JP6875311B2 - Transportation congestion prediction system and congestion prediction method - Google Patents

Description

The present invention relates to a system and a method for predicting and visualizing congestion of vehicles and boarding / alighting facilities for transportation.

It is very important for railway operators to understand the congestion status of stations and trains in order to realize safe and stable transportation services. In particular, the commander who manages the operation of the railway needs to immediately judge the necessity of increasing the number of trains and extending the stop time in the event of an abnormality, and wants to know the congestion status of stations and trains in real time as a basis for that judgment. There is a need.

Surveillance cameras have already been installed at many stations, allowing railway operators to grasp platform congestion in real time even from remote locations. Congestion grasping using surveillance cameras can be said to be an effective means for specific areas such as near ticket gates and escalator boarding / alighting locations, but in order to grasp the degree of congestion of the entire station, surveillance cameras should be used inside the station. It is necessary to place it in all areas without blind spots, and there are difficult issues in terms of cost and consideration for privacy for passengers.

On the other hand, there is known a method of obtaining train congestion using data obtained from a load sensor of a vehicle. For example, Patent Document 1 discloses a technique for calculating a train occupancy rate by using a vehicle weight and an average weight per passenger.

In addition, a technology that uses the passage record of an automatic ticket gate installed at a station is also known. In Patent Document 2, in order to predict future passenger flow, the passage records of automatic ticket gates are aggregated, and by comparing with the measured values on the day of prediction, statistically similar patterns are extracted on the day. The technology used for forecasting the flow of

Further, Patent Document 3 discloses a technique of calculating a passenger's route from a departure station to an arrival station by using a train by a utility function and predicting the number of passengers on the train.

Japanese Unexamined Patent Publication No. 2012-101640 Japanese Unexamined Patent Publication No. 2010-061321 Japanese Unexamined Patent Publication No. 2012-196987

The method of Patent Document 1 is a technique for estimating a train occupancy rate, and does not correspond to grasping a congestion situation at a station. In addition, in order to collect the data measured by the load sensor of the vehicle in real time, it is necessary to introduce a wireless system, and it takes a large development cost to build a mechanism to comprehensively collect it for all trains. ..

Further, using the method of Patent Document 2, it is possible to predict OD data (combination of entrance station and exit station) for each time zone, but this method alone directly estimates the congestion status of trains and stations. You can't.

Further, by using the method of Patent Document 3, it is possible to predict the degree of congestion of trains and the degree of congestion of stations by allocating the used routes and used trains to the OD data for each time zone based on the utility function. However, this method has a problem that as the number of target OD data and the scale of the route network increase, it takes time to calculate the allocation to trains, and it becomes difficult to present the prediction result immediately.

The present invention has been made in view of this point, and an object of the present invention is to predict and visualize the congestion status of trains and stations at high speed for a large-scale route network.

A preferred aspect of the present invention comprises an information processing device that can use station / route / train information and passenger demand information. The station / route / train information stores the railway network route information and timetable information. Passenger demand information includes a plurality of trip data, and one of the trip data includes the number of passengers moving, the departure station which is the station where the passenger departs, and the arrival station which is the target station of the passenger. Includes the departure time with respect to the time of departure from the departure station. Then, a boarding train estimation means that generates user behavior prediction information by inputting station / route / train information and passenger demand information, estimating and adding the train on which the passenger is boarding to each of the trip data. Be prepared. Further, the passenger demand information aggregating means for aggregating the passenger demand information processed by the boarding train estimation means by combining a plurality of trip data is provided. Then, the calculation time for estimation is shortened by processing the passenger demand information aggregated by the passenger demand information aggregation means by the boarding train estimation means.

Another preferred aspect of the present invention is a congestion prediction method using an information processing device that can use station / route / train information and passenger demand information. The station / route / train information stores the railway network route information and timetable information. Passenger demand information includes a plurality of trip data, and one of the trip data includes the number of passengers moving, the departure station which is the station where the passenger departs, and the arrival station which is the target station of the passenger. Includes the departure time with respect to the time of departure from the departure station. Then, for each of the input step for inputting the station / route / train information and the passenger demand information and the trip data, the train on which the passenger is boarding is estimated and added, and the boarding train estimation for generating the user behavior prediction information is generated. The passenger demand information aggregation step, which aggregates the passenger demand information processed in the step and the passenger train estimation step by combining a plurality of trip data, is provided, and the passenger demand information aggregated in the passenger demand information aggregation step is boarded. By processing in the train estimation step, the calculation time for estimation is shortened.

It is possible to predict and visualize the congestion status of trains and stations at high speed for large-scale railway networks. Issues, configurations and effects other than those described above will be clarified by the description of the following examples.

It is a block diagram which shows the system configuration of this Example. It is a table diagram which shows the data structure of a station / line / train information 123. It is a table diagram which shows the data structure of a station / line / train information 123 (continued). It is a table diagram which shows the data structure of the passenger demand information 124. It is a conceptual diagram which shows the example of the route map of the railway to which this embodiment is applied. It is a table diagram which shows the data structure of the train congestion prediction information 125. It is a table diagram which shows the data structure of the station congestion prediction information 126. It is a table which shows the data structure of the user behavior prediction information 127. It is a flowchart of the boarding train estimation program 134. It is a flowchart of a passenger demand information aggregation program 135. It is a graph which shows the relationship between the passenger demand information aggregation level and the aggregation ratio. It is a top view which shows an example of a passenger demand information aggregation level setting screen. It is a flowchart of a train crew totaling program 136. It is a flowchart of the station staying person totaling program 137. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a flowchart of the program which determines the display position of a station and a line on the screen distributed by an information distribution server 113. It is a flowchart of the program which determines the display size of a station on the screen distributed by an information distribution server 113. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes. It is a top view which shows an example of the screen which the information distribution server 113 distributes.

The embodiment will be described in detail with reference to the drawings. However, the present invention is not construed as being limited to the description of the embodiments shown below. It is easily understood by those skilled in the art that a specific configuration thereof can be changed without departing from the idea or gist of the present invention.

In the configuration of the invention described below, the same reference numerals may be used in common among different drawings for the same parts or parts having similar functions, and duplicate description may be omitted.

When there are a plurality of elements having the same or similar functions, they may be described by adding different subscripts to the same code. However, if it is not necessary to distinguish between a plurality of elements, the subscript may be omitted for explanation.

The notations such as "first", "second", and "third" in the present specification and the like are attached to identify the components, and do not necessarily limit the number, order, or contents thereof. is not it. In addition, numbers for identifying components are used for each context, and numbers used in one context do not always indicate the same composition in other contexts. Further, it does not prevent the component identified by a certain number from having the function of the component identified by another number.

The position, size, shape, range, etc. of each configuration shown in the drawings and the like may not represent the actual position, size, shape, range, etc. in order to facilitate understanding of the invention. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings and the like.

The publications, patents and patent applications cited herein form part of the description herein.

Components represented in the singular form herein shall include the plural form unless explicitly indicated in the context.

An example of an embodiment described below describes a train / station congestion prediction system that includes a processor that executes a program and a storage device that stores the program. This system is organically coupled with boarding / alighting facilities for using transportation means and measuring means for detecting the number of passengers installed in a part of the vehicle. The storage device stores the number of passengers partially measured by the measuring means, and the processor aggregates passenger data according to the specified estimated completion time of calculation, and all vehicles and all boarding / alighting facilities of the target transportation network. Calculate the number of people staying at high speed.

Embodiments of the present invention will be described with reference to FIGS. 1 to 27. Although the examples of the present invention are intended for railway transportation services, the present invention can be applied to transportation means (for example, buses) used by a large number of people.

FIG. 1 is a diagram showing an overall outline of a process for predicting congestion of transportation means at high speed and continuously visualizing the congestion in the system according to the embodiment of the present invention.

In this example, a sensor attached to a train and partial actual data that can be obtained from existing equipment in the station yard are used to simulate the movement of each passenger from a few minutes to a few hours ahead, and the railway to be calculated. Fastly predict congestion at all trains and stations on the network. When the scale of the railway network to be calculated is large and it takes time to calculate the movement of all passengers, the user of the system aggregates passenger data step by step while checking the estimated time to complete the calculation in advance. However, the time required to complete the calculation can be shortened. After the calculation for the movement of all passengers is completed, the congestion prediction results of trains and stations are visualized simultaneously or continuously on one screen. The commander and the station staff can utilize the presented congestion prediction result as a judgment material in carrying out their respective duties.

FIG. 1 is a diagram showing a system configuration of this embodiment. In recent years, automatic ticket gates 103 have been installed in many railway stations, and automatic ticket gates 103 read contactless IC cards (or mobile terminals with equivalent functions) 102 or magnetic tickets. User 101 can enter or exit the station. The information read by the automatic ticket gate 103 is transmitted to the data management server group 105 managed by the railway operator via the network 109, and is accumulated as ticket gate passage data. The ticket gate passage data includes information such as the station and time when each passenger entered to use the train, and the station and time when the passenger got off.

In addition, surveillance cameras 106 are often installed near ticket gates or on platforms. Since the spatial range that can be taken by one surveillance camera is limited, it is common to install multiple surveillance cameras near the ticket gate, each platform, aisle, stairs, etc. at one station. The video data acquired by the surveillance camera 106 can be confirmed in real time via the network 109 at a command center or the like that manages the operation of the railway. A technique has also been developed in which a person is detected from an image taken by a surveillance camera 106 and the number of people existing in a certain space is totaled.

Further, with the spread of public wireless LANs that can be used in stations and trains, access points 104 are being installed in various places. Here, the public wireless LAN may be provided by a business operator other than the railway business operator. The public wireless LAN is a service that provides a connection to the Internet by a wireless LAN, and the user 101 connects to the Internet from a mobile terminal such as a notebook PC, a tablet PC, or a smartphone via an access point 104. Since the range in which radio waves can reach from one access point is generally about several tens of emails, multiple access points are installed in a wide space such as a station. In order to prevent interference when the mobile terminal can communicate with a plurality of access points 104, communication is performed by an SSID that identifies the network. Therefore, the access point 104 side can acquire the connection start time and the connection end time of each mobile terminal.

Generally, a plurality of access points 104 installed in a station yard are installed near a ticket gate or on a platform, so the position of the mobile terminal can be roughly determined from the signal strength between each access point 104 and the mobile terminal of the user 101. Can be estimated. Alternatively, by tracking the access point 104 to which each mobile terminal is connected in chronological order, the movement of passengers in the station yard can be estimated. In addition, when the user 101 is using a public wireless LAN service that can be used by registering a MAC address or the like which is an identifier of the mobile terminal in advance, the movement history of the mobile terminal is connected to the wireless LAN. Can be obtained from the data. The connection information of the access points 104 installed in these station premises is transmitted to the data management server group 105 in real time.

In this embodiment, the number of station users 101, the routes used by each user 101, and the like are estimated based on the information from the automatic ticket gate 103, the surveillance camera 106, and the access point 104. Of course, such estimates are not limited to those using these devices.

Train 107 is operated by a railway management system owned by a railway operator. The railway management system has a timetable planning system, an operation management system, a train information management system, etc. as subsystems, and the operation management system manages whether the train 107 is operated according to the timetable created by the timetable planning system. To do. The train information management system is a system that acquires and collects various information from the train 107 in operation and transmits the information to the crew and the operation management center. By utilizing the train information management system, train positions, train numbers, train failure information, etc. can be transmitted from each train and aggregated in the operation management center. Generally, the train 107 is equipped with a sensor that can measure the weight of the train in order to adjust the braking strength. It is possible to collect the measured values of the train weight and estimate the passenger information of the train 107 from the average weight per passenger (see, for example, Patent Document 1).

The transportation congestion prediction system 108 may be owned by the railway operator as a part of the business system, or may be owned by a service operator different from the railway operator and the congestion prediction result may be distributed to the railway operator. It may be a business form to be carried out. The transportation congestion prediction system 108 has a data server 111, a calculation server 112, and an information distribution server 113. Each server is connected to a computer used by the system operator (or the railway operator if the railway operator owns the transportation congestion prediction system 108) 114 via the network 118. In addition, the transportation congestion prediction system 108 can communicate with the operation management center 163 in which the commander 162 is performing business, and the terminal 161 used by the passengers and the like 160 via the network 117.

Hereinafter, the servers constituting the transportation congestion prediction system 108 will be described. Although described as three server groups in this embodiment, they can be configured as one computer physically or as a computer system composed of a plurality of computers logically or physically configured, and the same computer. It may run on a separate thread above, or it may run on a virtual computer built on multiple physical computer resources.

The basic configuration of each server is the same, and it is a computer having network interfaces (I / F) 130 and 150, processors (CPU) 131 and 151, memories 132 and 152, and storage units 133 and 153. The network interfaces 130 and 150 are interfaces for connecting to the networks 109, 117 and 118. Processors 131 and 151 execute programs stored in memories 132 and 152. The memories 132 and 152 include a ROM (Read Only Memory) which is a non-volatile storage element and a RAM (Random Access Memory) which is a volatile storage element. An immutable program, such as a BIOS (Basic Input Output System), is stored in the ROM. Further, the RAM is a high-speed and volatile storage element such as a DRAM (Dynamic Random Access Memory), and temporarily stores a program executed by the processors 131 and 151 and data used when the program is executed. The storage units 133 and 153 are composed of a large-capacity and non-volatile storage device such as a magnetic storage device (HDD (Hard Disc Drive)), a flash memory (SSD (Solid State Drive)), and an optical drive, and the processor 131. , 151 stores the program to be executed and the data used when the program is executed. A plurality of recording devices may be provided as the storage unit, and the program or data may be divided into a plurality of recording devices for recording.

In addition, each server is connected to an input interface to which a keyboard and mouse are connected to receive input from the operator, and an output interface to which a display device and a printer are connected to output the program execution result in a format that the operator can see. You may have. In addition, the program executed by each server is provided to each server via a network or removable media (optical disk, flash memory, etc.). Therefore, each server may have an interface for reading data from removable media.

First, the data server 111 will be described. Information from the equipment (103, 106) in the station yard, access point 104, information from the railway management system that manages the train 107, etc. can be obtained at the timing when new data is acquired or at a predetermined time interval (every few minutes). , Every few hours, etc.), it is transmitted to the data server 111 via the network 109. For data collection, various known techniques as described in Patent Documents 1 to 3 may be used.

The data server 111 records the received data in the data storage unit (DB) 121. The data storage unit 121 stores station / route / train information 123, passenger demand information 124, train congestion prediction information 125, station congestion prediction information 126, and user behavior prediction information 127. Details of these data will be described later.

Next, the calculation server 112 executes the congestion prediction process of the train or the station by using the data group stored in the data server 111. The storage unit 133 stores the boarding train estimation program 134, the passenger demand information aggregation program 135, the train occupant aggregation program 136, the station resident number aggregation program 137, and intermediate data generated in the process of calculation processing. The program is read from the storage unit 133, loaded into the memory 132, and executed by the processor 131. The data to be analyzed is acquired from the intermediate data stored in the data server 111 or the storage unit 133, temporarily stored in the memory 132, and the processor 131 reads the program from the storage unit 133 and executes it. These programs may be automatically executed according to a predetermined time interval (for example, every few seconds, every few minutes, etc.), or may be executed at a timing instructed by the system operator 114. Details of these programs will be described later.

Next, the information distribution server 113 stores programs such as the information distribution program 155 and intermediate data generated in the process of calculation processing in the storage unit 153. The information distribution server 113 is accessed from the terminal 161 used by the system operator 114, the commander 162 of the operation management center 163, or the passenger 160 via the networks 117 and 118 to provide information.

The system operator 114 of the transportation congestion prediction system 108 uses the terminal to the network 118 via the network 118 to configure and status the data stored in the transportation congestion prediction system 108, the status and calculation results of the calculation server 112, and the user. You can check the status of search requests.

2 and 3 are diagrams showing the data structure of the station / line / train information 123 stored in the data server 111. The station / route / train information 123 includes the route information and the timetable (operation plan) information of the railway to which the system is applied. The station / route / train information 123 indicates the physical configuration of the transportation means and the like, and is master data stored in advance by the system operator 114 and the like.

In FIG. 2, the station information 300 stores information on each node (node) constituting the transportation network. Nodes include railway stations, bus stops, aircraft airports, and so on. In the present specification and the like, these may be collectively referred to as a "station". In this embodiment, the station information 300 includes information such as station ID 301, station name 302, owning company 303, location 304, latitude / longitude information 305, and path information 306 in which the station yard map is stored.

The route information 310 stores information on each link (connection between nodes) constituting the transportation network. Links include railroad routes, bus routes, aircraft routes, and the like. In the present specification and the like, these may be collectively referred to as "route". In this embodiment, the line information 310 includes information such as a line ID 311, a line name 312, an operating company 313, a train type 314, and a line shape 315, and represents a line structure. The route shape 315 is information used when visualizing the congestion prediction result on the screen, and may be arbitrarily specified by the system administrator or the user, or the latitude / longitude information 305 of each station included in the station information 300 may be specified. You may add a process that is automatically determined by using.

The station / line-related information 320 stores information related to the node defined in the station information 300 and the link defined in the line information 310. In this embodiment, the station / line-related information 320 includes information such as line ID 321 and station ID 322, and travel order 323. The stations that make up the line are basically stored according to the actual order. The order can be known from the traveling order 323.

The train timetable information 330 stores information on transportation means that move between nodes via a link. Transportation means include trains, buses, aircraft, etc. In the present specification and the like, these may be collectively referred to as "train". Stores information about when each node on each link departs, arrives, or passes. In this embodiment, the train timetable information 330 includes information such as train ID 331, line ID 332, station ID 333, traveling order 334, stop classification 335, arrival time 336, departure time 337, and capacity information 338. A value indicating passing or stopping is stored in the traffic stop category 335. When the traffic stop classification is passing, the arrival time and departure time values do not necessarily have to be stored. The train timetable information 330 receives data from the timetable planning system of the railway management system at any time and is updated.

In FIG. 3, the station yard equipment information 460 divides the physical configuration of each node defined in the station information 300 into a plurality of areas and stores information about each area. The station yard equipment information 460 includes information such as area ID 461, station ID 462, line ID 463, area name 464, area area information 465, and staying number threshold value 466. The definition regarding the area delimiter may be uniquely specified at the time of system introduction, or may be arbitrarily specified for each system administrator or user. The resident number threshold value 466 may be automatically obtained and stored by using the area area information 465, or may be arbitrarily specified by the system administrator or the user. The information of the resident number threshold value 466 is information necessary for alert expression when visualizing the congestion prediction result of the station.

FIG. 4 is a diagram showing a data structure of passenger demand information 124 stored in the data server 111. Passenger demand information 124 includes information such as trip ID 401, departure station ID 402, arrival station ID 403, departure time 404, number of passengers 405, and route information 406, and represents information on where and when passengers start moving. .. The passenger demand information 124 can be created based on various data collected via the network 109 and data separately input by the system operator 114. The passenger demand information 124 is used as the original data of the simulation executed by the transportation congestion prediction system of this embodiment.

The trip ID is a unique value. The time granularity of the departure time 404 may be 1 second unit, or may be stored in any unit such as 1 minute unit or 10 minute unit. Passenger demand information 124 can be calculated by aggregating the number of passengers who have entered a station in a predetermined time zone and got off at which station, for example, using past ticket gate passage data. The first record of FIG. 4 shows an example in which one passenger enters the departure station (10001) at the departure time (2016/3/21 12:00:10) and heads for the arrival station (10002). ..

As a means of creating the passenger demand information 124, there is also a method of utilizing the survey results such as a questionnaire. In addition, current or future demand may be estimated based on past data. For example, a method of utilizing the history of IC ticket data of passengers who have passed through an automatic ticket gate installed at a station, analysis of connection information of public wireless LAN access points and surveillance camera images, and the departure station of the user. And how to guess the arrival station. In addition, it may be created in consideration of irregular events such as weather, surrounding traffic conditions, and events. For example, a plurality of passenger demand information may be held according to the purpose of business such as weekdays and holidays. The passenger demand information 124 may be updated automatically according to a predetermined time interval (for example, every few seconds, every few minutes, every few hours, every few days, etc.), or the data server 111 may newly update the ticket gate. It may be executed at the timing of receiving the passing data.

The route information 406 is information including the line to be used, the boarding station, and the getting-off station, and the above information is stored for all the lines used for the movement using a plurality of lines.

With reference to FIG. 5, the route information 406 will be described using an example of a route network. For example, the record with trip ID 401 of FIG. 4 is a trip departing from the station (10001) and arriving at the station (10006), but "20002-10001-10003-20004-10003-10006" is used as the route information. Given. In this case, the passenger first gets on the line of the line ID (20002) from the station (10001) and gets off at the station (1003). At the station (1003), change to the line (20004), get on at the station (1003), and get off at the station (10006). A record with trip ID 401 also departs from the station (10001) and arrives at the station (10006), but "20003-10001-10006" is given as route information. In this case, it means that the passenger uses only the route with the route ID (20003).

In this way, if the actual route for each passenger is unknown from the acquired data, the route information 406 may set only one route for the combination of the departure station ID and the arrival station ID. However, a plurality of routes may be distributed and stored in some way. Regarding route allocation, it is useful to use the actual value of tracing the movement route of each passenger using the connection information with the wireless access point installed in the station yard, or to list multiple route candidates by route search. There is a method of allocating using a function. In addition, route candidates may be listed for each time zone using the train timetable, and the allocation rate may be changed and specified.

FIG. 6 is a diagram showing a data structure of train congestion prediction information 125 stored in the data server 111. The train congestion prediction information 125 stores congestion prediction results for all trains of the transportation means to be calculated in a format such as train passenger number prediction information 420 and boarding / alighting number prediction information 430. The train congestion prediction information 125 is created by performing a simulation based on the station / route / train information 123 and the passenger demand information 124.

The train passenger number prediction information 420 represents the number of passengers in the running train. The train occupancy prediction information 420 includes information such as train ID 421, line ID 422, station ID 423, travel order 424, and occupant 425, and represents occupancy information for each train traveling between stations. Here, instead of the passenger number 425, the passenger rate obtained from the passenger number and train capacity information may be stored. For example, the data in the first line of the train occupancy prediction information 420 indicates that the number of personnel in the train of train ID 8763 is 401 at the station of station ID 10001, which means that the train occupies the station of station ID 10002.

The boarding / alighting number prediction information 430 represents the boarding station and the getting-off station of passengers for each train. The boarding / alighting number prediction information 430 includes information such as train ID 431, line ID 432, boarding station ID 433, getting-off station ID 434, and number of passengers 435. Is stored. For example, the data in the first line of the boarding / alighting number prediction information 430 indicates that for the train with train ID 8765, 10 passengers get on at the station with station ID 10001 and the 10 passengers get off at the station with station ID 10002.

As described above, the train passenger number prediction information 420 can be estimated using the value of the weight sensor attached to the train. Alternatively, a measuring means such as a camera or an infrared sensor may be attached near the door of the train to count the number of people getting on and off the train. Alternatively, if a public wireless LAN access point is provided in the train, the number of passengers in the train may be calculated in the same way as the method for estimating the number of passengers waiting for the train on the platform of the station. It may be used to estimate and aggregate the routes used by all passengers and the trains used in the same manner as the above-mentioned estimation of the number of passengers waiting for trains.

FIG. 7 is a diagram showing a data structure of station congestion prediction information 126 stored in the data server 111. In the station congestion prediction information 126, the congestion prediction results are stored in the format of the number of people prediction information 440 for each moving direction and the number of people staying for each destination 450 for all stations included in the transportation network to be calculated. The station congestion forecast information 126 is created by performing a simulation based on the station / route / train information 123 and the passenger demand information 124.

The number prediction information 440 for each movement direction represents the movement of passengers in the station yard. The number of people prediction information 440 for each moving direction includes information such as station ID 441, time 442, start point area ID 443, end point area ID 444, and number of people 445. The time unit of aggregation follows a predetermined time unit such as every 1 minute, every 5 minutes, every 10 minutes. The number of people prediction information 440 for each moving direction is data that records the number of visitors, participants, and transferees at each station in units of designated times.

For example, the following three types of station yard movement patterns can be expressed by the values of the start point area ID and the end point area ID.
(1) Passengers entering from the ticket gate and boarding the train Start point area ID = ticket gate area of a certain station, end point area ID = home of a certain station Time 442 stores, for example, the time when the movement started.
(2) Passengers getting off the train and entering the ticket gate Start point area ID = platform of a certain station, end point area ID = ticket gate area of a certain station Time 442 stores, for example, the time when the train got off.
(3) Passengers who get off the train on one line and transfer to another line Start area ID = platform of a certain station, end point area ID = home of a certain station End point area ID ≠ start point area ID
At time 442, for example, the time when the train of the transfer source is disembarked is stored.

The destination-specific staying number prediction information 450 includes information such as station ID 451 and time 453, arrival station ID 453, and number of people 454. The staying number prediction information 450 for each destination is data obtained by totaling the staying number of people at each station as cross-sectional information for each destination. Similar to the number of people prediction information 440 for each moving direction, the time unit of aggregation follows a predetermined time unit such as every 1 minute, every 5 minutes, every 10 minutes.

In the first and second lines of the data structure of the information 450 for predicting the number of passengers staying by destination, the total number of passengers staying at the station ID (10001) at the time (2016/03/21 12:00:00) is 50. An example is shown in which there are 10 passengers whose arrival station (destination) is (10010) and 40 passengers whose arrival station (destination) is (10012). The resident passengers include passengers whose departure station or arrival station is that station, as well as transfer passengers.

One of the measuring means of the station congestion prediction information 126 is the utilization of the connection information of the access point of the public wireless LAN. If an access point is installed for each platform, find the number of connections. If there are multiple access points on one platform, the total number of connections can be calculated, and if two tracks are placed facing each other on one platform, the number of connections can be decomposed for each track. Just do it. For disassembly, the communication strength from a plurality of access points may be used to estimate the position of each terminal on the platform, or the image of the surveillance camera 106 shooting on the platform may be used for the track number. A method such as detecting the approximate number of people staying in each case can be considered. In addition, since all passengers using the station are not public wireless LAN users, the ratio of "total number of passengers / number of public wireless LAN users" is obtained in advance and corrected. To set the ratio, a questionnaire survey may be conducted, or the number of passengers who passed through the automatic ticket gate 103 within a predetermined time obtained from the automatic ticket gate 103 and the access point to which each mobile terminal is connected are determined. It may be determined by comparing with the number of passengers who have passed through the automatic ticket gate 103, which can be estimated from the movement of passengers obtained by tracking in time series.

Further, the number of passengers staying may be obtained by detecting a person by image processing using the image of the surveillance camera 106 installed in the station yard and totaling the number of passengers waiting for the train for each line. In addition to the above, a measuring means such as an infrared sensor may be used.

FIG. 8 is a diagram showing a data structure of user behavior prediction information 127 stored in the data server 111. User behavior prediction information 127 includes trip ID 471, departure station ID 472, arrival station ID 473, number of people 474, departure time 475, arrival time 476, route ID (1) 477, train ID (1) 478, boarding station ID (1) 479. , Disembarkation station ID (1) 480, train waiting time (1) 481, train boarding time (1) 482, and the like. The user behavior prediction information 127 is created by performing a simulation based on the station / route / train information 123 and the passenger demand information 124.

The train waiting time is the time from the departure time to the time of getting on the first train, or the time from getting off the train of one line to getting on the train of the next line. The train boarding time is the time from the time when the train was boarded at a certain station to the time when the train was disembarked at a certain station.

Here, the trip ID 471, the departure station ID 472, the arrival station ID 473, the number of people 474, and the departure time 475 correspond to the records stored in the passenger demand information 124. In the items after the arrival time 476, the result calculated by the boarding train estimation program 134 described later is stored.

The six items are line ID (1) 477, train ID (1) 478, boarding station ID (1) 479, getting off station ID (1) 480, train waiting time (1) 481, and train boarding time (1) 482. , It corresponds to the route information 406 of the passenger demand information 124, and is repeated as many times as the number of routes used by the passengers. In the case of passengers using a route including transfer, the route ID (2), train ID (2), and so on are recorded in succession.

FIG. 9 is a flowchart of the boarding train estimation program 134. The boarding train estimation program 134 reads the passenger demand information 124 stored in the data server 111, refers to the route information of each trip data, and links the available trains to obtain the train waiting time and the final arrival time. Is calculated, and the user behavior prediction information 127 is output. The boarding train estimation program 134 shall be automatically executed at regular intervals such as 5 minutes, 10 minutes, and 30 minutes, or executed at the timing instructed by the system operator.

First, the station / route / train information 123 to be calculated is read (step 901). At this time, a variable for storing the number of passengers between all trains and all stations is prepared and initialized.

Next, the passenger demand information 124 is read and sorted in order of the combination of the departure station ID and the arrival station ID and the departure time (step 902). Sorting is not essential, but it is desirable to include it as a calculation device because it can be expected to speed up the process of searching for available trains.

If the amount of passenger demand information 124 to be processed here is large, there is a problem that the calculation time is long, so the passenger demand information is aggregated as necessary (step 903). By performing aggregation, the number of repetitions after the subsequent processing 904 can be reduced. However, this aggregation process has a trade-off relationship with the accuracy of boarding train estimation. The aggregation process of this passenger demand information will be described later.

Next, the following processing is repeated with reference to the station / route / train information 123 for each record of the passenger demand information 124 to be calculated (step 904).

First, a variable t related to time is prepared and initialized (step 905). The departure time is stored as an initial value in the time-related variable t.

Further, the following processing is repeated for the number of routes to be used (step 906).

First, the route ID to be used first is obtained from the route information, and among the trains running on the relevant route, the one that arrives at the boarding station earliest after the time t is searched for (step 907).

Next, it is determined whether or not the searched train can be boarded (step 908). Specifically, it refers to the number of passengers on the searched train, determines that it is possible to board the train if it does not exceed the capacity even if the number of people on the trip is added, and if it exceeds the capacity, it is not possible to board. If it is determined that the train cannot be boarded, the same determination process is performed with the next train arriving at the relevant station as a candidate. At this time, as the capacity information, the information included in the train timetable information 330 of the station / route / train information 123 may be used as it is, or a margin of up to 100 people or up to 1.5 times is allowed. May be provided. Further, the degree of margin may be adjusted for each route or time zone. If a train that can be boarded can be searched for, the boarding section of the trip (between the boarding station and the getting-off station) is referred to, and the boarding personnel information of the corresponding train is updated.

The information of the searched train ID is retained. Further, the train waiting time is obtained from the difference between the time t and the departure time of the corresponding train ID, and is retained (step 909). Actually, it takes time to walk to the platform where the train of the first line to use arrives after entering the ticket gate of the departure station, so even if you take this walking time into consideration and deduct a certain amount of time Good.

Next, the time when the corresponding train ID arrives at the getting-off station is obtained and stored at the time t (step 910). In addition, the time from the time when the corresponding train ID departs from the boarding station to the time when it arrives at the getting-off station is retained as the boarding time. The above process is repeated for the number of route IDs used.

If the search for trains that can be boarded on all routes is successful and it is confirmed that the train arrives at the arrival station described in the passenger demand information, the retained information is stored in the user behavior prediction information 127. (Step 911).

FIG. 10 is a flowchart of the passenger demand information aggregation program 135. The passenger demand information aggregation program 135 reads the passenger demand information 124 stored in the data server 111, aggregates according to the aggregation level, and compresses the number of records to be processed in order to shorten the calculation time of the passenger train estimation program 134. To do.

Here, the concept of aggregation level will be described with reference to FIGS. 11 and 12. The aggregation level is an index showing the degree of decrease when compared with the original number of passenger demand information 124. As described above, the boarding train estimation program 134 is a process of searching for and allocating trains that can be boarded for each trip, so the calculation time is proportional to the number of trips. Therefore, in order to reduce the calculation time, it is effective to consolidate by some rule and reduce the number of trips. Specifically, for example, there is a method of totaling the departure time in a set time unit such as 1 minute or 2 minutes instead of the second unit.

As shown in FIG. 11, the larger the time unit of aggregation, the higher the aggregation ratio. However, the larger the time unit of aggregation, the higher the probability that a bias will occur in the train search process of the boarding train estimation program, and therefore the accuracy of the congestion prediction result will be lower. This passenger demand information aggregation process is an effective function when it is desired to know the congestion prediction result as soon as possible because it may be rough.

FIG. 12 is an example of the aggregation level setting screen. As a method of utilizing the aggregation level, the user can set the aggregation level while confirming the estimated time until the processing of the boarding train estimation program is completed by using the setting screen as shown in FIG. Although FIG. 12 shows the aggregation level as a discrete numerical value, it may change continuously. When the aggregation level is represented by a discrete numerical value, it can be defined by extracting representative points in the graph of the aggregation unit of the departure time and the aggregation ratio shown in FIG. 11, for example. As for the definition of the aggregation level, various rules can be considered as shown below in addition to the aggregation unit of the departure time, and a plurality of rules may be used in any combination. It is desirable that the rule can be set by the option of the setting screen shown in FIG.
・ Select and aggregate a specific departure station-arrival station combination ・ Aggregate only in a specific departure time zone ・ Change the aggregation unit for each departure time zone (for example, in the case of commuting rush hour) Set the aggregation unit small, and set the aggregation unit large during the daytime hours, etc.)
-Refer to the route used on route 406 of passenger demand information 124, and change the aggregation unit based on the route used (refer to the timetable, change the aggregation unit according to the operation interval, the number of trains per hour For routes with few, set a large aggregation unit, etc.)
-Set an upper limit so that the number of passenger demand information after aggregation does not exceed a certain threshold.

The passenger demand information aggregation program 135 is automatically executed at the timing when the passenger demand information 124 is updated, is automatically executed at the execution timing of the boarding train estimation program 134, or is automatically executed from the system operator. It shall be executed at the specified timing.

The processing of the passenger demand information aggregation program 135 will be described with reference to FIG. As an example, the process of aggregating in the aggregation unit of the departure time will be described.

First, the station / route / train information 123 to be calculated is read (step 1001). Next, the passenger demand information 124 is read (step 1002).

According to the aggregation level specified on the setting screen, the aggregation unit of the departure time is acquired (step 1003).

In addition, the variables for storing the passenger demand information after aggregation are initialized (step 1004).

The following processing is repeated for the trip ID for the passenger demand information read in step 10002 (step 1005).
The trip ID is totaled using the "departure station, arrival station, departure time rounded according to the counting unit" as a key (step 1006).

After completing the processing for all trip IDs, the number of original passenger demand information before aggregation is compared with the number after aggregation to determine how much the calculation time can be shortened, which is shown in FIG. Display on the setting screen. The setting screen may be displayed using a normal image display device or printer of the information processing device. Further, the passenger demand information 1008 after aggregation is output as intermediate data (step 1007). The passenger demand information 1008 may be stored in a temporary memory or the like of the information processing device so that it can be used.

How long it takes to execute the boarding train estimation program for the original passenger demand information before aggregation can be roughly estimated from the specifications (CPU and memory) of the execution environment, and in step 1002. I will ask for it.

The system operator 114 and the commander 162 can confirm the calculation time reduction effect displayed on the setting screen and decide whether or not to execute the calculation. When executing the calculation, the calculation is performed using the passenger demand information 1008 after aggregation output as intermediate data.

The data structure of the passenger demand information 1008 after aggregation is the same as that of the original passenger demand information 124. Therefore, the processes after the process 904 of FIG. 9 can be performed in the same manner by using the passenger demand information 1008 after aggregation. However, since the number of data (trip ID) is reduced, the time for processing is shortened. Further, since the number of data (trip ID) of the output user behavior prediction information 127 is also reduced, the particle size of the information is increased.

FIG. 13 is a flowchart of the train occupant counting program 136. The train occupant aggregation program 136 reads the user behavior prediction information 127 output as a result of the boarding train estimation program 134, and outputs the train congestion prediction information 125. The train occupant aggregation program 136 is automatically executed after the execution of the boarding train estimation program 134 is completed, in association with the execution timing of the boarding train estimation program 134.

First, the user behavior prediction information 127 is read (step 1301).

A variable for aggregation in the form of train congestion prediction information 125 is defined and initialized (step 1302).

The following processing is repeated for the trip ID of the user behavior prediction information 127 (step 1303).

The following processing is repeated for each trip ID as many times as the number of routes used (step 1304).

In order to output the train boarding personnel prediction information 420, the train ID, the boarding station ID, and the getting-off station ID stored in the user behavior prediction information 127 are acquired, and the trip data between the boarding station and the getting-off station is obtained. The value of the number of passengers is added (step 1305).

Further, in order to output the boarding / alighting number prediction information 430, the value of the number of people in the trip data is added using the "train ID, route ID, boarding station ID, and getting-off station ID" as keys (step 1306).

After completing the processing for all trip IDs, the aggregation result is stored in the train congestion prediction information 125 (step 1307).

FIG. 14 is a flowchart of the station staying number counting program 137. The station staying number counting program 137 reads the user behavior prediction information 127 output as a result of the boarding train estimation program 134, and outputs the station congestion prediction information 126. The station staying number counting program 137 is automatically executed after the execution of the boarding train estimation program 134 is completed in association with the execution timing of the boarding train estimation program 134.

First, the user behavior prediction information 127 is read (step 1401).

A variable for totaling in the format of station congestion prediction information 126 is defined and initialized (step 1402).

The following processing is repeated for the trip ID of the user behavior prediction information 127 (step 1403).

The following processing is repeated for each trip ID as many times as the number of routes used (step 1404).

In order to output the number of people prediction information 440 for each movement direction, the information is totaled for each movement direction in the station yard (step 1405). Specifically, first, the departure station ID stored in the user behavior prediction information 127 is acquired, converted into an area ID with reference to the station yard equipment information 460, and held as the start point area ID. In addition, the route ID to be boarded first is acquired, similarly, the station premises equipment information 460 is referred to, converted into an area ID, and held as the end point area ID. The number of people is totaled using the "start point area ID, end point area ID, and departure time" as keys. Here, as the time of the number of people prediction information 440 for each movement direction, the departure time stored in the user behavior prediction information 127 may be used as it is, or it is aggregated according to a designated time unit such as 1 minute unit or 5 minute unit. Values may be used. For trips including transfers, the same process as above is repeated for all routes used. That is, the number of people is similarly converted by converting to the start point area ID based on the route ID before transfer, converting to the end point area ID based on the route ID after transfer, and using the time when the train before transfer is disembarked. I will add up. The same applies to the processing of the part that exits the ticket gate of the arrival station after getting off the last line.

Next, in order to output the information 450 for predicting the number of people staying at each destination, the information is totaled for each combination of the staying station and the arriving station (step 1406). Specifically, first, the departure station ID stored in the user behavior prediction information 127 is acquired, and the number of people is added using the "departure station ID, arrival station ID, departure time" as keys. For trips including transfers, the number of people is added for all transfer stations using the "transfer station ID, arrival station ID, and train disembarkation time before transfer" as keys (step 1406).

After completing the processing for all trip IDs, the aggregation result is stored in the station congestion prediction information 126 (step 1407). Here, as the time of the staying number prediction information 450 for each destination, the departure time stored in the user behavior prediction information 127 or the train disembarkation time at the transfer station may be used as it is, or in 1-minute units, 5 Values aggregated according to a specified time unit such as minutes may be used.

15 to 19 and 22 to 27 are diagrams showing an example of a screen distributed by the information distribution server 113. 15 to 19 and 22 to 27 are screens distributed to the terminals of the system operator 114, the operation management center 163, and the terminals 161 of 160 such as passengers, and display the congestion prediction results of trains and stations. The features of each screen will be described.

FIG. 15 is a screen for taking a bird's-eye view of the congestion degree of trains and stations in the entire metropolitan area to be calculated, and the map screen 1501 and the congestion degree of trains running on each line are displayed in the screen 1500. It has a congestion degree display area 1502. The map screen 1501 can be operated using an input interface such as a mouse or keyboard. For example, the map screen can be zoomed in / out with a wheel button, or the map display position can be changed by dragging the mouse. Perform the operation of.

FIG. 16 shows an example in which the display is changed by using the input interface. FIG. 16 shows an example in which the map screen is enlarged and displayed by the zoom-in operation on the map screen 1501 of FIG.

In FIGS. 15 and 16, the congestion degree display area 1502 surrounding the map screen 1501 is not limited to the annular shape shown in FIGS. 15 and 16, and only one of the sides may be used. The arrangement of the routes in the congestion degree display area 1502 may be determined in consideration of the actual spatial position information, or may be arranged in the order of names. Further, in a large-scale metropolitan area, the number of routes is large, and it can be assumed that it will be difficult to list all the routes due to the space on the screen. Therefore, the routes to be displayed may be selected. Alternatively, only the routes displayed in the map screen 1501 may be displayed in the congestion degree display area 1502. In that case, the list of routes displayed in the congestion degree display area 1502 also changes dynamically according to the viewpoint position of the map screen 1501.

As a typical example of the contents displayed in the congestion degree display area 1502 surrounding the map screen 1501, the average value of the congestion degree of the stations constituting each line and the average number of passengers of the train traveling on each line Values and maximum values can be mentioned. The degree of congestion at stations is obtained from the area information for each station and area and the predicted number of people staying. Alternatively, the train occupancy rate (a value obtained by referring to the capacity information 338 of the train timetable information 330 and dividing the number of passengers by the capacity information) may be displayed. In addition, different colors are defined for each of several stages, such as 0% to less than 50%, 50% to less than 100%, 100% to less than 150%, and 150% or more with respect to the numerical value of the occupancy rate. The color of the rectangle of each line in the congestion degree display area 1502 may be changed. In addition, the threshold value of the number of people staying in each area included in the equipment information in the station yard may be referred to, and the color of the line including the station exceeding the threshold value may be changed. In addition, a warning may be displayed to inform the user that the congestion is higher than expected by enclosing it in a frame, changing the color or size, or displaying a pop-up.

It is desirable that the display contents and color coding of the congestion degree display area 1502 can be arbitrarily set by the user. The display content may be selected by an interface in which a single candidate can be selected or an interface in which a plurality of indexes can be selected at the same time. When the information distribution server 113 receives the condition input by the user or the operation by the input interface of the mouse or the keyboard by the condition setting function on the screen, the information distribution server 113 recreates the display screen according to the condition. Deliver the request to the device that sent it. These processes are executed by the display screen generation program 154 and the information distribution program 155.

In this way, by presenting a screen that gives a bird's-eye view of the current situation regarding train and station congestion to system operators and railway operators, it is easy to understand the need for passenger guidance services at stations and train operation adjustments. You can get information to understand the situation.

FIG. 17 is a diagram showing an example of a setting screen for selecting a route to be displayed. It is called from the screen 1500 shown in FIG. 15 and used. Although FIG. 17 shows an example of selecting a display route, the route may be divided into several subsections and listed, and the routes may be selected by the same operation.

FIG. 18 is an example in which the display contents of the congestion degree display area 1502 are displayed for each subsection of a specific route. The method of dividing the partial section may be set in advance by the system operator or the user, or it may be automatically divided by an arbitrary number of divisions centering on the main station (station with a large number of users). You may. The arrangement of the partial sections in the congestion degree display area 1502 may be determined in consideration of the spatial position information of the head station and the main station of each section, or may be arranged in the order of names. Alternatively, as shown in FIG. 18, it may be displayed on the extension of the drawing position on the central map screen 1501, and the station position on the central map screen and the position on the congestion degree display area 1502 are indicated by a line. The correspondence may be displayed in an easy-to-understand manner by connecting them.

As for the display contents of the congestion degree display area 1502, as described above, the average value of the congestion degree of the stations constituting each subsection, the average value and the maximum value of the number of passengers of the train traveling in each section, and the like can be mentioned. .. Further, the occupancy rate may be used instead of the number of passengers on the train. At this time, the display size of each subsection may be adjusted according to the size of the index to be displayed, such as the total number of people staying in each subsection and the average value of the degree of congestion. The same applies to the color coding of each section displayed in the congestion degree display area 1502.

FIG. 19 is an example in which the display contents of the congestion degree display area 1502 are displayed for each station on a specific line. The arrangement of stations in the congestion degree display area 1502 may be determined using latitude / longitude information in consideration of the spatial position information of each station, or may be arranged in order of name. Alternatively, as shown in FIG. 19, it may be displayed on the extension of the drawing position on the central map screen 1501, and the station position on the central map screen and the position on the congestion degree display area 1502 are indicated by a line. The correspondence may be displayed in an easy-to-understand manner by connecting them.

As for the display contents of the congestion degree display area 1502, as described above, the total number of people staying at each station, the average value of the congestion degree, the average value and the maximum value of the number of passengers of the train traveling in each section, and the like can be mentioned. Further, the occupancy rate may be used instead of the number of passengers on the train. At this time, the display size of each station may be adjusted according to the size of the index to be displayed, such as the total number of people staying at each station and the average congestion degree. The same applies to the color coding of each station displayed in the congestion degree display area 1502 as in FIG. When it is difficult to list all stations due to the space of the screen, the user may be able to select the station to be displayed by using the same setting screen as the route selection screen shown in FIG.

FIG. 20 is a flowchart of a process for determining the positions of lines and stations to be displayed in the congestion degree display area 1502. For example, as shown in FIG. 19, the present invention relates to a process of determining at which position the information of each station is to be displayed on the screen for displaying the congestion status for each station in the congestion degree display area 1502.

In FIG. 20, as an example, processing for a mode in which the drawing position of each station on the central map screen 1501 and the drawing position on the congestion degree display area 1502 correspond to each other will be described. First, with respect to the selected line, the station / line / train information 123 is referred to, and the station information is read (step 2001). Next, the latitude / longitude information (Cx, Cy) of the center point of the drawing range of the map screen 1501 is acquired (step 2002). The following process is repeated for all stations constituting the selected line (step 2003). With reference to the latitude / longitude information (Sx, Sy) of each station included in the station / route / train information 123, the rotation angle θ with respect to the center point (Cx, Cy) of the map screen 1501 is calculated (step 2004).

θ = tan ^-1 ((Sy-Cy) / (Sx -Cx))
After obtaining the rotation angle θ for all stations, sorting is performed (step 2005). The sorted station order is stored as intermediate data (step 2006).

FIG. 21 is a flowchart of a process for determining the display size of a line or station to be displayed in the congestion degree display area 1502. In FIG. 21, as an example, a diagram showing the display contents of the congestion degree display area 1502 shown in FIG. 19 for each station on a specific line will be described.

First, with respect to the selected line, the station / line / train information 123 is referred to, and the station information is read (step 2101). Next, the station congestion prediction information 126 is read (step 2102). The following process is repeated for all stations constituting the selected line (step 2103). The following process is repeated for all areas constituting the target station (step 2104). The congestion rate Vn (n = 1 to n) of each area is calculated from the station congestion prediction information and the area area information (step 2105). After completing the processing for all stations, the sum Vtotal is calculated (step 2106). Finally, the occupancy rate of each station is calculated by Vn / Vtotal (step 2107).

22 and later are diagrams showing another example of the screen distributed by the information distribution server 113. FIG. 22 is an example of a screen displaying congestion in a station yard on a specific line and a specific station. FIG. 22 shows an example in which the screen is divided and displayed for each main area so that the transition of congestion from the ticket gate to the platform (or vice versa) can be seen in the premises of each station. With such an expression, it becomes possible to visualize the congestion situation with a common expression at any station, and it becomes easy to deploy to a plurality of stations. It also has the advantage that it is easy to compare congestion between stations. The division of the area of each station can be realized by referring to the station yard facility information 460.

Regarding switching of stations to be displayed, for example, as shown in FIG. 22AA, a list of stations is displayed on the left side, and switching can be performed arbitrarily by clicking a mouse or the like. As for the display position of the station list, it may be displayed side by side at the upper position of the screen as shown in Fig. 22B, and there are various other variations such as using the lower side and right side of the screen, and the upper side and the right side in combination. Conceivable.

As for the position to be displayed, a setting common to all routes may be used, or the display may be performed at a different position for each route. In the latter case, the system operator or the user may set in advance in consideration of the shape of each route. The display size of each area may be a fixed width, a fixed height, or as described above, the display size (width, height, or both) may be adjusted according to the degree of congestion. Further, the display color of the rectangle, the size of the text, and the color may be changed according to the degree of congestion.

FIG. 23 is an example of a screen displaying congestion at a specific station. In the case of a station that holds a layout diagram of the station yard with station / route / train information 123, it is possible to express such as superimposing the degree of congestion on the image data. Specifically, there are methods such as displaying the degree of congestion for each main area on the layout diagram and creating a heat map using colors defined according to the degree of congestion. For stations consisting of multiple floors, it is desirable to have a function that allows the user to specify / switch the floor to be viewed.

FIG. 24 is an example of a screen that visualizes the equipment operation status in the station yard for a specific station. The screen of FIG. 24 is configured to notify the current status of whether or not the equipment installed in the station yard, such as a ticket gate and an elevator, is operating normally. For example, it is specified in FIGS. 19 and 22. When it is found that the station is unusually crowded compared to usual, it can be used as a judgment material for identifying the cause. As usual, for devices that are not in operation, you may change the color or automatically notify you of an alert with a pop-up or the like. The operating status of each device may be automatically uploaded from the device to the center system, or may be updated based on the information manually input by the station staff at the site. As a result, when the station is congested, the railway operator can share the situation among the parties concerned without actually going to the site to check it, which leads to improvement of work efficiency.

FIG. 25 is a diagram showing another example of a screen for visualizing the degree of congestion of a partial section or each station for a specific line. The intention of the screen of FIG. 25 is to provide a screen that maximizes the spatial structure of the target route in order to intuitively notify the situation to the user. For example, when a line having the shape of a ring road is selected, the ring road is represented as a plan view as shown in the left region 2501 of FIG. 25. Alternatively, as shown in the right region 2502 of FIG. 25, a method of three-dimensionally expressing the ring shape and displaying the degree of congestion by the height of the shaded portion can be considered. In addition, as shown in the lower area 2503 of FIG. 25, stations may be lined up on the line. One of the plan view expression, the three-dimensional expression, and the line expression may be selected, or a plurality of combinations may be arranged on one screen as shown in FIG. 25. In the representation of the plan view, the display size of each station may be displayed evenly, or as described above, the display size may be determined according to the degree of congestion and dynamically changed. The same applies to the representation of 3D drawings and the representation on lines.

FIG. 26 is a diagram showing another example of a screen for visualizing the degree of congestion of a partial section or each station for a specific line. When it is desired to compare the congestion status of a plurality of stations, the expression on the dashboard as shown in FIG. 26 is also effective. In the example of FIG. 26, the congestion degree of each station is shown by a pie chart and a bar graph divided for each main area, but the present invention is not limited to this, and may be displayed in the form of a radar chart or the like. Further, it may have a function of selecting and arranging arbitrary stations instead of comparing only stations within a specific line. Furthermore, the same station may be compared side by side with different analysis periods.

FIG. 27 is a diagram showing another example of a screen for visualizing the degree of congestion of a partial section or each station for a specific line. The intention of the screen of FIG. 27 is to provide a screen that allows one to grasp the degree of congestion of a specific station and at the same time to see the degree of congestion of trains arriving and departing from that station at the same time. For example, in FIG. 27, the degree of congestion at a station is shown in the left area of the screen as an allowable rate. The permissible rate is synonymous with the capacity of the number of people staying at the station, and can be calculated based on the area area and the threshold value of the number of people staying at the station. In addition, the same concept of tolerance can be applied to the degree of congestion of trains departing from and arriving at the target station. That is, since it is possible to calculate how many more passengers can be boarded from the difference between the capacity information of each train and the current passenger information, the value is displayed as the allowable number of people on the icon of each train as text. Expression is possible.

The information distribution server 113 may create a screen to be distributed by combining a plurality of display screen generation programs 154 according to the characteristics of the device to be distributed and the content of the information to be distributed. For example, the technology of the web server can be used for the distribution of the screen, and the information to be distributed can be viewed by the web browser executed by the device of the distribution destination. A dedicated application executed on the distribution destination device may create a screen to be displayed using the data transmitted from the information distribution server 113.

As described above, a typical example of the embodiment is a train / station congestion prediction system including a processor that executes a program and a storage device that stores the program. Some of the boarding / alighting facilities and vehicles for using the means of transportation are equipped with measuring means for detecting the number of passengers. The storage device stores the number of passengers partially measured by the measuring means. The processor aggregates passenger data according to the specified estimated end time of calculation, and quickly calculates the number of people staying in all vehicles and all boarding / alighting facilities of the target transportation network. With such a configuration, congestion of trains and stations can be predicted at high speed based on the number of passengers partially collected from boarding / alighting facilities and vehicles, and can be continuously visualized on one screen.

As described above, according to the embodiment of the present invention, it is possible to predict the congestion of trains and stations, and visualize and distribute the predicted result to railway operators and users. As a result, the railway operator can judge the necessity of business improvement such as reviewing the train operation plan when an abnormality occurs and making a decision to increase the number of station staff. In addition, users can make travel plans to avoid crowded stations. As a result, in the case of passengers who can use a plurality of stations, the vacant stations can be preferentially selected and used, which leads to the dispersion of congestion. Transport operators can ensure the safety of stations and the like by allocating personnel and regulating them according to the degree of congestion at the stations and the like. In addition, many means of transportation (buses, taxis, etc.) can be dispatched to crowded landings to alleviate congestion in a short time.

In addition, since the number of people using transportation means is estimated from any of the data sources such as ticket gate passage data, surveillance camera images, and public wireless LAN connection information, train and station congestion can be achieved without making a large capital investment. You can know the degree. In addition, passengers can select and use vacant trains. Transport operators can use it as basic data when formulating congestion mitigation measures (for example, timetable revision).

The present invention is not limited to the above-described embodiment, and includes various modifications and equivalent configurations within the scope of the attached claims. For example, the above-described examples have been described in detail in order to explain the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to those having all the described configurations. Further, a part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Further, the configuration of another embodiment may be added to the configuration of one embodiment. In addition, other configurations may be added / deleted / replaced with respect to a part of the configurations of each embodiment.

Further, each of the above-described configurations, functions, processing units, processing means, etc. may be realized by hardware by designing a part or all of them by, for example, an integrated circuit, and the processor realizes each function. It may be realized by software by interpreting and executing the program to be executed.

Information such as programs, tables, and files that realize each function can be stored in a memory, a hard disk, a storage device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

In addition, the control lines and information lines indicate those that are considered necessary for explanation, and do not necessarily indicate all the control lines and information lines necessary for implementation. In practice, it can be considered that almost all configurations are interconnected.

Data server 111, data storage unit 121, station / route / train information 123, passenger demand information 124, train congestion prediction information 125, station congestion prediction information 126, user behavior prediction information 127, calculation server 112, storage unit 133, boarding Train estimation program 134, passenger demand information aggregation program 135, train occupant aggregation program 136, station staying number aggregation program 137

Claims (15)

It consists of an information processing device that can use station / route / train information and passenger demand information.
The station / route / train information stores the railway network route information and timetable information.
The passenger demand information includes a plurality of trip data and includes a plurality of trip data.
One of the trip data is the number of passengers moving, the departure station which is the station where the passenger departs, the arrival station which is the target station of the passenger, and the departure time relating to the time when the departure station departs. Including and
The station / route / train information and the passenger demand information are input, and the trip data is used by referring to the departure station, the arrival station, the departure time, the route information, and the timetable information. By associating possible trains, it is equipped with a boarding train estimation means that estimates and adds the train on which the passenger is boarding and generates user behavior prediction information.
A passenger demand information aggregating means for aggregating the passenger demand information processed by the boarding train estimation means by combining a plurality of the trip data is provided.
That processes the passenger demand information aggregated by the passenger demand information collecting means by the boarding train estimation means, transportation congestion prediction system. The passenger demand information aggregation means determines how much the calculation time can be shortened based on the number of trip data of the passenger demand information before aggregation and the number of trip data of the passenger demand information after aggregation. Obtained and displayed on the output device in association with the degree of aggregation,
The transportation congestion prediction system according to claim 1. The passenger demand information aggregating means combines a plurality of the trip data by aggregating the departure times in a predetermined aggregation unit.
The transportation congestion prediction system according to claim 1. The passenger demand information aggregating means limits the departure time to a specific combination of the departure station and the arrival station and summarizes the departure time in a predetermined aggregation unit.
The transportation congestion prediction system according to claim 3. The passenger demand information aggregating means collects the departure time in a predetermined aggregation unit only in a predetermined time zone.
The transportation congestion prediction system according to claim 3. The passenger demand information aggregation means changes the aggregation unit for each time zone.
The transportation congestion prediction system according to claim 3. One of the trip data further includes a route from the departure station to the arrival station.
The passenger demand information aggregation means changes the aggregation unit based on the route used in the route.
The transportation congestion prediction system according to claim 3. The passenger demand information aggregation means sets an upper limit so that the number of passengers in the trip data of the passenger demand information after aggregation does not exceed a predetermined threshold value.
The transportation congestion prediction system according to claim 3. The boarding train estimation means further estimates and adds a boarding station and a disembarking station for each of the boarding trains to each of the trip data, and generates the user behavior prediction information.
The transportation congestion prediction system according to claim 1. In addition, it is equipped with a means for counting train crews.
The train occupant counting means receives the user behavior prediction information as an input, and uses the above-mentioned user behavior prediction information as an input.
Generates train passenger number prediction information, which is the result of totaling the number of passengers between predetermined stations for each train.
The transportation congestion prediction system according to claim 9. In addition, it is equipped with a means for counting train crews.
The train occupant counting means receives the user behavior prediction information as an input and
Generates passenger number prediction information, which is the result of totaling passengers for each train by the combination of the boarding station and the getting-off station.
The transportation congestion prediction system according to claim 9. In addition, it is equipped with a means for counting the number of people staying at stations.
The station staying number counting means inputs the user behavior prediction information, and uses the user behavior prediction information as an input.
The destination-specific resident number prediction information, which is the result of totaling the resident number of passengers at each station for each arriving station, is generated.
The transportation congestion prediction system according to claim 9. It is a congestion prediction method that uses an information processing device that can use station / route / train information and passenger demand information.
The station / route / train information stores the railway network route information and timetable information.
The passenger demand information includes a plurality of trip data and includes a plurality of trip data.
One of the trip data is the number of passengers moving, the departure station which is the station where the passenger departs, the arrival station which is the target station of the passenger, and the departure time relating to the time when the departure station departs. Including and
Input step for inputting the station / route / train information and the passenger demand information,
By associating each of the trip data with a train that can be used by referring to the departure station, the arrival station, the departure time, the route information, and the timetable information, the train on which the passenger boarded can be obtained. Boarding train estimation step, which estimates and adds, and generates user behavior prediction information,
A passenger demand information aggregation step, which aggregates the passenger demand information processed in the boarding train estimation step by combining a plurality of the trip data, is provided.
That processes the passenger demand information aggregated by the passenger demand aggregation step in the boarding train estimation step, the congestion prediction method.
The passenger demand information aggregation step determines how much reduction in calculation time is expected based on the number of trip data of the passenger demand information before aggregation and the number of trip data of the passenger demand information after aggregation. Obtained and displayed on the output device in association with the degree of aggregation,
The congestion prediction method according to claim 13. The passenger demand information aggregation step combines a plurality of the trip data by summarizing the departure times in a predetermined aggregation unit.
The congestion prediction method according to claim 13.

Images