 in the acquired commuter ticket information table 320, the commuter ticket end. It is determined whether the station (1) 325, the commuter pass start station (2) 326, or the commuter pass end station (2) 327, and if they match, the flag 328 is set to 1. Considering round trips, the commuter pass start station (1) 324, the commuter pass end station (1) 325, the commuter pass start station (2) 326, and the commuter pass end station (2) 327 can all be entrance stations.

  Subsequently, in step 408, the boarding / alighting station prediction calculation unit 111 searches the usage record information table 330 for the usage record information of the IC ticket of the acquired record, and acquires all of them.

  In step 409, if the flag 328 is 1, the boarding / alighting station prediction calculation unit 111 determines that the train is used with a commuter pass and performs the process of step 411. If the flag 328 is 0, the use of a train that is not a commuter pass is used. The process of step 410 is performed after determination.

  The boarding / alighting station prediction calculation unit 111 accumulates information on the correspondence relationship between the entry station and the entry station when using the railway in the past for each user. In the correspondence relationship information, a participating station having a frequency that is special with respect to the entering station in the acquired record is determined, and the participating station is determined as a participating station (plan) 515 of the boarding station prediction information table 510, and the participating station is determined. A time obtained by adding the time required until the entry time 514 is determined as an entry time (scheduled) 516.

  In the case of use of a train with a commuter pass, the boarding / alighting station prediction calculation unit 111, in step 411, selects either the commuter pass start station (1) 324 or the commuter pass end station (2) 327 in the commuter pass information table 320. Determined as (planned) 515.

  If an unprocessed record remains in the ticket information table 310 after step 410 or step 411, the boarding / alighting station prediction calculation unit 111 returns to step 404 and searches for the next ticket information. If all the records have been processed, the boarding / alighting station prediction calculation unit 111, in step 412, the effective date 312, the entry station 313 and the entry time 314 of the commuter pass information, the entry station (plan) 515 and the entry time (plan) 516 and the estimated transfer station 517 are registered in the boarding station prediction information table 510 as boarding station prediction information 116. The boarding / alighting station prediction calculation unit 111 once terminates the process after step 412 and starts the process again from step 401 at regular time intervals.

  FIG. 6 is a diagram illustrating an example of the event information table. In the event information table 610, the number of users who enter the station at the end of an event where the user concentrates on the station, such as a concert, and the time of entry are registered. Information in the event information table 610 is set in advance from, for example, the event input terminal 104 before the event is held. In the event information table 610, an ID 611 given in the order of input from the event input terminal 104, an event enforcement date 612, a station name 613, an entrance time (plan) 614, and the number of visitors (plan) are recorded. The event information table 610 can be referred to from the station congestion degree prediction calculation unit 112.

  FIG. 7 is a diagram illustrating an example of a station congestion information table. The station congestion information table 710 stores the station congestion information 118 notified from the passenger count server 105. Passenger totaling server 105 analyzes in real time the video obtained from the measurement start time to the measurement end time from the camera installed in the station including the platform, and the number of users entering the station (number of visitors) Are counted. The sampling interval is preferably several tens of seconds to several minutes. In the station congestion information table 710, an ID 711, a station name 712, a measurement start time 713, a measurement end time 714, and the number of visitors 715 that are given in order are recorded. The station congestion information table 710 can be referred to from the station congestion degree prediction calculation unit 112.

  FIG. 8 is a diagram illustrating an example of a prediction diagram information table. In the prediction diagram information table 810, information of the prediction diagram 119 notified from the operation management system 106 is stored for each train. The operation management system 106 is a system that manages a plan diagram and a prediction diagram and manages a train operation based on them. In the prediction diagram 119, for each train, ID 811, header information 812, and diagram information 813 that are assigned in the stored order are stored. The header information 812 includes an enforcement date 814 on which the train is operated and a train number 815. The diamond information 813 includes station information 816 stored for each station. The station information 816 includes a planned arrival time 817 that is a time when the train arrives at the station in the operation plan, a planned departure time 819 that is a time when the train leaves the station, and a prediction simulation of the train operation by the operation management system 106. The predicted arrival time 818, which is the time when the train arrives at the station in the predicted diagram obtained by the above, and the predicted departure time 820, which is the time when the train leaves the station. The prediction diagram information table 810 can be referred to from the station congestion degree prediction calculation unit 112.

  FIG. 9 is a flowchart showing the processing of the station congestion degree prediction calculation unit 112. FIG. 10 is a diagram illustrating an example of a station congestion degree prediction information table. The station congestion degree prediction calculation unit 112 uses the boarding / alighting station prediction information 116, event information 117, station congestion information 118, and prediction diagram 119 to create station congestion degree prediction information 120. In the station congestion degree prediction information table 1010, the station congestion degree prediction information 120 generated by the station congestion degree prediction calculation unit 112 is recorded.

  In step 901, the station congestion degree prediction calculation unit 112 acquires the boarding / alighting station prediction information table 510. Next, the station congestion degree prediction calculation unit 112 acquires the event information table 610 in step 902. Furthermore, the station congestion degree prediction calculation unit 112 acquires the station congestion information table 710 in step 903. Furthermore, the station congestion degree prediction calculation unit 112 acquires the prediction diagram information table 810 in step 904.

  In the station congestion information table 710 illustrated in FIG. 7, the number of users in the premises of the station collected from the measurement start time to the measurement end time at each station is set as an entry, and the information is shown from the past to the present. ing. In step 905, the station congestion degree prediction calculation unit 112 calculates the average value of the increase / decrease rate of the number of visitors for the most recent sampling in the station congestion information table 710 for each station. In a simple example, the rate of increase / decrease in the number of visitors can be calculated by dividing the number of visitors in the latest (most recent) entry by the number of visitors in the latest previous entry.

  On average, users who enter the station from the automatic ticket gates enter the station by taking a train within a certain period of time, so if the number of visitors is accumulated for a certain period of time, the number of users at the station (number of people on the premises) ) Can be calculated.

  In step 906, the station congestion degree prediction calculation unit 112 calculates a value obtained by adding a value obtained by multiplying the on-site number calculated from the number of visitors 715 in the station congestion information table 710 by the increase / decrease rate calculated in step 905. The number of users after one sampling in the future is assumed to be the predicted number of people on the premises. After the next time, the same calculation is repeated for the predicted number of people on the premises, and the predicted number of people on the premises for a predetermined time in the future is calculated.

  In step 907, the station congestion degree prediction calculation unit 112 predicts the time calculated in step 906 by adding the number of people at the entry station (plan) 515 and the entry time (plan) 516 in the boarding / alighting station prediction information table 510. Add to the number of people on campus. This assumes that the user gets off from the train to the participating station (planned) 515 at the participating time (scheduled) 516, and adds the number of users to the future number of people in the participating station (planned) 515). Is.

  In step 908, the station congestion degree prediction calculation unit 112 converts the number of visitors (scheduled) 615 shown in the event information table 610 into the entry time (scheduled) 615 calculated in step 907 ( (Planned) It is added to the predicted number of people on the premises in 614.

  Further, in step 909, the station congestion degree prediction calculation unit 112 calculates the predicted arrival time 818 at which the train that stops at the entry station (plan) 515 in the boarding / alighting station prediction information table 510 arrives at the entry station 513. And subtracted from the predicted number of people on the premises at the time calculated in step 908.

  Further, in step 910, the station congestion degree prediction calculation unit 112 performs the process in step 911 when the predicted number of people on the station at the time calculated in step 909 is less than the predefined number of people on the station at each station. If the predicted number of people on the premises is equal to or greater than the number of people on the premises, the station congestion degree prediction calculation unit 112 performs the process of step 912.

  In step 911, the station congestion degree prediction calculation unit 112 sets the prediction time and the predicted number of people on the station calculated in step 909 as the predicted time 1013 and the predicted number of people on the station 1014 in the station congestion degree prediction information table 1010 illustrated in FIG. sign up. On the other hand, in step 912, the station congestion degree prediction calculation unit 112 sets the predicted time and the number of people on the premises calculated in step 909 as the prediction time 1013 and the predicted number of people on the premises 1014, and the station congestion degree prediction illustrated in FIG. Register in the information table 1010.

  After the processing from step 905 to step 912 is repeated until the predicted time 1013 reaches a predetermined future time, the processing is once ended, and the processing is started again from step 901 at a constant time period.

  By the above-described processing, the station congestion degree prediction information 120 calculated by the station congestion degree prediction calculation unit 112 is stored in the station congestion degree prediction information table 1010. Referring to FIG. 10, in the station congestion degree prediction information table 1010, ID 1011, station name 1012, prediction time 1013 calculated by the station congestion degree prediction calculation unit 112, and predicted number of people 1014 are registered. Managed. The station congestion degree prediction information table 1010 can be referred to from the train congestion degree prediction calculation unit 113. The station congestion degree prediction information 120 is distributed to end users including users in the station premises and users on the train via the information distribution system 108 connected to the outside via a network.

  FIG. 11 is a diagram illustrating an example of the in-vehicle number information table. The in-vehicle number information 121 notified from the vehicle information system 107 connected to the outside network is stored in the in-vehicle number information table 1110. The in-car number information table 1110 is a table showing how many users are on each train. The number of users on the train is calculated by the vehicle information system 107 based on the acquired load of each vehicle of the train. In the in-vehicle number information table 1110, ID 1111, train number 1112, measurement start time 1113, measurement end time 1114, travel section 1115, and in-vehicle number 1116 obtained from the vehicle weighting device etc. are registered, Managed. The in-vehicle number information table 1110 can be referred to from the train congestion degree prediction calculation unit 113.

  FIG. 12 is a flowchart illustrating processing of the train congestion degree prediction calculation unit. The train congestion degree prediction calculation unit 113 creates the train congestion degree prediction information 122 using the boarding / alighting station prediction information 116, the station congestion degree prediction information 120, and the in-car number information 121.

  The train congestion degree prediction calculation unit 113 acquires the boarding / alighting station prediction information table 510 in step 1201. In addition, the train congestion degree prediction calculation unit 113 acquires the station congestion degree prediction information table 1010 in step 1202. Furthermore, the train congestion degree prediction calculation unit 113 acquires the in-vehicle number information table 1110 in step 1203.

  Thereafter, in step 1204, the train congestion degree prediction calculation unit 113 performs prediction in the station congestion degree prediction information table 1010 with respect to the in-car number 1116 of a train described later among the trains indicated by the train number 1112 in the in-car number information table 1110. Add the number of people on campus 1014. The reason why the predicted number of people 1014 is added to the number of people 1116 in the vehicle is that the number of people in the vehicle increases when the user inside the station gets on the train.

  The target train to which the predicted number of people 1014 is added is determined from the measurement start time 1113, the measurement end time 1114, and the travel section 1115 in the in-vehicle number information table 1110. Specifically, it can be understood from which information the train was traveling at which time, and the time at which the train arrives at the station can be predicted. The predicted number of people 1014 at the predicted arrival station at the predicted time 1013 corresponding to the predicted arrival time may be added to the in-car number 1116 of the arriving train.

  The method for determining the target train to which the predicted number of people 1014 is added is an example, and other methods may be used. For example, the target train to which the predicted number of people 1014 is added may be determined based on the prediction diagram 119. Since it is known from the prediction diagram 119 which train arrives at which station at which time, the predicted number of people on the predicted time corresponding to the arrival time may be added to the in-vehicle number 1116. Here, the predicted number of people on campus 1014 is added to the number of people 116 in one train, but the present invention is not limited to this. As another example, it may be distributed to the number of people 1116 in a plurality of trains according to a predetermined condition. For example, at a station where a plurality of routes such as a terminal station enter, the allocation condition may be assigned to a plurality of routes at a preset ratio of the predicted number of people 1014.

  Next, in step 1205, the train congestion degree prediction calculation unit 113 determines whether the train arrives at a station near the time at which the train arrives and the time near the train from the number 1116 of the train indicated by the train number 1112 in the number of people information table 1110. The total number of users corresponding to the participating station (planned) 510 and the participating time (planned) 511 in the prediction information table 510 is subtracted.

  Further, in step 1206, the train congestion degree prediction calculation unit 113 determines whether or not the number of passengers calculated in step 1205 is less than a predefined boarding limit value. When the number of passengers in the vehicle is less than the boarding limit value, the train congestion degree prediction calculation unit 113 performs the process of step 1207. When the number of passengers in the vehicle is equal to or greater than the boarding limit value, the train congestion degree prediction calculation unit 113 performs the process of step 1208.

  In step 1207, the train congestion degree prediction calculation unit 113 obtains the train number indicated by the train number 1112 in the in-car number information table 1110 and the train indicated by the train number from the station name 1012 in the station congestion degree prediction information table 1010. The time shown in the predicted time 1013 of the travel section and station congestion degree prediction information table 1010 and the number of people in the vehicle calculated in the processing from step 1204 to step 1205 of the train that travels in the travel section at that time correspond to each other In addition, the train number 1312, the travel section 1314, the predicted time 1313, and the predicted in-car number 1315 are registered in the train congestion degree prediction information table 1310.

  On the other hand, in step 1208, the train congestion degree prediction calculation unit 113 associates with the train number, travel section, and time in the processing of step 1207, and sets the boarding limit value as the predicted passenger number 1315, and the train congestion degree prediction information table 1310. Register with.

  After step 1207 or step 1208, the train congestion degree prediction calculation unit 113 is the number of records indicated by the train number 1112 and the measurement time from the measurement start time 1113 to the measurement end time 1114 in the in-vehicle number information table 1110. After repeating step 1204 to step 1208, the process is temporarily ended, and the process is started again from step 1201 at a constant time period.

  In the train congestion degree prediction information table 1310, an ID 1311, a train number 1312, a prediction time 1313, a traveling section 1314, and a predicted in-vehicle number 1315 that are assigned in order are registered and managed as the train congestion degree prediction information 122. The train congestion degree prediction information 122 is distributed to the end user via the information distribution system 108 connected to the outside via a network. The train congestion degree prediction information 122 may also be notified to the operation management system 106 connected to the outside of the network and used as a parameter for a train operation prediction simulation executed by the operation management system 106.

  The above-described embodiments of the present invention are examples for explaining the present invention, and are not intended to limit the scope of the present invention only to those embodiments. Those skilled in the art can implement the present invention in various other modes without departing from the gist of the present invention.

DESCRIPTION OF SYMBOLS 100 ... Congestion degree prediction information delivery system, 101 ... Congestion degree prediction server, 1010 ... Station congestion degree prediction information table, 1012 ... Station name, 1013 ... Prediction time, 1014 ... Predicted number of people on premises, 102 ... Seat reservation system, 103 ... IC boarding Ticket system 104 ... Event input terminal 105 ... Passenger totaling server 106 ... Operation management system 107 ... Vehicle information system 108 ... Information distribution system 111 ... Boarding station prediction calculation unit 1110 ... In-car number information table 1111 ... ID 1112 ... Train number, 1113 ... Measurement start time, 1114 ... Measurement end time, 1115 ... Traveling section, 1116 ... Number of passengers in the car, 112 ... Station congestion degree prediction calculation part, 113 ... Train congestion degree prediction calculation part, 114 ... Seat Reservation information, 115 ... ticket information, 116 ... boarding / departure station prediction information, 117 ... event information 118 ... Station congestion information, 119 ... Prediction diagram, 120 ... Station congestion degree prediction information, 121 ... Number of passengers information, 122 ... Train congestion degree prediction information, 1310 ... Train congestion degree prediction information table, 1311 ... ID, 1312 ... Train number , 1313 ... predicted time, 1314 ... travel section, 1315 ... predicted number of passengers, 210 ... seat reservation information table, 211 ... ID, 212 ... enforcement date, 213 ... boarding station, 214 ... drop-off station, 215 ... train information, 216 ... The train boarding station, 217 ... the train getting-off station, 310 ... boarding ticket information table, 311 ... search ID, 312 ... effective date, 313 ... entrance station, 314 ... entrance time, 315 ... entrance flag, 320 ... commuter pass information table 321 ... Search ID, 322 ... commuter pass start date, 323 ... commuter pass end date, 328 ... flag, 330 ... use record information table, 31 ... Search ID, 332 ... Effective date, 333 ... Entrance station (result), 334 ... Entrance time (result), 335 ... Participating station (result), 336 ... Participation time (result), 337 ... Estimated transfer station, 510 ... Boarding / alighting station prediction information table, 511 ... search ID, 512 ... enforcement date, 513 ... entrance station, 514 ... entrance time, 515 ... participation station (plan), 516 ... participation time (plan), 517 ... estimated transfer station, 610 ... Event information table, 611 ... ID, 612 ... Effective date, 613 ... Station name, 710 ... Station congestion information table, 711 ... ID, 712 ... Station name, 713 ... Measurement start time, 714 ... Measurement end time, 715 ... Number of visitors, 810 ... Predictive diamond information table, 811 ... ID, 812 ... Header information, 813 ... Diamond information, 814 ... Effective date, 815 ... Train number, 816 ... Station information, 817 ... Planned arrival Time, 818 ... Predicted arrival time, 819 ... Planned departure time, 820 ... Predicted departure time

Claims (10)

The time at which the user enters and the time at which the station and the user are expected to enter based on the seat reservation information indicating the seat reservation made in advance and the ticket information read from the ticket by the ticket gate And a boarding / alighting station prediction calculation unit for generating boarding / alighting station prediction information indicating the station,
Based on the above-mentioned boarding / alighting station prediction information and a prediction diagram indicating the arrival time and departure time at each station of each future train, the station congestion degree prediction information predicting the number of users in each station at each future time A station congestion degree prediction calculation unit to be generated;
Based on the station congestion degree prediction information and the in-vehicle number information indicating the current boarding status of each train, train congestion degree prediction information is generated by predicting the number of passengers in the vehicle at each future time of each train. A congestion degree prediction apparatus having a train congestion degree prediction calculation unit. The station congestion degree prediction calculation unit
Calculate the rate of increase / decrease in the number of people in the station on the basis of the station congestion information indicating the actual number of people in each station from the past to the present, Calculate the predicted number of people on site predicting the number of users at
Based on the information of the user's entry station and entry time in the boarding / alighting station prediction information, the number of users who get off the train at the entry time to the entry station is added to the predicted number of people on the premises,
Obtaining information on the user's entry station, entry station, and entry time from the boarding / alighting station prediction information, and from the prediction diagram, a scheduled arrival time at which the train that arrives at the entry station at the entry time arrives at the entry station And subtracting the number of users on the train from the entrance station at the estimated arrival time from the predicted number of people on the premises,
The congestion degree prediction apparatus according to claim 1. The station congestion degree prediction calculation unit is further configured to enter the station at the event entrance time based on event information indicating the event entrance time and the number of event visitors that are the number of users who enter the station due to an event. Add the number of people entering the event to the predicted number of people on the premises,
The congestion degree prediction apparatus according to claim 2.   2. The congestion degree prediction apparatus according to claim 1, wherein the station congestion degree prediction calculation unit uses the local number limit value as the predicted local number if the calculated predicted local number is equal to or greater than the local number limit value. The train congestion degree prediction calculation unit,
From the number of people in the vehicle by the number of people in the vehicle, based on the predicted number of people in each station in the station congestion degree prediction information, add the number of people predicted to board the train at a future time at the station,
From the number of people in the car, based on the boarding / alighting station prediction information, subtract the number of people expected to get off from the train at a future time at each station,
The congestion degree prediction apparatus according to claim 1.   The congestion degree prediction apparatus according to claim 1, wherein the train congestion degree prediction calculation unit sets the in-vehicle number limit value as the in-vehicle number limit if the calculated in-vehicle number is equal to or greater than the in-vehicle number limit value. The congestion degree prediction device according to claim 1;
An information distribution device for distributing the station congestion degree prediction information and the train congestion degree prediction information to the terminal of the user;
Congestion degree prediction information distribution system. A congestion degree prediction method for predicting the congestion degree of a station and a train,
Based on the seat reservation information indicating the reservation of seats made in advance and the ticket information read from the ticket by the ticket gate, the boarding / alighting station prediction calculation means, the station when the user entered, the station and the user Generate boarding / prediction station prediction information that shows the time and station that is expected to appear,
The station congestion degree prediction calculation means calculates the number of users on the premises at each future time of each station based on the above-mentioned boarding / alighting station prediction information and a prediction diagram indicating arrival time and departure time at each station of each future train. Generate predicted station congestion prediction information,
The train congestion degree prediction calculation means predicts the number of passengers in the vehicle at each future time of each train based on the station congestion degree prediction information and the in-vehicle number information indicating the current boarding status of each train. Generate train congestion prediction information,
Congestion prediction method. The station congestion degree prediction calculation means,
Calculate the rate of increase / decrease in the number of people in the station on the basis of the station congestion information indicating the actual number of people in each station from the past to the present, Calculate the predicted number of people on site predicting the number of users at
Based on the information of the user's entry station and entry time in the boarding / alighting station prediction information, the number of users who get off the train at the entry time to the entry station is added to the predicted number of people on the premises,
Obtaining information on the user's entry station, entry station, and entry time from the boarding / alighting station prediction information, and from the prediction diagram, a scheduled arrival time at which the train that arrives at the entry station at the entry time arrives at the entry station And subtracting the number of users on the train from the entrance station at the estimated arrival time from the predicted number of people on the premises,
The congestion degree prediction method according to claim 8. The train congestion degree prediction calculation means,
From the number of people in the vehicle by the number of people in the vehicle, based on the predicted number of people in each station in the station congestion degree prediction information, add the number of people predicted to board the train at a future time at the station,
From the number of people in the car, based on the boarding / alighting station prediction information, subtract the number of people expected to get off from the train at a future time at each station,
The congestion degree prediction method according to claim 8.

Images