Passenger flow forecasting device, passenger flow forecasting method, and passenger flow forecasting program
The passenger flow forecasting device improves prediction accuracy by using real-time passage data from ticket gates to update OD data and estimate congestion levels, addressing the limitations of conventional methods in predicting future passenger flows.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- OMRON CORP
- Filing Date
- 2024-08-29
- Publication Date
- 2026-06-30
Smart Images
Figure 0007882294000001 
Figure 0007882294000002 
Figure 0007882294000003
Abstract
Description
Technical Field
[0001] The present invention relates to a passenger flow prediction device, a passenger flow prediction method, and a passenger flow prediction program for predicting the flow of passengers in railway transportation.
Background Art
[0002] Conventionally, several methods for predicting the flow of passengers in railway transportation have been proposed. According to the method disclosed in Patent Document 1, based on the passing data obtained from automatic ticket gates, the flow of passengers up to the current time on the day is measured. Next, the past data most similar to the measurement result is extracted from past OD (Origin-Destination) data. Next, the extracted past OD data itself or a finely adjusted version thereof is set as the predicted value of the flow of passengers after the current time on the day.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] According to the above method, past OD data is directly or strongly reflected in the predicted value. However, even if the flow of passengers up to the current time coincides with past data, the flow of passengers after the current time does not necessarily coincide with the past data. Also, although both the passing data and the OD data indicate the flow of passengers in the sense of showing the entry / exit location and time, they do not directly indicate the flow of passengers during the period between entry and exit.
[0005] Therefore, it is considered that there is room for improvement in the prediction results (for example, prediction accuracy and prediction content) of the conventional methods for predicting the flow of passengers in railway transportation. Therefore, the present invention aims to improve the prediction results of passenger flow in rail transport. [Means for solving the problem]
[0006] A passenger flow forecasting device according to one embodiment of the present invention predicts the flow of passengers in rail transport. The passenger flow forecasting device comprises a passage data acquisition unit, an OD forecasting unit, and a congestion level estimation unit. The passage data acquisition unit sequentially acquires passage data for the day, indicating the entry and exit stations and entry and exit times for each passenger. The OD forecasting unit updates the OD forecasting data, which shows the prediction results of the OD data for the day, based on the acquired passage data for the day. The congestion level estimation unit estimates the congestion level of each train by allocating the number of entering passengers indicated by the OD forecasting data as the number of crew members for the train on the day, based on the OD forecasting data and the train operation plan data for the day.
[0007] "Passage data" shows the entry station, entry time, exit station, and exit time for each passenger. This passage data is acquired when each passenger passes through the ticket gate. "OD data" can be generated by aggregating transit data. OD data shows the number of passengers for each combination of entry station (origin) and exit station (destination) among multiple stations within the data acquisition range (for example, a railway line operated by a railway operator that has introduced the passenger flow forecasting device). If the number of stations within the data acquisition range is n, there are n × (n-1) combinations of entry and exit stations.
[0008] According to the above configuration, the OD prediction unit updates the OD prediction data, which shows the prediction result of the OD data, in real time using the traffic data for the day acquired by the traffic data acquisition unit. Because the traffic data for the day is used, the OD data can be predicted in real time, improving the prediction accuracy. The congestion estimation unit then estimates the congestion level of the train based on this real-time, continuously updated OD prediction data. This allows for real-time estimation of congestion levels, improving estimation accuracy. Congestion levels are an indicator of passenger flow during the period between entry and exit, which is difficult to show using only transit data or OD data. Therefore, it is possible to expand the content of passenger flow predictions in rail transport; in other words, the prediction results can be improved.
[0009] The passage data acquisition unit may acquire passage data from ticket gates installed at each station. According to the above configuration, passage data is obtained from the ticket gates. The entry and exit stations can be easily linked from the ticket used to pass through the gates. Therefore, highly reliable passage data can be generated, and based on this, highly reliable OD prediction data can be generated. Since the degree of congestion is estimated based on this OD prediction data, the estimation accuracy is improved.
[0010] Generally, ticket gates are railway equipment owned by railway operators. When the passenger flow prediction device is introduced to a railway operator, the railway operator can predict OD data and estimate congestion levels by applying the railway equipment it owns. Congestion levels can be estimated with reduced or no reliance on data provided from external sources. Furthermore, congestion levels can be estimated even if the train itself is not equipped with hardware for estimating congestion (e.g., weight sensors). Therefore, railway operators can obtain congestion estimation results at low cost.
[0011] The OD prediction unit may be configured to continuously update multiple OD prediction data sets, each showing the predicted OD data for the day broken down by time period. The congestion level estimation unit may be configured to estimate the congestion level based on the multiple OD prediction data sets. According to the above configuration, multiple OD (Destination) prediction data sets are generated for each time period. For example, the number of passengers from 15:00 to 15:05 is XX people, and the number of passengers from 15:05 to 15:10 is YY people, and so on, with OD data predicted by subdividing the time. This makes it possible to instantly and easily determine the number of passengers that occurred within the target time period of each OD prediction data based on the day's transit data. Therefore, it is possible to instantly and easily measure how many passengers traveled to which destinations within the target time period, improving the accuracy of OD data prediction. In addition, because congestion is estimated based on OD prediction data generated by subdividing the time, it becomes easier to identify the trains that passengers that occurred within each target time period are expected to be on, improving the accuracy and timeliness of congestion estimation.
[0012] The OD prediction unit may include an initialization unit that performs an initialization process to set initial values for OD prediction data based on past OD data, and an update unit that performs an update process to update the OD prediction data by correcting the initial values based on the day's transit data. According to the above configuration, first, the initialization unit sets initial values for the OD prediction data based on past OD data, and then the update unit corrects the initial values and updates the OD prediction data. Compared to cases where the initial values are set arbitrarily, the prediction accuracy of the OD prediction data is improved. Since the update of the OD prediction data is based on the day's transit data, the prediction accuracy is improved.
[0013] The initialization unit may extract data from past OD data that is estimated to be similar to the OD prediction data, and set the extracted past OD data as the initial value. According to the above configuration, when setting initial values based on past OD data, the initialization unit extracts past OD data that is estimated to be similar to the OD prediction data. Since similar data is set as the initial values, the prediction accuracy of the OD prediction data is improved.
[0014] In a configuration where the OD prediction unit is configured to update multiple OD prediction data sets, each showing the prediction results of the OD data for the day by time period, the update unit may determine the number of passengers entering each station during the time period covered by the OD prediction data based on the transit data acquired at the end of the time period covered by the OD prediction data. According to the above configuration, the OD prediction unit generates OD prediction data for each time period, and determines the number of passengers entering each station during the target time period of the OD prediction data based on the passage data acquired at the end of the target time period. Since the number of passengers entering is determined, the prediction accuracy is improved.
[0015] The update unit may update the OD prediction data by allocating the confirmed number of passengers entering each station to the number of passengers exiting at other stations, using the same ratio as the initial values. According to the above configuration, initial values generated based on past OD data are updated based on the current day's transit data. This enables real-time updating of OD prediction data, improving prediction accuracy.
[0016] The update unit may update the OD prediction data by partially determining the number of passengers who entered during the target time period and have already exited, based on transit data acquired at a point after the end of the target time period. According to the above configuration, the number of passengers who have already departed can be partially determined based on transit data acquired after the end of the target time period, thus improving prediction accuracy.
[0017] The update unit may determine the number of passengers who entered the station within the target time period of the OD prediction data for combinations where the longest travel time is shorter than the period from the start of the target time period to that point, based on the passage data acquired at a point after the end of the target time period and the longest travel time data which defines the longest travel time expected to be required to travel between two stations for each combination of entry station and exit station.
[0018] According to the above configuration, OD data is predicted using the concept of the longest travel time. The longest travel time is the time expected to be required for moving between two stations, defined for each combination of any two stations of the entrance station and the exit station. For example, if the longest travel time from Station A to Station B is 10 minutes, it is estimated that all passengers who enter Station A and attempt to move to Station B within the target time period have exited Station B within 10 minutes from the start of the target time period. When the passing data is reacquired at a time point 10 minutes or more after the start of the target time period, the number of passengers in the combination with Station A as the entrance station and Station B as the exit station is determined based on this passing data. In this way, the number of passengers who have made relatively short-distance movements is determined as those who have completed exiting, and the prediction accuracy is improved.
[0019] The update unit may update the OD prediction data by estimating the number of passengers who have not exited by subtracting the number of passengers moving between two stations that have been determined from the number of passengers entering that have been determined, and allocating the estimated number of passengers who have not exited as the number of passengers exiting between two stations in a combination where the longest travel time is longer from the start to that point. According to the above configuration, the number of passengers who have not exited is estimated from the number of passengers entering that have been determined and the number of passengers who have already exited. This number of passengers who have not exited is allocated as the number of passengers making long-distance movements. After partially determining the number of passengers who have already exited, it is possible to predict where and how many of the passengers who have not exited will exit, and the prediction accuracy is improved.
[0020] The congestion degree estimation unit may have a train allocation unit that searches for trains used by passengers entering each station within the target time period of the OD prediction data to move to the exit station based on predetermined search conditions in the operation plan data, and executes a train allocation process of allocating the passengers to the searched trains as the number of passengers. According to the above configuration, by searching for trains used by passengers based on the predicted value of the OD data, the number of passengers on the train is estimated. The congestion degree of the train can be estimated based on the operation plan data and the OD prediction data. Therefore, it is possible to reduce the dependence on external data and estimate the congestion degree at low cost.
[0021] The predetermined search conditions may include two or more conditions. The train allocation unit allocates the passengers entering each station to a predetermined ratio according to the number of conditions of the search conditions, searches for the trains used by the allocated passengers to move to the departure station based on the corresponding conditions, and may allocate the passengers to the searched trains as the number of passengers. In addition, when moving between two stations, passengers may select a route according to their preferences, such as selecting a route with an earlier arrival time or a route with fewer transfer times. The two or more search conditions correspond to such preferences, and include conditions for preferentially extracting a route that can reach the destination at the fastest speed and conditions for preferentially extracting a route with the fewest transfer times.
[0022] According to the above configuration, passengers are allocated at a predetermined ratio according to the number of conditions. The trains that will be used by the passengers allocated for each condition are searched according to the condition. Then, the passengers are allocated as the number of passengers of the searched trains. Thereby, even if there is a possibility that the trains boarded by passengers differ depending on the preferences of passengers moving between the same two stations, it is possible to cope with this. Therefore, the estimation accuracy of the congestion level of the train is improved.
[0023] A passenger flow prediction method according to an aspect of the present invention is a passenger flow prediction method for predicting the flow of passengers in railway transportation, and includes a passing data acquisition step, an OD prediction step, and a congestion level prediction step. In the passing data acquisition step, the passing data of the day indicating the entry and exit stations and the entry and exit times of each passenger is sequentially acquired. In the OD prediction step, based on the acquired passing data of the day, the OD prediction data indicating the prediction result of the OD data of the day is updated at any time. In the congestion level estimation step, based on the OD prediction data and the train operation plan of the day, the trains boarded by passengers are estimated, the number of passengers entering shown by the OD prediction data is allocated as the number of passengers of the trains in operation, and the congestion level of each train is estimated.
[0024] A passenger flow prediction program according to an aspect of the present invention causes a computer to execute the above passenger flow prediction method. The above method and program possess the same technical features as the passenger flow forecasting device described above and produce similar effects. [Effects of the Invention]
[0025] According to the present invention, it is possible to improve the prediction results of passenger flow in rail transport. [Brief explanation of the drawing]
[0026] [Figure 1] This is a block diagram showing a passenger flow prediction device according to an embodiment of the present invention. [Figure 2] This figure shows an example of an application of an embodiment of the present invention. [Figure 3] This diagram shows an example of transit data in a table format. [Figure 4] (A) is a diagram showing an example of OD data in matrix form. (B) is a diagram showing an example of OD data by time period in matrix form. [Figure 5] This is a block diagram for passenger flow forecasting systems. [Figure 6] This is a time chart of the processes performed by the OD prediction unit and the congestion level estimation unit. [Figure 7] This diagram shows the initialization process and the first update process. [Figure 8] This is a diagram showing the second update process. [Figure 9] This is a diagram showing the third update process. [Figure 10] This diagram shows the train allocation process. [Figure 11] This is a flowchart showing a passenger flow prediction method according to an embodiment of the present invention. [Modes for carrying out the invention]
[0027] Embodiments of the present invention will be described below with reference to the drawings. (Passenger flow forecasting device) Figure 1 shows a passenger flow forecasting device 1 according to one embodiment of the present invention. The passenger flow forecasting device 1 forecasts the flow of passengers in rail transport. The passenger flow forecasting device 1 can be used, for example, in the passenger transport operations of a railway operator, and is also suitably applied to information provision services for passengers.
[0028] The passenger flow forecasting device 1 comprises a central processing unit 11, a storage unit 12, an input unit 13, and an output unit 14, which are interconnected via a communication bus 15. The central processing unit 11 executes a passenger flow forecasting method according to the passenger flow forecasting program P stored in the storage unit 12. The central processing unit 11 or the passenger flow forecasting device 1 equipped therewith is an example of a computer that executes a passenger flow forecasting method.
[0029] The storage unit 12 temporarily or permanently stores the passenger flow forecasting program P, as well as data necessary for executing the passenger flow forecasting method. In this embodiment, the storage unit 12 includes a main storage unit 12a and an auxiliary storage unit 12b. The main memory unit 12a is implemented by a storage device such as ROM (Read-Only Memory), RAM (Random Access Memory), or EEPROM (Electrically Erasable Programmable Read-Only Memory). The main memory unit 12a stores the passenger flow forecasting program P. The main memory unit 12a also temporarily stores the OD forecasting data D20, which will be described later.
[0030] The auxiliary storage unit 12b is a large storage device, such as a hard disk drive. The auxiliary storage unit 12b may also be implemented by a cloud server separated from the communication bus 15. The auxiliary storage unit 12b stores various types of data that are referenced when predicting passenger flow. These types of data include historical OD data D1, longest travel time data D2, operation data D3, holiday data D4, station data D5, route data D6, and path data D7.
[0031] The input unit 13 collects transit data D10 (see Figure 3) indicating the movement of each passenger from the transit data collection server 2 via the communication network 9. The central processing unit 11 predicts passenger flow based on the collected transit data and various data stored in the storage unit 12. The output unit 14 outputs information indicating the passenger flow prediction results to the administrator terminal 3, signage 4, and user terminal 5 via the communication network 9.
[0032] The administrator terminal 3 is an information terminal managed by an employee of the railway operator where the passenger flow forecasting device 1 is installed, and is equipped with a display device that shows the information output from the output unit 14. The signage 4 is an electronic sign installed at each station, and is equipped with a display unit that shows the information output from the output unit 14. The user terminal 5 is an information terminal carried by passengers, such as a smartphone or tablet. If a dedicated application 5A is installed on the user terminal 5, passengers can check information showing the passenger flow forecast results on the user terminal 5. The provider of this application 5A is, for example, the railway operator that installed the passenger flow forecasting device 1.
[0033] (Transit data, OD data) As shown in Figure 2, a railway operator owns a complete set of railway facilities, including tracks and multiple stations, and operates trains on the tracks to transport passengers. The complete set of railway facilities includes ticket gates 6 installed at each station. In Figure 2, as a mere example, a railway operator operates trains on two lines: the P line from stations A to F, and the Q line which branches off from the P line at station D and goes to stations G and H. A "train" refers to a train consisting of a single railway car or a train formation composed of two or more railway cars coupled together.
[0034] Each passenger prepares the ticket required to pass through ticket gate 6, passes through ticket gate 6 to enter the departure station (hereinafter also referred to as the "entrance station"), travels by train operated by the railway company, and passes through ticket gate 6 to exit the destination station (hereinafter also referred to as the "exit station"). Hereafter, the entire sequence of travel for each passenger from entry at the departure station to exit at the destination station will be referred to as a "trip". "Train tickets" include magnetic tickets and contactless IC (integrated circuit) cards, and each ticket records various information such as a ticket ID (Identity Document) to identify the ticket. "Train tickets" include regular tickets, commuter passes, and fare adjustment tickets.
[0035] The ticket gate 6 reads the information recorded on the ticket. If there is no entry record on the ticket and the passenger passes through the ticket gate 6 in the direction of moving from outside the station to inside the station, the ticket gate 6 records an entry on the ticket, treating it as if the passenger is entering. If there is an entry record on the ticket and the passenger passes through the ticket gate 6 in the direction of moving from inside the station to outside the station, the ticket gate 6 records an exit on the ticket, treating it as if the passenger is exiting (or collects the ticket).
[0036] The ticket gate 6 transmits information about the passenger's entry and exit, along with information about the ticket, to the pass-through data collection server 2. When a passenger passes through the ticket gate 6 to enter, the ticket gate 6 transmits at least the entry station information and entry time information. When a passenger passes through the ticket gate 6 to exit, the ticket gate 6 transmits at least the exit station information and exit time information. In addition, the ticket gate 6 at the exit station transmits the entry record (entry station information and entry time information) read from the ticket, along with the exit station information and exit time information.
[0037] As shown in Figure 3, the passage data collection server 2 generates passage data D10 representing one trip, which consists of entry and exit as one set, based on information transmitted from ticket gates 6 at numerous stations. The passage data D10 includes entry station information, entry time information, exit station information, and exit time information. The entry station information indicates the entry station where the ticket gate 6 that read the ticket information is installed. The entry time information indicates the time when the ticket gate 6 was passed through at the entry station. The exit station information indicates the exit station where the ticket gate 6 that read the ticket information is installed. The exit time information indicates the time when the ticket gate 6 was passed through at the exit station.
[0038] While a passenger is in transit, there is no information about the departure station or departure time (see the passage data D10 on the second line). Therefore, the passage data collection server 2 stores the incomplete passage data D10 until the passenger departs. The passage data collection server 2 can also transmit the incomplete passage data D10 to the passenger flow forecasting device 1 upon request from the passenger flow forecasting device 1. Once the day's train operations are complete, all passengers depart from their destination stations, and no passengers enter their departure stations until the first train of the following day. Thus, all trips for the day are finalized. Based on the finalized group of transit data D10, the transit data server 2 generates the OD data D1 for this day (see Figure 4(A)).
[0039] As shown in Figure 4(A), the OD data D1 shows the number of passengers (i.e., number of trips) for each combination of entry station and exit station among multiple stations within the data measurement range (in the example in Figure 2, stations A to H on lines P and Q operated by a single railway operator). The OD data D1 is represented, for example, in a matrix with entry stations in rows and exit stations in columns. Referring to Figures 2 and 4(A), the symbols "a~h" correspond to stations A~H. Although not shown, the symbols "i" and "j" used in this book represent any two different stations and can be appropriately replaced with the symbols "a~h". The symbol "Tij" represents the total number of passengers who entered at station i and exited at station j. In other words, it is the number of trips where station i is the entry station and station j is the exit station. Further in other words, it is the number of passage data D10 sets among the aggregated passage data D10 sets where the entry station information is station i and the exit station information is station j. The symbol "Gi" represents the total number of passengers entering at station i, and the symbol "Ai" represents the total number of passengers exiting at station i. Ga is the sum of Tab, Tac, ..., Tag, Tah, and Aa is the sum of Tba, Tca, ..., Tga, Tha. The symbol "Σ" represents the total number of trips (the number of aggregated transit data D10), which is the sum of Ga, Gb, ..., Gg, Gh, and also the sum of Aa, Ab, ..., Ag, Ah.
[0040] The collection period for transit data D10 to generate OD data D1 is not limited to a daily period, but may be a predetermined time unit shorter than one day (for example, an hourly period, a 10-minute period, etc.). In this case, as shown in Figure 4(B), multiple "time-specific OD data D1A" are generated, each with a different collection period for transit data D10. Hereinafter, the collection period for transit data D10 corresponding to each time-specific OD data D1A will be referred to as the "target time period T1".
[0041] Figure 4(B) shows a simplified example of time-based OD data D1A for the same day, where the collection period is set to 5-minute intervals. Multiple time-based OD data D1A examples are shown, with target time period T1 being 14:50-14:55, the next target time period T1 being 14:55-15:00, and the next target time period T1 being 15:00-15:05. If the collection period is short, such as in 5-minute increments, the duration of a trip may be longer than the collection period. Time-based OD data D1A includes all trips where the entry time falls within the target time period T1, regardless of whether the exit time falls within the target time period T1. In time-based OD data D1A, the symbol "Σ" indicates the total number of trips where the entry time falls within the target time period T1, and the symbol "Gi" indicates the number of passengers entering at station i during the target time period T1.
[0042] The transit data collection server 2 generates OD data D1 after the last train of the day and before the first train of the next day. The input unit 13 (see Figure 1) acquires the OD data D1 from the transit data collection server 2 before the first train of the day, and the acquired OD data D1 is stored in the storage unit 12 (see Figure 1) as past OD data D1. The passenger flow forecasting device 1 may also store time-specific OD data D1A as past OD data D1. Time-specific OD data D1A may be generated by the transit data collection server 2 or by the passenger flow forecasting device 1.
[0043] Figure 5 is a block diagram of the passenger flow forecasting device 1. The passenger flow forecasting device 1 includes a passage data acquisition unit 21, an OD forecasting unit 22, and a congestion level estimation unit 23, which are activated by executing the passenger flow forecasting program P. The passage data acquisition unit 21 sequentially acquires the day's passage data D10 from the passage data collection server 2 at a predetermined acquisition interval. The passage data D10 is acquired by the ticket gates 6 installed at each station. The passage data D10 shows the entry and exit stations and entry and exit times for each passenger (see Figure 1(B)). However, the acquired passage data D10 may include incomplete data that only shows the entry station and entry time because the passenger has not yet exited. The acquisition interval for passage data D10 is not particularly limited. For example, the acquisition interval may be set to the same as the collection period for passage data D10 at the passage data collection server 2 (for example, 5 minutes).
[0044] (OD prediction) The OD prediction unit 22 updates the OD prediction data D20, which shows the prediction result of the OD data for the day, based on the acquired transit data D10 for the day. In particular, the OD prediction unit 22 generates multiple time-segment OD prediction data D21. The length of the target time period T1 for each time-segment OD prediction data D21 is not particularly limited. For example, the length of the target time period T1 is set to be the same as the acquisition cycle of transit data D10 in the transit data acquisition unit 21 (for example, 5 minutes).
[0045] The OD prediction unit 22 includes an initialization unit 31 and an update unit 32. The initialization unit 31 performs an initialization process to set the initial value of the OD prediction data D20 based on past OD data D1 stored in the storage unit 12. The update unit 32 corrects the initial value of the OD prediction data D20 based on the day's transit data D10, which is acquired as it progresses, and performs an update process to update the OD prediction data D20 as it progresses. In other words, the update unit 32 updates the OD prediction data D20 multiple times based on the latest version of the transit data D10, which is acquired intermittently as time progresses.
[0046] In this embodiment, updates are performed three times. The update process includes a first update process, a second update process, and a third update process, and the update unit 32 comprises a first update unit 36, a second update unit 37, and a third update unit 38. The first update unit 36 executes the first update process on the OD prediction data after the initialization process. The second update unit 37 executes the second update process on the OD prediction data after the first update process. The third update unit 38 executes the third update process on the OD prediction data after the second update process.
[0047] Referring also to Figure 6, the first update unit 36 executes the first update process at the end time t1 of the target time period T1 of the OD prediction data D20. The second update unit 37 and the third update unit 38 execute the second update process and the third update process, respectively, at the second update time t2 and the third update time t3, which are after the end time (first update time) t1 of the target time period T1. The second update time t2 and the third update time t3 may be set in any way. In this embodiment, as just one example, the second update time t2 is set to a point in time that is the length of the target time period T1 after the end time t1 of the target time period T1 (for example, 10 minutes after the start time t0 and 5 minutes after the end time t1). The third update time t3 is set to a point in time that is the same length after the second update time t2 (15 minutes after the start time t0, 10 minutes after the end time t1 and 5 minutes after the second update time t2).
[0048] Furthermore, the target time zones T1 of multiple time-based OD prediction data D21 are set to be continuous without interruption. The end time t1 of the target time zone T1 of one time-based OD prediction data D21 is the same as the start time t0 of the target time zone T1 of the next time-based OD prediction data D21. The second update time t2 of one time-based OD prediction data D21 is the same as the start time t0 of the target time zone T1 of the second-to-last time-based OD prediction data D21. Therefore, multiple time-based OD prediction data D21 are generated simultaneously.
[0049] (initialization) Figure 7 shows the initialization process and the first update process. In Figure 7 and subsequent diagrams, for the sake of simplicity, only stations A, D, F, and H are shown as entry and exit stations, but the number of passengers traveling between other two stations is predicted in the same way. Note that A and F are terminal stations of the P line. D is an intermediate station on the P line and a terminal station on the Q line, and is a transfer station between the P line and the Q line. H is the terminal station on the opposite side of the Q line.
[0050] During the initialization process, the initialization unit 31 extracts data from past OD data D1 that is estimated to be similar to the time-of-day OD prediction data D21 that will be generated this time. The initialization unit 31 sets the extracted data as initial values. This generates the initialized time-of-day OD prediction data (initial data D21(t0)). The extraction method, the number of past OD data D1 to extract, and whether or not the extracted data needs to be processed are not particularly limited. In the illustrated example, the numerical value of the past time-specific OD data D1 for the same day of the week and time of day one week prior is set as the initial value. On routes with high commuter demand, the same day of the week makes it easier to extract similar data. The initialization unit 31 may also extract past time-specific OD data for the same time of day yesterday. On days when events with high drawing power (e.g., sports matches, concerts, festivals, etc.) are held along the route, the initialization unit 31 extracts past time-specific OD data for the same time of day on days when similar events were held. In addition to setting the numerical value of a single past time-specific OD data as the initial value, the initialization unit 31 may also set the average value of multiple time-specific OD data as the initial value. The initialization process can be executed at any time as long as it is completed by the end time t1 of the target time period T1. For example, the initialization process is executed at the start time t0 of the target time period T1.
[0051] (1st update) The first update process is executed at the end of the target time period T1, time t1. The first update unit 36 aggregates the passage data D10(t1) acquired at the end of the target time period T1, specifically the passage data D10 whose entry time falls within the target time period T1. Based on this, the first update unit 36 determines the number of passengers Gi entering each station during the target time period T1. Note that some passengers may have already exited during the target time period T1, but the first update unit 36 does not require exit station information or exit time information from the passage data D10(t1) in the first update process.
[0052] The first update unit 36 allocates the confirmed number of passengers entering each station, Gi, as the number of passengers exiting other stations, using the same ratio as the initial values. This corrects the initial values and generates the time-specific OD prediction data (first update data D21(t1)) that has undergone the first update process. If we let Gi(t0) be the number of passengers entering station i in the initial data D21(t0), and Tij(t0) be the number of passengers entering station i and exiting station j in the initial data D21(t0), then the number of passengers entering station i and exiting station j, Tij(t1), is predicted based on the formula: Tij(t1) = Gi × (Tij(t0) / Gi(t0)). This predicted value Tij(t1) is then used to replace the initial value Tij(t0).
[0053] Specifically, in the first updated data D21(t1), the number of passengers Tad(t1) entering at station A and exiting at station D is predicted to be 10. This is derived by multiplying the confirmed number of passengers entering at station A, Ga(80), by the ratio (11 / (11+33+44)) of the number of passengers entering at station A, Tad(t0), in the initial data D21(t0), to the number of passengers entering at station A, Ga(t0).
[0054] In other words, in the initial data D21(t0), 88 passengers who entered at Station A exited at other stations (Station D, Station F, Station H) in a ratio of 1:3:4. When generating the first updated data D21(t1), the confirmed number of passengers entering Station A, Ga (80 people), is allocated to the other stations (Station D, Station F, Station H) as the number of passengers exiting (10 people, 30 people, 40 people), in the same ratio as the initial values of 1:3:4.
[0055] (2nd update) Figure 8 shows the second update process. The second update process is executed at the second update time t2, which is after the end time t1 of the target time period T1. At the second update time t2, some of the passengers who entered during the target time period T1 begin to exit. In the second update process, the exit station information and exit time information are referenced from the transit data D10 whose entry time falls within the target time period T1.
[0056] In the second update process, the second update unit 37 extracts from the transit data D10(t2) acquired at the second update time t2 the passenger numbers for travel between two stations that are greater than the corresponding passenger numbers in the first update data D21(t1). Then, the numerical value in the first update data D21(t1) is corrected by increasing it to the extracted passenger numbers. Next, the second update unit 37 offsets the increase correction amount by decreasing the number of passengers in other combinations where the entry station is the same but the exit station is different, thereby maintaining the number of passengers entering the entry station at the confirmed number. The distribution ratio of the decrease correction amount is determined by the ratio of the number of passengers in other combinations in the first update data D21(t1). In this way, the time-specific OD prediction data (second update data D21(t2)) that has undergone the second update process is generated.
[0057] Specifically, at the second update time t2, the number of passengers Tfd(t2) who entered at station F and exited at station D is known to be 12 from the transit data D10(t2). This is more than the corresponding number of passengers Tfd(t1) of 10 from the first update data D21(t1). Therefore, the second update unit 37 replaces the number of passengers Tfd(t1) related to the movement between the two stations with the number of passengers Tfd(t2) determined from the transit data D10(t2) acquired at the second update time t2. In this way, an increase correction is made.
[0058] The increase correction amount ΔTfd(Tfd(t2)-Tfd(t1)) is 2 people. If the increase correction were to be applied individually, the number of passengers entering station F, Gf, would increase to 47, and the confirmed number of 45 would not be maintained. Therefore, the number of passengers Tfa(t1) and Tfh(t1) for other combinations, using the same station F as the entry station but with a different exit station, is reduced. The decrease adjustment applies to the passenger numbers Tfa(t1) and Tfh(t1) for all combinations other than those for which an increase adjustment was applied, assuming the entry station is the same. The amount of the decrease adjustment is equal to the amount of the increase adjustment ΔTfd, thereby maintaining the number of entering passengers Gf. The ratio of the amount of the decrease adjustment is the ratio of the passenger numbers Tfa(t1) and Tfh(t1) for the combinations for which the decrease adjustment was applied, which in this example is 3:2.
[0059] The number of passengers Tfa(t1) departing from station A will be subject to a 60% reduction adjustment, decreasing from 21 to 1.2 passengers. The number of passengers Tfh(t1) departing from station H will be subject to a 40% reduction adjustment, decreasing from 14 to 0.8 passengers. The ratio of passenger numbers Tfa and Tfh for the combinations subject to the reduction adjustment will remain at 3:2, the same as before the adjustment. Note that while the unit of passenger numbers is people, the predicted values of passenger numbers do not need to be integers.
[0060] (3rd update) Figure 9 shows the third update process. The third update process is executed at the third update time t3, which is after the end time t1 of the target time period T1 and the second update time t2. By the third update time t3, it is assumed that all passengers who entered during the target time period T1 and whose destinations were close to the entry station have already exited. Therefore, the third update unit 38 determines the number of passengers traveling between two stations in combinations where the entry station and exit station are close to each other.
[0061] The third update unit 38 refers to the longest travel time data D2 stored in the memory unit 12 in order to determine the number of passengers. The longest travel time data D2 defines the correspondence between combinations of entry and exit stations and the longest expected travel time required to travel from the entry station to the exit station. The longest travel time for each combination is a value obtained empirically by referring to the passage data D10 accumulated so far, and can be predetermined during the design and manufacturing stages of the passenger flow prediction device 1.
[0062] In the illustrated example, the cell where the row for Station A and the column for Station D intersect is labeled "8". That is, in the longest travel time data D2, the combination of Station A as the entry station and Station D as the exit station is associated with a longest travel time of 8 minutes. It is also possible that the longest time may change if the entry and exit stations are swapped. This embodiment also handles this case (see "11" in the cell where the row for Station D and the column for Station A intersect). In this embodiment, when the number of stations within the data measurement range is n, n × (n-1) correspondence relationships are defined by the longest travel time data D2.
[0063] The third update unit 38 determines the number of passengers for two-station travel where the longest travel time is shorter than the period T3 from the start time t0 of the target time period T1 to the third update time t3, based on the passage data D10(t3) acquired at the third update time t3. The closer the exit station is to the entry station, the shorter the longest travel time. By executing this process, the number of passengers for combinations where the exit station is close to the entry station is determined.
[0064] Then, the third update unit 38 estimates the number of passengers who have not yet exited by subtracting the number of passengers in the confirmed combination from the number of passengers who have entered for the confirmed combinations of combinations where the entry station is the same. The estimated number of passengers who have not yet exited is then allocated as the number of passengers in the unconfirmed combinations. This generates the OD prediction data that has undergone the third update process (hereinafter referred to as the third update data D21(t3)).
[0065] Specifically, if a passenger enters at Station A, the longest travel time to exit at Station D (8 minutes) and the longest travel time to exit at Station F (10 minutes) are both shorter than the total travel time during the above period T3 (15 minutes). From the transit data D10 (t3) obtained at the third update time t3, it was determined that the number of passengers Tad who entered at Station A and exited at Station D was 9. These 9 people are determined to be the number of passengers Tad (t3) for the combination of Station A as the entry station and Station D as the exit station. Similarly, 25 people are determined to be the number of passengers Taf (t3) for the combination of Station A as the entry station and Station F as the exit station.
[0066] On the other hand, the longest travel time (16 minutes) when H station is the exit station is longer than the above period T3. Therefore, the value (10 people) obtained from the transit data D10(t3) is not used to estimate the number of passengers Tah(t3) for the combination of A station as the entry station and H station as the exit station. Instead, the sum of the number of passengers Tad(t3) and Taf(t3) for combinations in which passengers entered A station and exited is confirmed is subtracted from the number of passengers Ga entering A station. This estimates the number of passengers who entered A station but did not exit (80-(9+25)=46).
[0067] In this example, there is only one combination remaining where the longest travel time is longer than period T3 and departure cannot be confirmed. Therefore, the estimated number of passengers who have not yet departed (46 people) is directly allocated as the number of passengers Tah(t3) for that combination. If there are two or more combinations where the longest travel time is longer than period T3, the estimated number of passengers who have not yet departed may be allocated as the number of passengers for the two or more remaining undetermined combinations in proportion to the number of passengers in the second update data D21(t2). Furthermore, the determination of the number of passengers using this longest travel time may be performed in the second update process at the second update time t2.
[0068] (Congestion level estimation) Returning to Figures 5 and 6, the congestion estimation unit 23 includes a train allocation unit 41 that performs train allocation processing and a congestion derivation unit 42 that performs congestion derivation processing.
[0069] In the train allocation process, the train allocation unit 41 allocates the number of incoming passengers indicated by the OD prediction data D20, which is generated by the OD prediction unit 22, and the train operation plan data for the day, as the number of crew members for the trains to be operated on that day. In the congestion derivation process, the congestion degree derivation unit 42 estimates the congestion degree of the trains based on the number of train crew members estimated by the train allocation unit 41.
[0070] As described above, the OD prediction unit 22 generates multiple time-based OD prediction data D21 simultaneously. The timing (estimation time) at which the congestion level estimation unit 23 estimates the congestion level is the same as the start time t0 of the target time period of a given time-based OD prediction data D21, the same as the end time t1 (first update time) of the target time period of the previous time-based OD prediction data D21, the same as the second update time t2 of the time-based OD prediction data D21 before that, and the same as the third update time t3 of the time-based OD prediction data D21 before that.
[0071] According to the OD prediction method described above, at each estimation time, there exists at least initial data D21(t0), first update data D21(t1) and second update data D21(t2) which are in the process of being updated, and third update data D21(t3) which is immediately after the update is completed. In addition, the storage unit 12 may store one or more third update data D21(t3) which have been updated before the estimation time.
[0072] The congestion estimation unit 23 may, at each estimation point, perform train allocation and congestion estimation based on initial data D21(t0), where the estimation point is the start time t0 of the target time period. In this case, congestion estimation can be performed even when there are no passengers entering during the target time period T1 at the estimation point. The congestion estimation unit 23 may, at each estimation point, take into account one or more update data D21(t1) to (t3) that are in the process of being updated or have been updated, to perform train allocation and congestion estimation. In this case, congestion can be estimated with high accuracy based on OD prediction data that is updated in real time using the day's passing data D10.
[0073] (Train allocation) The operation plan data referenced by the train allocation unit 41 is stored in the memory unit 12. The operation plan data includes various information regarding the train's operation schedule and specifications, such as which trains will run on which days and at what times, the operating section and stations for each train, the stopping and departure times at each station, the number of cars that make up each train, and the transport capacity (passenger capacity) of each train.
[0074] The memory unit 12 stores operation data D3 and holiday data D4 as examples of operation plan data. The operation data D3 includes operation pattern information that shows the train's operating pattern, and train information that shows the train's specifications (for example, the train's transport capacity). The operation pattern information is like a timetable. If trains operate according to different timetables on weekdays, weekends, and holidays, the operation data D3 includes multiple operation pattern information corresponding to multiple timetables. If extra trains operate on event days when events are held along the line, the memory unit 12 includes operation pattern information corresponding to these extra trains as operation data D3.
[0075] The holiday data D4 includes calendar information to determine whether the day is a weekday, weekend, or public holiday. This calendar information also includes information indicating whether or not it is an event day. By referring to the holiday data D4, it is possible to determine which operational pattern information should be applied when the operational data D3 contains multiple operational pattern information. The train allocation unit 41, in addition to the operation plan data, refers to search data necessary to find out which train each passenger entering each station will board. The storage unit 12 stores station data D5, route data D6, and path data D7 as examples of search data.
[0076] Station data D5 contains information about the time required for transfers (in this example, Station D is the transfer station) and the time required to move from the ticket gate 6 to the platform at each station. Route data D6 contains information about the route map within the data measurement range. Path data D7 contains information about the route used when moving from a certain time, when on the platform of a certain entry station, to a certain exit station. By combining this path data D7 with past passage data D10, a path table is generated. Travel path tables are generated in advance for each station, for each time period, and for each condition such as the earliest travel route and the shortest transfer travel time, and these multiple travel path tables are stored in device 1.
[0077] Figure 8 shows the process executed by the train allocation unit 41. In the illustrated example, the train allocation unit 41 executes the train allocation process by referring to the first update data D21(t1), where the current estimated time is the end time t1 of the target time period T1. As mentioned above, in the first update data D21(t1), the number of passengers entering each station in the target time period T1 is determined using the transit data D10(t1) acquired at the end time t1 (i.e., the estimated time), and the number of passengers for each combination of entering and exiting stations is predicted based on this determined number of entering passengers and the initial value. Furthermore, in the first update data D21(t1), the number of passengers who have already exited has not been determined.
[0078] The train allocation unit 41 searches for trains that passengers entering an entry station will use to travel to an exit station based on predetermined search conditions, and allocates the passengers to the searched trains as crew members. The search conditions include two or more conditions. Examples of search conditions include the condition for reaching the exit station as quickly as possible (condition 1) and the condition for minimizing the number of transfers (condition 2). The train allocation unit 41 allocates passengers to each condition at a predetermined ratio according to the number of conditions in the search conditions.
[0079] The train allocation unit 41 further distributes the passengers allocated for each condition according to their entry time. The train allocation unit 41 sets multiple estimated entry times by dividing the target time period T1 into predetermined time units (for example, 1-minute units). Then, the passengers allocated for each condition are further distributed as passengers who entered at each of the multiple estimated entry times. As an example, the train allocation unit 41 evenly distributes the passengers allocated for each condition as passengers who entered at each estimated entry time.
[0080] The train allocation unit 41 refers to the station data D5 to estimate the time when passengers who entered at each estimated entry time will arrive at the platform. Then, the train allocation unit 41 refers to the operation data D3, route data D7, or route table to predict which train passengers who arrived at the platform at the estimated time will use to get to their departure station. The above process will be explained in detail. For example, in the first update data D21(t1), it is predicted that 30 passengers will enter station A and exit station F. The train allocation unit 41 predicts which train these 30 people will take to travel to station F.
[0081] Passengers' preferences also play a role in the trains they choose. These 30 passengers are then sorted according to multiple search criteria that correspond to their preferences. In this example, two conditions are set: Condition 1, which ensures the fastest arrival at the departure station, and Condition 2, which ensures the fewest transfers. The ratio of passengers traveling according to Condition 1 to those traveling according to Condition 2 is set to 6:4 as an example. The train allocation unit 41 estimates that of the 30 passengers, 18 will travel according to condition 1 and 12 will travel according to condition 2. Thus, the 30 passengers are allocated to the two conditions.
[0082] In this example, the target time period T1 is 15:00 to 15:05, and five estimated entry times are set to the minute. The train allocation unit 41 evenly distributes the 18 passengers assigned to condition 1 as passengers who entered at each of the five estimated entry times. That is, the train allocation unit 41 estimates that 18 / 5 passengers entered Station A by passing through the ticket gate 6 at Station A at each estimated entry time. Similarly, the train allocation unit 41 evenly distributes the 12 passengers assigned to condition 2 as passengers who entered at each of the five estimated entry times. Note that the entry time "15:00" is more precisely the time period "15:00:00 to 15:00:59", but for convenience, the time period with seconds omitted is referred to as "time". Also, the estimated number of passengers does not need to be an integer.
[0083] Here, as illustrated in the operation data D3, on Saturday, at Station A, trains bound for Station F, the terminus of Line P, and trains bound for Station H, which connects directly from Station D to Line Q, depart alternately every 5 minutes. Although detailed illustrations are omitted, trains operate on Line P, running back and forth between Station D and Station F. Passengers who board a train bound for Station H at Station A and alight at Station D can smoothly transfer to the aforementioned round-trip train. However, this operation pattern is merely a simplified example for the sake of explanation.
[0084] Furthermore, as exemplified in station data D5, it is assumed that at Station A, it takes 2 minutes to travel from ticket gate 6 to the platform. The train allocation unit 41 searches for the trains that passengers who entered Station A at each estimated entry time are likely to board in order to travel to Station F, according to the search criteria. A passenger who passes through ticket gate 6 at Station A at 15:00 will reach the platform at Station A at 15:02. If the passenger who arrives on the platform at 15:02 wants to travel to Station F as quickly as possible, it is estimated that they will board train A bound for Station F departing at 15:04. Therefore, the train allocation unit 41 estimates that the 18 out of 5 passengers who entered at 15:00 will board train A departing at 15:04. Similarly, the train allocation unit 41 estimates that the 18 out of 5 passengers who entered at 15:01 and the 18 out of 5 passengers who entered at 15:02 will also board the same train A in order to travel to Station F as quickly as possible.
[0085] Passengers who entered at 15:03 will reach the platform at 15:05, after train A has departed. Passengers arriving at the platform at 15:05 are presumed to board train B bound for station H, departing at 15:09, and transfer to a train bound for station F at station D, in order to reach station F as quickly as possible. Train allocation department 41 presumes that the 18 out of 5 passengers who entered at 15:03 will board train B departing at 15:09 and the round-trip train between stations D and F. Similarly, train allocation department 41 presumes that the 18 out of 5 passengers who entered at 15:04 will also board train B and the round-trip train in order to reach station F as quickly as possible.
[0086] On the other hand, a passenger who enters Station A at 15:00 and arrives at the platform at 15:02 is estimated to board train A bound for Station F departing at 15:04 if they wish to travel to Station F with the fewest possible transfers. The train allocation unit 41 estimates that 12 out of 5 passengers who entered at 15:00 will board train A. The train allocation unit 41 also estimates that the 12 out of 5 passengers who entered at 15:01 and the 12 out of 5 passengers who entered at 15:02 will also board train A in order to travel to Station F with the fewest possible transfers.
[0087] A passenger who enters Station A at 15:03 and arrives at the platform at 15:05 is presumed to have boarded the next train C bound for Station F at 15:14, which requires a transfer at Station D, instead of Train B departing at 15:09 bound for Station H, if they wish to travel to Station F with the fewest possible transfers. The train allocation unit 41 presumes that 12 out of 5 passengers who enter at 15:03 and travel according to condition 2 will board train C departing at 15:14. Similarly, the train allocation unit 41 presumes that 12 out of 5 passengers who enter at 15:04 will also board train C in order to travel to Station F with the fewest possible transfers.
[0088] The above train allocation process applies to the number of passengers Taf(t1) for one combination of all combinations of entry and exit stations in the first update data D21(t1). The train allocation unit 41 performs the same process for all remaining combinations in the first update data D21(t1). Then, the train allocation unit 41 refers to the initialization data D21(t0), whose estimated time is the start time t0 of the target time period T1, and performs the same processing as described above.
[0089] Furthermore, the train allocation unit 41 performs the same processing as above by referring to the second update data D21(t2), whose estimated time is the second update period t2, and by referring to the third update data D21(t3), whose estimated time is the third update period t3. In the second update data D21(t2) and the third update data D21(t3), unlike the initialization data D21(t0) and the first update data D21(t1), the number of passengers who have already departed is partially determined. Passengers who have already departed have already completed their travel by train. Therefore, when using these update data D21(t2) and D21(t3), the train allocation unit 41 estimates how many passengers boarded which trains for combinations of passengers whose departure is not yet determined.
[0090] (Derivation of congestion level) The congestion derivation unit 42 performs congestion derivation processing after the train allocation processing by the train allocation unit 41. The congestion derivation unit 42 derives an estimated congestion value for each train by dividing the estimated number of passengers for each train by the transport capacity (passenger limit) of that train. As an example, information indicating the transport capacity of a train may be included in the operation data D3 as part of the train specifications (see bottom of Figure 10).
[0091] (output) Returning to Figures 1 and 5, the output unit 14 outputs the passenger flow prediction results to the administrator terminal 3, the signage 4, and the user terminal 5. The signage 4 and user terminal 5 may output only the congestion level of each train from the prediction results. The administrator terminal 3 may output the OD prediction data D20 along with the congestion level as part of the prediction results.
[0092] (Passenger flow forecasting method) Figure 11 is a flowchart of the passenger flow prediction method according to this embodiment. The operation of each part of the passenger flow prediction device 1 is realized by the passenger flow prediction program P that executes the passenger flow prediction method. Therefore, the passenger flow prediction method will be described briefly.
[0093] The flow shown in Figure 11 is executed sequentially at the acquisition cycle of the aforementioned passage data D10 (for example, every 5 minutes). First, the passage data acquisition unit 21 executes the passage data acquisition process S1 and acquires passage data D10 from the ticket gate 6 via the communication network 9 and the passage data collection server 2. Next, the OD prediction unit 22 executes the OD prediction process S2. Then, the congestion level estimation unit 23 executes the congestion level estimation process S3.
[0094] In the OD prediction process S2, the OD prediction unit 22 executes a process to generate or update multiple time-specific OD prediction data D21, each with a different target time zone T1, at the time of this processing. In this embodiment, one time-specific OD prediction data D21 is initialized and updated three times. Therefore, different processes are executed simultaneously for the four time-specific OD prediction data D21 based on the transit data D10 acquired at the time of this processing.
[0095] The OD prediction process S2 includes an initialization process S20, a first update process S21, a second update process S22, and a third update process S23. In the initialization process S20, the initialization unit 31 initializes time-specific OD prediction data D21, with the current processing time being the start time t0 of the target time period T1. In the first update process S21, the first update unit 36 of the update unit 32 updates the time-specific OD prediction data D21(t1), which has the current processing time as the end time t1 of the target time period, based on the transit data D10 acquired at the time of the current processing.
[0096] In the second update process S22, the second update unit 37 updates the time-specific OD prediction data D21(t2), which is set to the current processing time as the second update time t2, based on the transit data D10 acquired at the current processing time. In the third update process S23, the third update unit 38 updates the time-specific OD prediction data D21(t3), which has the current processing time as the third update time t3, based on the transit data D10 acquired at the current processing time.
[0097] The congestion estimation process S3 includes the train allocation process S31 and the congestion derivation process S32. In the train allocation process S31, the train allocation unit 41 estimates how many passengers will be on which train using the time-specific OD prediction data D21(t1) for which the first update process S21 was performed at the time of this processing. In the congestion derivation process S32, the congestion derivation unit 42 estimates the congestion level of each train according to the estimated number of passengers.
[0098] (Effects / Actions) According to the passenger flow forecasting device 1 of this embodiment, the OD forecasting unit 22 updates the OD forecasting data D20, which shows the forecasting result of the OD data, in real time using the passage data D10 for the day acquired by the passage data acquisition unit 21. Because the passage data D10 for the day is used, the OD data can be predicted in real time, and the forecasting accuracy is improved.
[0099] The congestion estimation unit 23 then estimates the train's congestion level based on this real-time, continuously updated OD prediction data. This allows for real-time estimation of congestion levels, improving estimation accuracy. Congestion level is an indicator that shows the flow of passengers during the period between entry and exit, which is difficult to represent using transit data D10 or OD data D1. It can expand the content of passenger flow predictions in rail transport; in other words, it can improve the results of passenger flow predictions.
[0100] Passage data D10 is acquired from the ticket gate 6 via the passage data collection server 2. The entry station and exit station can be easily linked from the ticket used to pass through the ticket gate 6. Therefore, highly reliable passage data D10 can be generated, and based on this, highly reliable OD data can be generated. Furthermore, the estimation accuracy of the congestion level estimation result is improved based on this OD data.
[0101] The ticket gate 6 is railway equipment owned by the railway operator. When the passenger flow prediction device 1 is introduced to the railway operator, the railway operator can predict OD data D1 and estimate the degree of congestion by applying the railway equipment it owns. The degree of congestion can be estimated with low or no reliance on data provided from external sources. Furthermore, the degree of congestion can be estimated even if the train itself is not equipped with hardware for estimating the degree of congestion (e.g., weight sensors). Therefore, the railway operator can obtain congestion estimation results at low cost.
[0102] Multiple OD prediction data sets D20 are generated for each time period. By subdividing time and generating OD prediction data, the number of passengers that occurred in the target time period T1 of each time period prediction data D21 can be immediately and easily determined based on the day's transit data D10. Therefore, it is possible to immediately and easily measure how many passengers traveled to which destinations in the target time period T1, improving the accuracy of OD data prediction. In addition, since congestion is estimated based on OD prediction data generated by dividing time into small segments, it becomes easier to identify the trains that passengers that occurred in each target time period are expected to be on, improving the accuracy of congestion estimation.
[0103] Embodiments of the present invention have been described above, but the above configurations can be modified, deleted, and / or added as appropriate within the scope of the present invention. [Explanation of symbols]
[0104] 1. Passenger flow forecasting device 6 Ticket gates 21. Passage Data Acquisition Unit 22 OD prediction unit 23. Congestion level estimation unit 31 Initialization section 32 Update section 41 Train Allocation Department 42 Congestion level derivation unit D1 Past OD Data D2 Longest Travel Time Data D3 Operation Data D20 OD prediction data D21 Time-based OD forecast data S1 Passing data acquisition process S2 OD prediction process S3 Congestion level estimation process P Passenger Flow Forecasting Program
Claims
1. A passenger flow forecasting device that predicts the flow of passengers in rail transport, A passage data acquisition unit acquires passage data indicating the entry and exit stations and entry and exit times for each passenger. A congestion estimation unit estimates the degree of congestion of a train by allocating the number of passengers entering the station during a predetermined time period indicated by the passage data as the number of train crew members, based on the acquired passage data and train operation plan data. Equipped with, The congestion estimation unit includes a train allocation unit that, based on the operation plan data, searches for trains that passengers who entered each station during a predetermined time period will use to travel to their departure station, based on predetermined search conditions, and performs a train allocation process that assigns the passengers to the searched trains as crew members. The train allocation unit allocates the number of entering passengers according to the predetermined search conditions, and further allocates the number of entering passengers allocated for each predetermined search condition as passengers who entered at each of the multiple entry times obtained by dividing the predetermined time period into predetermined time units. Passenger flow forecasting device.
2. A passenger flow forecasting device for predicting the flow of passengers in rail transport, A passage data acquisition unit acquires passage data indicating the entry and exit stations and entry and exit times for each passenger. A congestion estimation unit estimates the degree of congestion of a train by allocating the number of passengers entering the station during a predetermined time period indicated by the passage data as the number of train crew members, based on the acquired passage data and train operation plan data. Equipped with, The congestion estimation unit includes a train allocation unit that, based on the operation plan data, searches for trains that passengers who enter each station during a predetermined time period will use to travel to their exit station, based on predetermined search conditions including two or more conditions, and performs a train allocation process that assigns the passengers to the searched trains as crew members. The train allocation unit allocates passengers who enter each station in a predetermined ratio according to the number of conditions in the search criteria, and assigns those passengers as crew members to the trains found based on the corresponding conditions. Passenger flow forecasting device.
3. The aforementioned predetermined search conditions include a condition for prioritizing the extraction of the fastest route to the destination, and a condition for prioritizing the extraction of the route with the fewest number of transfers. The passenger flow forecasting device according to claim 1 or 2.
4. The train allocation unit distributes the number of passengers allocated for each predetermined search condition equally among the passengers who entered at each of the multiple entry times. A passenger flow forecasting device according to claim 1, or claim 3 dependent on claim 1.
5. The aforementioned predetermined time unit is one minute. A passenger flow forecasting device according to claim 1, claim 3 dependent on claim 1, or claim 4 dependent on claim 1.
6. The aforementioned passage data acquisition unit acquires the passage data from ticket gates installed at each station. The passenger flow prediction device according to claim 5.
7. The length of the predetermined time period is set to be the same as the acquisition cycle of the passage data in the passage data acquisition unit. The passenger flow prediction device according to claim 6.
8. The congestion estimation unit estimates the train's congestion level by allocating the number of passengers entering the train during multiple time periods, which are set to be continuous and seamless, as the number of train crew members, based on the acquired passage data and the train operation plan data. The passenger flow prediction device according to claim 7.
9. The train allocation unit estimates the time at which passengers who entered at the multiple entry times will arrive at the platform, based on search data including the operation plan data as well as information on the time required to travel from the ticket gate at each station to the platform, and searches for trains that passengers who entered at each station during the predetermined time period will use to travel to the exit station. The passenger flow prediction device according to claim 1.
10. A passenger flow forecasting method for predicting the flow of passengers in rail transport, A process for acquiring passage data that obtains passage data indicating the entry and exit stations and entry and exit times for each passenger, A congestion estimation step is performed to estimate the degree of congestion of a train by allocating the number of passengers entering the train during a predetermined time period indicated by the passage data as the number of train crew members, based on the acquired passage data and train operation plan data. Equipped with, The congestion estimation step includes a train allocation step that, based on the operation plan data, searches for trains that passengers who entered each station during a predetermined time period will use to travel to their departure station, based on predetermined search conditions, and then executes a train allocation process that assigns the passengers to the searched trains as crew members. In the train allocation process, the number of entering passengers is allocated according to the predetermined search conditions, and the number of entering passengers allocated for each predetermined search condition is further allocated as passengers who entered at each of the multiple entry times obtained by dividing the predetermined time period into predetermined time units. A method for predicting passenger flow.
11. A passenger flow forecasting method for predicting the flow of passengers in rail transport, A process for acquiring passage data that obtains passage data indicating the entry and exit stations and entry and exit times for each passenger, A congestion estimation step is performed to estimate the degree of congestion of a train by allocating the number of passengers entering the train during a predetermined time period indicated by the passage data as the number of train crew members, based on the acquired passage data and train operation plan data. Equipped with, The congestion estimation step includes a train allocation step that, based on the operation plan data, searches for trains that passengers who entered each station during a predetermined time period will use to travel to their departure station, based on predetermined search conditions including two or more conditions, and then executes a train allocation process that assigns the passengers to the searched trains as crew members. In the aforementioned train allocation process, passengers who enter each station are allocated in a predetermined ratio according to the number of conditions in the search criteria, and these passengers are allocated as crew members to the trains found based on the corresponding conditions. A method for predicting passenger flow.
12. The passenger flow prediction method according to claim 10 or 11 is performed on a computer. Passenger flow forecasting program.