JP2014213697A - System for collection and delivery correspondence of information on congestion situation - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a high-speed and high-precision system for collection and delivery correspondence of information on congestion situation.SOLUTION: In a system 100 for collection and delivery correspondence of information on congestion situation, a plurality of computers are connected via a network. The system includes a ticket gate 101, a control monitoring server 102, a gate information collection and delivery correspondence server 103, an analysis server 106, a management information server 107, a delivery server 104, a supervisor terminal 109, and a user terminal 105.

Description

  The present invention relates to a system for collecting and distributing human flow information in a transportation facility such as a railway, and an equipment control system.

  The railway network is becoming increasingly complex. In particular, timely and appropriate information provision and equipment control are important in the event of a large-scale disaster or abnormality. Furthermore, the ticket gate information is an information source that can grasp the flow of passengers, and research and development are actively conducted on a new service for collecting and distributing the information in real time while protecting the privacy.

  Once the ticket gate information is accumulated, the user's behavior pattern can be extracted using means such as data mining (Patent Documents 1 to 4).

JP 2001-55145 A JP 2010-247728 A JP 2007-118705 A JP 2007-22497

  However, simulation based on past data and real-time status grasp are different issues. Especially in urban areas, the ticket gate information is big data where a large number of transactions fly, so it is difficult to aggregate in real time as the flow of people and congestion. From the viewpoint of utilizing recent computer performance such as a processor dedicated to floating-point processing and a parallel processing OS, the analysis logic for simulation purposes using many IF-THEN rules is likely to become a processing speed bottleneck and is not suitable for real-time stream processing. .

  In a system that collects and distributes ticket gate information, it has been a challenge to achieve both high-speed processing and improved accuracy of situation estimation.

  In order to solve the above problems, the present invention has the following configuration according to one aspect.

  The congestion status information collection / distribution system is an information collection / distribution system in which a plurality of computers are connected to a network, including a ticket gate, a control monitoring server, a ticket inspection information collection server, an analysis server, a management information server, a distribution server, an administrator terminal, Consists of user terminals. The computer includes a memory, a CPU, a disk, and input / output means and is connected to each other via a network.

  The management information server includes a management database, and stores and manages management information in the management database. The management information includes performance information such as congestion status information, analysis information, and related information. The related information includes operation information, vehicle overload information, route information, and the like. Further, the related information includes attribute information such as a ticket, a user, contract information, a station, and a ticket gate. For example, by searching the management database, the station name character string can be referred to from the boarding station ID, and the sex of the passenger can be referred to the management information server from the ticket gate information. The related information may be stored using the input means of the administrator terminal, or may be received from a vehicle or the like by communication. A table of operation information, route information, ticket gate information, prediction information, and current status information may be managed in advance and / or sequentially.

  The ticket gate is equipped with an open / close door and a ticket reading device. When a passenger with a ticket passes through the ticket gate, it acquires ticket information including ticket gate passage time, boarding station ID, ticket ID, and commuter pass information. The ticket gate information is transmitted to the control monitoring server. The control monitoring server controls one or more ticket gates and receives ticket gate information. Control of the ticket gate includes opening and closing of the door and limitation of the passing direction. The ticket gate information collection server collects and accumulates ticket gate information from one or more control monitoring servers, and stores the ticket gate information in the ticket gate information table.

  The analysis server analyzes the ticket gate information, calculates congestion status information, and stores it in the congestion status information table. The distribution server distributes with reference to the congestion status information. The user terminal receives the congestion status information and outputs the congestion status information.

The analysis server
Summing up the preliminary ticket gate information table using the ticket gate ID as a key, and storing the congestion status of each ticket gate in the current status information table;
Using a time lookup table prepared in advance or a primary function of time, the step of determining the departure station, arrival station or other station of the commuter pass section as the scheduled departure station of the passenger according to the ticket passing time;
Calculating the residence time and the residence time at the boarding station, scheduled departure station and / or halfway station, and storing in the prediction information table;
Calculating the total estimated residence time stored in the prediction information table, and storing the congestion status for each station and / or time zone in the current status information table;
Using the route specified by the operation information as a key, updating the record by adding the expected time until operation restart to the residence time stored in the prediction information table;
Compare the congestion status for each time zone of the current status information table and the stay limit number of people for each station registered in advance, check which time zone will exceed the stay limit, set the attention flag,
The system user can be alerted at the user terminal when the attention flag is set.

  Users of the congestion status information collection and distribution system are station staff and general users. General users include passengers. Station staff can access all components using an administrator terminal and a user terminal, and general users use the user terminal to obtain distribution information.

  The technology disclosed in this specification can be applied to software, devices, services, systems, contents, media, and applications for processing.

  The effects obtained by the above embodiment according to one aspect are as follows.

    According to the present invention, it is possible to provide a high-speed and high-precision congestion status information collection and distribution system. Furthermore, in the application system of the present invention, an equipment control system that collects and distributes congestion status information can be realized, which contributes to alleviating the congestion status and optimizing human flow.

It is a figure which shows the whole system structure of the collection and distribution system of congestion condition information. It is a figure which shows the structure and data example of a ticket gate information table. It is a figure which shows the structure and data example of a prediction information table. It is a figure which shows the structure and data example of the present condition information table A (counter unit total). It is a figure which shows the structure and example of data of the present condition information table B (station unit total). It is a figure which shows the structure of an analysis server. It is a figure which shows the process sequence of the prediction process of an analysis server. It is a figure which shows the process sequence of the total process of an analysis server. It is a figure which shows the process sequence of the correction process at the time of abnormality of an analysis server. It is a figure which shows the structure of the delivery server of congestion condition information. It is a figure which shows the method of calculating an estimated destination station from ticket gate information and ticket gate time in the processing procedure of the prediction process of an analysis server. It is a figure which shows the table | surface which arranged the heuristic rule.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

  FIG. 1 shows the overall system configuration of a congestion status information collection / distribution system 100.

    The congestion status information collection / distribution system 100 is an information collection / distribution system in which a plurality of computers are connected to the network, and includes a ticket gate 101, a control monitoring server 102, a ticket gate information collection server 103, an analysis server 106, a management information server 107, a distribution. The computer is composed of a server 104, an administrator terminal 109, and a user terminal 105. A computer for the equipment control 110 may be included. Each of these computers includes a memory, a CPU, a disk, input / output means, and a timer, and is connected to each other via a network.

  The management information server 107 includes a management database, and the management information 112 is stored and managed in the management database. The management information 112 includes performance information such as congestion status information, analysis information, and related information. The related information includes operation information, vehicle overload information, route information, and the like. Further, the related information includes attribute information such as a ticket, a user, contract information, a station, and a ticket gate. For example, by searching the management database, a station name character string can be referred to from the boarding station ID, or the sex of the passenger can be referred to the management information server 107 using the ticket gate information 111. The related information may be stored using the input means 116 of the administrator terminal 109, or may be received from the vehicle 108 or the like by communication. There may be a ticket gate in the vehicle 108. A table of operation information, route information, ticket gate information 111, prediction information 113, and current status information 114 may be pre-registered and / or sequentially registered and managed in a management database.

  The ticket gate 101 includes an opening / closing door and a ticket reading device. When a passenger with a ticket passes through the ticket gate, the ticket gate information including ticket gate passage time, boarding station ID, ticket ID, and commuter pass information is acquired. The ticket gate information is transmitted to the control monitoring server 103. You may transmit to the equipment control 110. The facility control 110 and / or the control monitoring server 102 includes one or more ticket gates 101, elevators, control means such as passage gates, and means for transmitting and receiving information. Control of the ticket gate 101 includes opening / closing of the door and limitation of the passing direction. The ticket gate information collection server 103 collects ticket gate information from one or more control monitoring servers 102 and stores it in the ticket gate information table 111.

  The analysis server 106 includes a database, and has functions of prediction processing, tabulation processing, and distribution information creation processing in the analysis server memory, and analyzes the ticket gate information 111 under the control of the analysis server CPU, and predicts and summarizes the congestion situation. These are stored in the database table as prediction information 113 and current status information 114. The distribution server 104 distributes the congestion status information 115 with reference to the current status information 114 and the like. The user terminal 105 receives the congestion status information 115 and outputs it using the output means 117.

  The analysis server 106 includes an analysis server CPU, an analysis server memory, and an analysis server input / output unit. The analysis server memory is loaded with a program that defines various processing methods including prediction 118, aggregation 119, and distribution information creation 120, and the analysis server CPU controls execution of the program.

An outline of the operation of the analysis server 106 is as follows.
(1) The ticket gate ID is counted from the table contents of the ticket gate information 111 as a key, and the congestion status for each ticket gate is stored in the table of the current status information 114.
(2) Using a time lookup table prepared in advance or a linear function of time, the departure station, arrival station, or other station of the commuter pass section is determined as the scheduled exit station for passengers based on the ticket gate passage time. .
(3) Calculate the staying time and the staying time at the boarding station, scheduled getting-off station and / or halfway station, and store them in the prediction information 113 table. The total estimated residence time stored in the prediction information 113 table is calculated, and the congestion status for each station and / or time zone is stored in the current status information 114 table.
(4) Using the route information specified by the operation information as a key, the expected time until restart of operation is added to the residence time stored in the table of the prediction information 113, and the record of the prediction information 113 is updated.
(5) Compare the congestion status for each time zone with the table of the current status information 114 and the stay limit number of people for each station registered in advance to determine which time zone will exceed the stay limit and set a caution flag.
(6) When the caution flag is set, the user can be alerted to the system user at the user terminal 105.

  Next, the structure of a database and an example of data will be described with reference to FIGS.

  FIG. 2 is a diagram illustrating the structure of the ticket gate information table 201 and data examples.

    The ticket gate information table 201 is stored in the database of ticket gate information 111 by the ticket gate information collecting server 103. The ticket gate information table 201 includes fields of cID 202, boarding station ID 203, ticket gate ID 204, ticket gate time 205, commuter pass section 206, entrance / exit 207, and attribute 208. The cID 202 is an ID that uniquely identifies one ticket gate information, and may be, for example, a card number of a contactless IC card ticket, or another system converted from a card number using an ID linkage technology or the like. It may be an ID number. The attribute 208 field stores boarding history information of a non-contact IC card ticket, a part of management information 112 obtained by searching the management information server 107, and the like.

  FIG. 3 is a diagram illustrating the structure of the prediction information table 209 and data examples.

  The prediction information table 209 is stored in the database of the prediction information 113. The prediction information table 209 includes fields of rID 210, station ID 211, entrance time 212, expected residence time 213, calculation source 214, operation condition 215, and attribute 216. The rID 210 field is an ID that uniquely identifies one piece of prediction information. In the calculation source 214 field, information for identifying the ticket gate information from which the prediction information of the record is generated and cID 202 are stored. The operation condition information 215 stores operation route information that is a premise of prediction.

  FIG. 2C is a diagram showing the structure and data example of the current status information table A217.

  The current status information table A217 is stored in the current status information 114 database. The current status information table A 217 is a table mainly used for ticket gate unit aggregation, and includes gID 218, station ID 219, ticket gate ID 220, entry 221, exit 222, time 223, caution flag 224, and attribute 225 fields. The gID 218 field is an ID that uniquely identifies a record. The attribute 225 field stores the total time, relationship information, and the like.

  FIG. 2D is a diagram showing the structure and data example of the current status information table B226.

  The current status information table B 226 is stored in the current status information 114 database. The current status information table B226 is a table mainly used for station unit aggregation, and includes sID227, station ID228, stay limit 229, attention flag 230, staying number 231, 5 minutes 232, 10 minutes 233, 15 minutes 234, attributes. It has 235 fields. The sID 227 is an ID that uniquely identifies a record. The attribute 235 field stores the total time, reliability information, prediction accuracy, relationship information, and the like.

  FIG. 3 shows a detailed configuration of the analysis server 106.

  The analysis server 106 includes a program describing the processing methods of the prediction 118, the summation 119, the abnormality correction 306, the behavior hypothesis scenario processing 307, the storage analysis 308, and the distribution information creation 120 on the memory 305, and a work such as a buffer. An area is provided, and processing is executed under the control of the CPU 301. The analysis server 106 inputs / outputs operation information 311, route information 312, performance analysis information 311 including performance information and analysis information, etc. using the input means 303 or output means 304. The database of the performance information 311 may be provided on the analysis server disk 302 or may be provided on the computer disk of the management information server 107. The analysis server 106 receives the ticket gate information 111 output from the ticket gate information collection server 103, and creates prediction information 309, current status information 310, analysis information 311 and the like on the memory 305 or the disk 302.

  The ticket gate information 111 is handled as real-time stream information with buffering by the input means 303 of the analysis server 106, and is received continuously and irregularly. The analysis server 106 performs processing such as prediction 118, totaling 119, etc. by taking out a plurality of ticket gate information in the buffer one by one or a plurality of pieces. If there is a change in the operation information 311, the abnormality correction 306 is processed. The distribution information creation 120 outputs current status information 310 and the like.

  In FIG. 3, the case where the passenger with the commuter pass between A station and C station processes ticket gate information when entering A station is illustrated. According to the behavior hypothesis scenario 307, the passenger gets off at the station C and is interpreted as movement for five people. However, not all commuter passes are necessarily commuter passes, so an overall estimate is required. The route information 312 is information including a graph structure 315 of a route including a halfway station, a timetable, a standard transit time at each station, and a walking time on the premises, and is managed.

  With reference to the route information 312, the prediction 118 process summarizes the expected stay time 314 of each station. In the totaling 119 process, the actual totaling result number of persons is evaluated five times and output. Further, when an abnormality occurs, the administrator operates the administrator terminal 109 and inputs by the input means 116 that the n-line is stopped during the time period of 08: 30-08: 45. Furthermore, the input contents are stored as operation information in the management information 112 and further transmitted to the analysis server 106 as operation information 311, so that it is possible to appropriately analyze and total even in an abnormal time.

  The storage and creation of the result analysis information 311 is performed by the storage analysis process 308 of the analysis server 106. For a large number of passengers, the actual number of times of getting on and off is tabulated based on the ticket gate information 111. You may include stations in the commuter pass section. In the totalization, the actual number of times of getting on and off at each station for each individual is divided by the sum of them and converted into a numerical value of occurrence frequency, and then stored as analysis information. The analysis information is a record set in the form of cID, boarding station ID, getting-off station ID, and occurrence frequency (%), and is associated with each ID number in the ticket gate information table 201. If cID is not included, the number of records is the square of the total number of station IDs. If cID is included, the number of records is further multiplied by the order of the number of passengers. Still other attributes may be included.

  Since the use frequency distribution at the boarding station is greatly biased, the prediction process can be performed more accurately by referring to the record information. Since the performance information has a large index scale, it takes a certain amount of calculation time even when using the search. However, in this embodiment, the frequency of reference to the record information is limited by using the commuter pass information in the ticket gate information, so the processing speed can be increased. That is, the entire ticket gate information real-time stream process can be processed efficiently.

  By facilitating the analysis and counting process of ticket gate information, there is a margin in calculation resources, and the calculation resources can be used appropriately by increasing the counting frequency and performing detailed analysis of important cases. In addition, the latest information can be obtained in order to grasp the current situation. It contributes to the appropriate control of equipment and equipment, the optimization of the campus environment, the improvement of passenger services by reducing congestion, and the improvement of system users.

  FIG. 4 is a diagram showing in detail the processing procedure of the prediction processing 118 of the analysis server 106.

  In the analysis server 106, after initialization of the buffer memory and the like in step 401, the following prediction process is repeatedly controlled in step 402 for the ticket gate information 111 that is continuously stream-inputted.

    1. In step 403, the ticket gate information 111 is read.

    2. In step 404, an expected destination station is estimated using information including the boarding station ID and the ticket gate time in the ticket gate information table 201 at the time of entrance. Details of the estimation means will be described later.

    3. In step 405, the route information 312 is referenced to obtain the route between the boarding station and the predicted destination station. For each boarding station (departure station), transfer station, and predicted destination station, the estimated arrival time, expected residence time, and route name as a route are acquired. The estimated residence time may be calculated as the difference from the latest train time included in the timetable information with respect to the scheduled arrival time of the passenger, or a fixed value for each station is registered in the route information 312 in advance. The value may be used as the expected residence time or may be calculated by combining both.

    4). In step 406, the information on the entrance time 212 (predicted arrival time and the actual value of the ticket gate information) and the expected residence time 213 for each station are stored in the prediction information table 209.

    5. In step 407, ticket gate information, prediction information, and time are stored as performance information 311. Analyze and accumulate as performance information when leaving.

    6). In step 408, the time is referenced, and an unnecessary record such as a record with an old time is updated or deleted every time a fixed number of cases are processed or periodically. In step 408, when the ticket gate information is an exit ticket, that is, a record that has exited the ticket gate, the prediction information having the value of the calculation source 214 field that matches the value of the cID 202 of the ticket gate information is deleted. Further, the relationship between the tables may be tracked, and 1 may be subtracted from the total value such as the number of people 231 staying at the related station in the current status information table B (total for each station). By appropriately invalidating old records, the accuracy of analysis increases, and the number of data to be processed that does not contribute to analysis and aggregation is reduced. Therefore, analysis and aggregation processing is accelerated and increased in frequency, and the situation can be grasped in real time. Will be able to.

  FIG. 5 is a diagram showing in detail the processing procedure of the aggregation processing 119 of the analysis server 106.

  After initialization of the buffer memory and the like in step 501, in step 502, the following aggregation processing is periodically controlled for the ticket gate information and the prediction information to summarize the current state of the congestion situation.

  In step 503, operation information is acquired.

  In step 504, if the operation information is updated, in step 505, an abnormality process is performed.

  In step 506, the following iterative process is performed for each ticket gate (ticket gate ID, station ID).

  In step 507, a record for a specific ticket gate ID and station ID is searched from the ticket gate information table, and the number of people entering and leaving in a predetermined counting time range (for example, 1 minute) is totaled. The information on the total time range is changed in accordance with the abnormal situation by storing a numerical value obtained by checking a normal value in advance in a memory or the like. The total time range may be added to the time included in the operation information 311 or may be changed according to the time range specified by the input means 116 of the administrator terminal 109. Further, the total results are stored in the current status information table A (counter unit total) as records corresponding to specific ticket gate IDs and station IDs.

  In step 508, the following repetitive processing is performed for each station (station ID).

  In step 515, the prediction information table 209 is searched and the number of staying persons is totaled. The total result is stored in the current status information table A (total for each station).

  In step 509, all the current status information tables are searched, and the difference value of the number of people entering and leaving the ticket gate and the number of people staying are acquired as current status information.

  In step 510, the current status information and the comparison information are compared with specific records in the current status information table A217 and the current status information table B218, and necessary correspondence is performed. The comparison information may be a fixed value determined in advance for each station or each ticket gate as an appropriate value such as a difference value of the entrance / exit of the ticket gate. The comparison information may be a retention limit 229. Alternatively, the value of the staying number may be calculated as a current status information evaluation value as a differential value or a second-order differential value indicating their temporal transition, and whether the current status information evaluation value is larger than the comparison information may be compared. If the current status information evaluation value is used, an appropriate response can be made because it is possible to detect sensitively, for example, when the temporal transition deteriorates at an accelerated rate.

  Further, in step 510, if the staying number information is larger than the comparison information, the attention flag 224 and / or the attention flag 234 is set in step 511. The comparison information may be a past value acquired from the performance information, or an average value or an analysis predicted value of the performance information may be used. Since passengers with commuter passes are a part of the whole, a numerical value obtained by multiplying the number of people indicated by the current status information by an appropriate variable value or fixed value may be calculated and compared with the calculation result.

  In step 512, distribution information is created based on the current status information and transmitted to the distribution server 104. It may be congestion status information 115 in which the distribution information reaches the user terminal 105 as it is.

  In step 513, the current status information, the total time, etc. are accumulated as performance information.

  In step 514, the current time is referred to by a timer to periodically search for and delete records having an old record registration time. The record registration time may be stored in the attribute field 208 or the like. This is the same as the processing in step 408, and the effectiveness of the tabulation result is enhanced by appropriately invalidating unnecessary records. Furthermore, since the number of data to be processed that does not contribute to analysis or aggregation is reduced, the aggregation process is speeded up and increased in frequency, and the situation in real time can be grasped.

  The cycle for regularly counting may be every other minute, or even real-time processing. There are a large number of people who are trying to use railways, or who are going to use railways in the future. As a whole, there are many people who can influence the action selection even if the information transmission is a difference of several seconds. If the situation can be grasped in real time, it will be possible to distribute more appropriate information such as congestion status, encourage passengers to wait a little, and recommend detours, etc. Contributes to In information distribution, broadcast, mail, twitter, sound, etc. can be used according to the output means of the user terminal.

  FIG. 6 is a diagram illustrating a processing procedure of the abnormality correction processing 306 of the analysis server 106.

  The analysis server 106 receives the input or change of the operation information 311 and performs the following abnormal process.

  In step 601, operation information 311 is read. The operation information 311 includes operation stop route information, operation resumption expected time information, delay time information, cause information, and the like.

  In step 602, all records that are affected by the operation stop are extracted for the operation condition 215 of the prediction information table 209 using the operation stop route information as a key.

  In step 603, with respect to the record that is affected by the operation stop, the route information 312 is referred to select a passenger to proceed to the alternative route, and the prediction content related to the passenger is changed. In the case where the destination remains the same and the route proceeds to another rail route, the station ID and the staying time of the record of the station on the way are updated using the route information content. The selection may be statistical or random, and there are rules such as "Persons whose time to change to an alternative route and go to the destination are shorter than the time to simply wait for resumption of driving will detour" May be selected. Passenger records leaving the railway and selecting an alternative transportation means are deleted by reflecting the ticket gate information content at the time of exiting the ticket gate in the prediction information table. The route information 312 is prepared in advance so as to obtain a plurality of information on transit routes, intermediate stations, transit times, station residence times, and priority orders including bus alternative transport cases corresponding to the departure station and arrival station. Register. The route information 312 is searched for the route with the highest priority without including the operation stop route specified by the operation information, and the passenger prediction process is performed again or excluded from the narrowing targets. You may narrow down records at a certain rate statistically.

  In step 604, the expected operation resumption time is added to the expected residence time to update the prediction information table. The total time range information is updated to a value obtained by adding the estimated recovery time to a predetermined value. If it returns to normal time by specifying the operation information, a predetermined value is used.

  In step 605, the current status information table is totaled again.

  In step 606, the residence limit is confirmed. Specifically, the field value indicating the prediction after 5 minutes or the like in the current status information table is compared with a predetermined value of the stay limit field.

  In step 607, the current status information table is searched, and the stay limit 229 is compared with the current value and the predicted value. As the predicted value, a value in a field such as the staying number 231 and 232 after 5 minutes is used. If there is a station exceeding the stay limit in the near future, a caution flag in the current status information table is set in step 607.

  In step 608, information before and after the update is accumulated as performance information.

  In step 609, the distribution information is recreated. The user terminal, user information, distribution method, etc. for monitoring the attention flag are registered in advance in the distribution server. If the attention flag is set, a signal is transmitted from the analysis server to the distribution server so as to push the warning to the user terminal, and immediate distribution processing is performed by the distribution server.

  FIG. 7 is a diagram illustrating a configuration of the distribution status information distribution server 104.

  The distribution server 104 is a computer, and functions that define processing procedures for device management 701, limited reception (PUSH) 702, status inquiry (PULL) 703, attention flag confirmation 704, and user screen configuration 705 installed in the distribution server memory. Provide modules. Each functional module is executed under the control of the distribution server CPU. Current status information 310 and management information 112 are acquired using the distribution server input means, and information is distributed to the user terminal 105 using the distribution server output means. The management information 112 includes vehicle response load, past situation, analysis information, map, attached information, guidance information, and the like.

  The user terminal 105 includes output means such as an input means 706 and a monitor 707. Users are station staff, passengers, or ordinary people. The user uses the input unit 706 to send a transmission request to the distribution server 104. In the distribution server 104, the requested information is returned to the user terminal 105 and output to the monitor 707 by the processing of the situation inquiry (PULL) 703.

  In limited reception (PUSH) 702, the current status information is sequentially transmitted or broadcast to the user terminal 105 in real time.

  The attention flag confirmation 704 process monitors whether there is a reception from the alert request 609 of the analysis server 106. If the attention flag is set, the information is distributed to the user terminal, and the color scheme of the screen is reversed or a warning is issued. Sound the user to call attention. Screen color inversion is performed in the user screen configuration 705.

  The distribution information is received using the user terminal input means, and is output to the user terminal using the user terminal output means such as a monitor. The distributed management information includes vehicle load, past situation, analysis information, map, store information, guidance information, and other information.

  The device management function performs registration of user terminals and equipment, user management, overall control, and the like.

  As functions for distributing information, there are a limited reception (PUSH) function and a situation inquiry (PULL) function, and distribution information is distributed from a distribution server to a user terminal client continuously or irregularly. It may be broadcast. When the attention flag confirmation function is used by the user's user terminal operation, when the attention flag is raised, control is performed to push the warning information from the distribution server. The warning information may be output to the monitor as text or an image, or a warning sound or the like may be emitted from a speaker provided as user terminal output means to prompt the user to pay attention.

  If the distribution information is distributed to the equipment control device and equipment control such as opening / closing of the ticket gate door and elevator stop is performed, the staying passenger can be appropriately guided. For example, the analysis server 106 includes a step of constantly monitoring the attention flag 224 or the attention flag 230, and a step of transmitting congestion state information including the ticket gate ID to the distribution server when the attention flag is set 704. The data is stored in the analysis server memory and controlled by the analysis server CPU. Similarly, when the distribution server 104 receives the congestion status information, the distribution server 104 transmits an opening / closing control signal of the ticket gate corresponding to the ticket gate ID to the control monitoring server. The ticket gate is provided with means for operating the door when receiving the opening / closing control signal. With such a configuration and operation, a ticket gate that is expected to be crowded can appropriately control a door in a specific direction when the opening / closing control signal is received. That is, the number of people who can pass the ticket gate per unit time can be suppressed and guided.

  FIG. 8 is a diagram illustrating a method for calculating a predicted destination station of a passenger from ticket information of a certain passenger in the processing procedure of the prediction process 118 of the analysis server 106. In addition, the details of the processing content of step 404 will be described.

  For passengers who have a commuter pass from the departure station 801 to the arrival station 802, the station attributes are configured as three types of departure station 801, arrival station 802, and other station 803, and each is encoded with values 1, 2, and 3 The railway movement model is used. There are six types of passenger behavior patterns for the combination of the boarding station and the getting-off station where the passenger actually gets on and off, except for the case of an admission ticket that gets off at the boarding station. The station that the passenger is about to go to is the information that only the passenger knows immediately after passing through the ticket gate, so it is called the predicted destination station in the sense that the information system estimates it.

  In this railway movement model, the following heuristic rules are established.

・ IF (time zone is around noon, the boarding station is the arrival station of the commuter pass section)
THEN Expected destination station is another station (3) BECAUSE business trip etc. ・ IF (time zone is weekday morning, boarding station is the departure station of commuter pass section)
THEN Expected destination station is arrival station (2) BECAUSE attendance etc. ・ IF (time zone is night, boarding station is out of commuter pass section)
THEN Expected destination station is departure station (1) BECAUSE Business trip bounce, etc. These heuristic rules can be summarized as Table 804 in FIG. The time zone 809 including the ticket gate passage time is divided into the first half 810 and the second half 811 and encoded as 1 and 2, respectively. According to Table 804, the expected destination station can be specified as either the departure station (1), the arrival station (2), or another station (3), depending on the attributes of the boarding station (3 types) and the ticket gate passage time (2 types). .

すなわち、改札情報テーブル２０１において、乗車駅ＩＤ２０３と改札時刻２０５のフィールド値を用いて予想目的駅を推定できる。予想目的駅の推定には、別の多次元の表を用いてもよいし別の関数を用いてもよい。 s: Station ID of boarding station
a, b: commuter pass section (station ID of departure station = a, station ID of arrival station = b)
x: Attribute of boarding station, x = f (s, a, b)
t: Ticket gate passage time y: Expected destination station ID y = g (x, t)
Since the time information is generally a continuous value expressed as a difference value from the reference time, the ticket gate passage time can be converted to the time zone information t at high speed by appropriate encoding means such as integerization by division. The same applies to x. Therefore, a high-speed logic having the same contents as the table 804 in FIG. 9 can be assembled as a primary expression of the time zone t as follows.

That is, in the ticket gate information table 201, the expected destination station can be estimated using the field values of the boarding station ID 203 and the ticket gate time 205. For estimation of the predicted destination station, another multi-dimensional table or another function may be used.

  If the predicted destination station is calculated as other station (3), the getting-off station is estimated from the history of the use station included in the ticket gate information. Alternatively, the getting-off station is estimated from the analysis information created in advance based on the result information obtained by combining the ticket gate information of the boarding station and the getting-off station. First, history and analysis information are searched using the boarding station ID of the ticket gate information as a key, and a search result is obtained in a form in which there are a plurality of combinations of the getting-off station ID and the occurrence frequency. Next, the getting-off station ID to be associated with specific ticket gate information is selected with a probability corresponding to the occurrence frequency. For example, the occurrence frequency percentage of the search result is added in descending order, and the getting-off station ID corresponding to the last occurrence frequency exceeding the pseudo random number value is designated.

  When the predicted destination station corresponding to one specific ticket information is determined as another station, the above method of setting the getting-off station ID calculated from the analysis information as the predicted destination station is one person corresponding to one ticket information. Although it is difficult to ensure consistency with the actual destinations of the passengers, it is consistent with the destination distribution of the plurality of passengers corresponding to the plurality of ticket gate information. Therefore, it is useful for grasping the overall congestion situation.

  Using the ticket ID as a key, it may be possible to search and set at a personal level or simulate which station the passenger is likely to get off. In that case, the search processing using the ticket ID as a key may be slow in processing time, for example, involving a huge table search equivalent to the number of issued tickets. On the other hand, using the above-described search method aggregated in the form of analysis information is effective in that processing results can be obtained quickly and easily with high accuracy without requiring special equipment based on commuter pass information.

  Analyzing the performance information and arranging the stations in order of the actual number of disembarkation times for a specific passenger, the number of disembarkation at the commuter pass departure and arrival stations is prominent, showing intermediate values at several major stations, It follows the Zipf law that there are many stations with almost no disembarkation record. In this embodiment, by utilizing the ticket gate time and commuter pass information, it is possible to estimate the departure station and the arrival station with prominent track record frequency as the predicted destination station with high speed and high accuracy. This speeds up the overall processing time. Furthermore, the congestion status based on the ticket gate information can be totaled in real time or with high speed and accuracy. By presenting the information together, invisible information becomes visible, improving the work efficiency of station staff and ensuring the safety and security of users and passengers.

In addition, this invention is not limited to an above-described Example, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files for realizing each function can be stored in a recording device such as a memory, a hard disk, an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.
Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

101: Ticket gate information collection server 103: Ticket gate information collecting server 104: Distribution server, analysis server 106, 107: Management information server 109: Administrator terminal 110: Equipment control 111: Ticket gate information

Claims (5)

In the congestion status information collection and distribution system,
A ticket gate equipped with a door and a ticket reading device, for obtaining ticket gate information including ticket gate passage time, boarding station ID, ticket gate ID, commuter pass information from the ticket;
A control and monitoring server connected to the ticket gate;
A ticket gate information collecting server connected to the control monitoring server and collecting information on the ticket gate from the control monitoring server;
An analysis server connected to the ticket gate information collecting server, analyzing ticket gate information, calculating congestion status information, and storing it in a congestion status information table;
A management information server connected to the analysis server;
A distribution server connected to the analysis server and distributing with reference to the congestion status information;
The analysis server
Aggregate the number of people using the ticket gate ID as a key from the ticket gate information, analyze the congestion status of each ticket gate, determine the planned exit station from the ticket gate information,
Calculate the estimated residence time at the boarding station and the scheduled exit station,
Calculate the total scheduled residence time, analyze the congestion situation for each station and each time zone,
A congestion status information collection and distribution system that analyzes a time zone exceeding a stay limit by comparing the congestion state and a stay limit number of persons for each station registered in advance. In the congestion status information collection and delivery system according to claim 1,
The management information server is provided with one or more databases, and pre-registers and / or sequentially registers and manages a table of operation information, route information, ticket gate information, prediction information, current status information, performance information, and analysis information,
The control and monitoring server controls one or more of the ticket gates and receives the ticket gate information, and the ticket gate information collecting server collects and stores the ticket gate information from one or more of the control and monitoring servers. And a congestion status information distribution system for storing the ticket gate information in a ticket gate information table. In the congestion status information collection and delivery system according to claim 2,
The operation information includes the station ID, route name information, and expected operation restart time information.
The management information server includes a step of performing communication for transmitting the operation information when the operation information is registered in the management information server,
The analysis server
When receiving the operation information, update the record by adding the expected operation resumption time to the arrival time and the scheduled residence time stored in the prediction information table using the station ID and / or the route name information as a key,
A congestion status information collection and distribution system that recounts congestion status information. In the congestion status information collection and delivery system according to claim 3,
The ticket gate information includes passenger entry / exit information as an attribute,
The analysis server of the previous period has time range information in memory,
Aggregate the difference between the person leaving the ticket gate and the person entering the ticket within the range indicated by the time range information,
A congestion status information collection and distribution system that counts the number of people staying from congestion status information using the ticket gate ID as a key. In the congestion status information collection and delivery system according to claim 3,
The analysis server includes a step of transmitting congestion status information including a ticket gate ID to a distribution server when a caution flag is set,
The distribution server includes a step of transmitting an opening / closing control signal of a corresponding ticket gate to the control monitoring server when the congestion status information is received,
The congestion gate information collection and distribution system, wherein the ticket gate includes a step of operating a door when the door control signal is received.

Images