Train diagram simulation device, train diagram simulation method, and train diagram simulation program

Abstract

To reflect an actual preference of each passenger when estimating a movement route of each passenger in a new train diagram.SOLUTION: A train diagram simulation device 10 comprises a data acquisition unit 12 and a new route estimation unit 15. The new route estimation unit 15 estimates a new route D7 indicating that each passenger uses which train in a new train diagram and moves from an entrance station to an exit station, based on an estimation current route D6 acquired by the data acquisition unit 12 and new diagram data D2.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a train diagram simulation device, a train diagram simulation method, and a train diagram simulation program for predicting passenger flow when a new train diagram proposal is applied. [Background technology]

[0002] Railway operators regularly revise train timetables to optimize passenger flow. Once a new train timetable is applied, it is practically difficult to repeatedly fine-tune it through trial and error while the new train timetable is in operation. Therefore, various simulators have been proposed to predict passenger flow when a new train timetable plan is applied, so that the quality of the created new train timetable plan can be evaluated before it is applied (see, for example, Patent Documents 1 to 3).

[0003] Patent Document 1 discloses a device that sets desired arrival times at entry and exit stations for each passenger and estimates which train each passenger will travel on in a new train timetable proposal. Patent Document 2 discloses a device that predicts the number of passengers and calculates the congestion rate for each train's running section in order to evaluate the merits of a new train timetable proposal. In the device disclosed in Patent Document 3, as the simulation time advances moment by moment, virtual passengers appear at each station based on data collected by automatic ticket gates, etc. Each virtual passenger is assigned a behavioral attribute of seat preference or fastest train selection based on a predetermined probability. The movement route of each virtual passenger is estimated according to the behavioral attribute. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2007-237948 [Patent Document 2] Japanese Patent Application Publication No. 2019-209769 [Patent Document 3] Japanese Patent Application Laid-Open No. 2015-229459 Summary of the Invention [Problem to be solved by the invention]

[0005] However, conventional devices do not fully consider the route that each passenger actually prefers to take to travel from their entry station to their exit station. For example, in Patent Documents 1 and 2, travel routes are not estimated taking preferences into consideration. In Patent Document 3, behavioral attributes are considered to be parameters that reflect preferences, but if the probabilities used to determine the behavioral attributes of each virtual passenger do not reflect reality, the accuracy of passenger flow prediction will decrease.

[0006] Therefore, the present invention aims to more strongly reflect the actual preferences of each passenger in the prediction results, thereby improving the accuracy of passenger flow predictions. [Means for solving the problem]

[0007] The train schedule simulation device according to the present invention comprises a data acquisition unit and a new route estimation unit. The data acquisition unit acquires an estimated current route estimated based on current schedule data and ticket gate passage data, and new schedule data. The current schedule data indicates the current train schedule. The ticket gate passage data includes the entry records of each passenger and the corresponding exit records collected from automatic ticket gates during operation of the current train schedule. The estimated current route indicates which train each passenger used to travel from the entry station to the exit station in the current train schedule. The new schedule data indicates the new train schedule. The new route estimation unit estimates a new route indicating which train each passenger will use to travel from the entry station to the exit station in the new train schedule, based on the estimated current route and the new schedule data.

[0008] Here, a "train schedule" is a train operation plan. A "current train schedule" is a train schedule currently in operation. A "new train schedule" is a passenger flow simulation target in the train schedule simulation device according to the present invention, and is, for example, one of the proposed train schedules created to be applied in place of the current train schedule in future revisions. "Schedule data" is information indicating an operation plan, and includes, for example, information indicating the starting station, terminal station, and intermediate stops for each train in operation, as well as information indicating the departure time from the starting station, the arrival time and departure time at intermediate stops, and the arrival time at the terminal station.

[0009] The "entrance record" in ticket gate passage data is information including the entrance station and the time when the ticket gate was passed through. The "exit record" in ticket gate passage data is information including the exit station and the time when the ticket gate was passed through. A set of data consisting of an entrance record and its corresponding exit record (hereinafter sometimes referred to as "trip data") identifies one trip from entry to exit, i.e., when and where a passenger departed and when and where they arrived.

[0010] Unless otherwise specified, a "passenger" refers to the subject of a trip. A trip corresponds one-to-one with one passenger as its subject. The "estimated current route" is information that indicates which train each passenger used in the current train schedule to complete a trip. Therefore, the estimated current route contains information related to passenger preferences, such as whether they chose an express train or stayed on a local train when an express train was available. The estimated current route also contains information related to passenger preferences, such as whether they transfer between trains to minimize the number of transfers, or whether they don't mind transferring and will transfer multiple times if it's the fastest option. The estimated current route reflects passenger preferences, such as whether they want to arrive at their destination as quickly as possible and whether they consider transfers to be a hassle.

[0011] According to the above configuration, the new route estimation unit estimates, based on the estimated current route, a new route that will be used in place of the estimated current route when the current train schedule is revised to a new train schedule. The estimated current route is based on ticket gate passage data collected from automatic ticket gates while the current train schedule is in operation, and reflects the actual preferences of passengers. The passenger preferences reflected in the estimated current route are also reflected in the new route. Such estimation is performed for each passenger (i.e., for each trip). Therefore, the predicted results of passenger flow when the new train schedule is applied reflect the actual preferences of each passenger, improving the accuracy of passenger flow prediction.

[0012] The information representing the estimated current route or new route may include a boarding time indicating the departure time from the entrance station of the train that each passenger is estimated to board at the entrance station, a disembarking time indicating the arrival time at the exit station of the train that each passenger is estimated to disembark at, and a number of transfers indicating the number of train changes required to travel from the entrance station to the disembarking station. According to the above configuration, the information representing the estimated current route and the new route includes the boarding time, the disembarking time, and the number of transfers, so the estimated current route is information that accurately reflects the passenger's preferences, and the new route based on it also accurately reflects the passenger's preferences.

[0013] Furthermore, it is thought that passengers will not like the time difference between the estimated current route and the new route to change significantly after the train schedule is revised, and will likely choose a route that minimizes the time difference. This is particularly true for passengers who, such as commuter passengers, have set fixed distances and times that are part of their daily routine. Therefore, the new route estimation unit may estimate a new route based on the boarding time and disembarking time of the estimated current route.

[0014] According to the above configuration, the new route estimation unit estimates a new route that is suited to the passenger's tendencies, thereby improving the accuracy of estimating the new route. In particular, it is considered highly likely that each passenger will select a route with as small a time difference as possible between boarding times. Therefore, the new route estimation unit may estimate the new route so as to reduce the time difference between the boarding time of the new route and the boarding time of the estimated current route.

[0015] According to the above configuration, the new route estimation unit estimates a new route that is suited to the passenger's tendencies, thereby improving the accuracy of estimating the new route. Furthermore, it is considered highly likely that each passenger will select a route with as small a time difference as possible between disembarking times. Therefore, the new route estimation unit may estimate a new route so as to reduce the time difference between the drop-off time on the new route and the drop-off time on the estimated current route.

[0016] According to the above configuration, the new route estimation unit estimates a new route that is suited to the passenger's tendencies, thereby improving the accuracy of estimating the new route. In particular, it is highly likely that passengers will choose an earlier disembarkation time than a later disembarkation time, even if the difference in time from the estimated disembarkation time of the current route is the same. Travel often involves business at the destination. It is highly likely that a delay in arrival time at the destination after a timetable revision will not be desirable.

[0017] Therefore, the new route estimation unit may estimate a new route such that the drop-off time on the new route is earlier than the drop-off time on the estimated current route. According to the above configuration, the new route estimation unit estimates a new route that is suited to the passenger's tendencies, thereby improving the accuracy of estimating the new route. Furthermore, transfers involve a significant burden on passengers, such as walking and waiting, and it is highly likely that passengers will not want to have to transfer more frequently after the train timetable revision.

[0018] Therefore, the new route estimation unit may estimate a new route so that the number of transfers on the new route is equal to or less than the number of transfers on the estimated current route. According to the above configuration, the new route estimation unit estimates a new route that is suited to the passenger's tendencies, thereby improving the accuracy of estimating the new route. Furthermore, if the new train timetable also contains a route that is the same as the estimated current route, it is thought that passengers will have no particular motivation to take the trouble to select a different train after the train timetable is revised.

[0019] Therefore, if a route identical to the estimated current route exists in the new train timetable, the new route estimation unit may estimate the identical route as the new route. According to the above configuration, the new route estimation unit estimates a new route that is suited to the passenger's tendencies, thereby improving the accuracy of estimating the new route. If there is no route in the new train timetable that is the same as the estimated current route, the new route estimation unit may extract multiple route candidates as new route candidates.

[0020] According to the above configuration, the new route estimation unit extracts a plurality of route candidates and estimates a new route, which improves the accuracy of passenger flow prediction compared to speculating on one route as the new route. The new route estimation unit may extract as a route candidate a route whose boarding time falls within a predetermined first time range based on the boarding time of the estimated current route and whose disembarking time falls within a predetermined second time range from the disembarking time of the estimated current route.

[0021] According to the above configuration, the new route estimation unit extracts as new route candidates only those routes whose time difference from the estimated current route falls within a predetermined time range, which makes it possible to extract route candidates that match passenger habits, thereby improving the accuracy of passenger flow prediction. The new route estimation unit may calculate a route time difference representing the magnitude of the time difference from the estimated current route for each of the extracted multiple route candidates, and allocate passengers to the multiple route candidates according to the calculated route time difference. The smaller the route time difference of a route candidate, the higher the allocation rate of passengers to the route candidate.

[0022] Here, the "route time difference" is a parameter that represents the magnitude of the time difference between the route candidate and the estimated current route. It is considered that passengers do not like the time difference between the estimated current route and the new route to change significantly after a train schedule revision, and are likely to select a route with as small a time difference as possible. According to the above configuration, in the estimation process (particularly the passenger allocation process) by the new route estimation unit, the smaller the route time difference between the route candidates, the higher the allocation rate to the route candidate. Therefore, the new route estimation unit can allocate passengers to multiple route candidates in accordance with the passenger's tendencies, improving the accuracy of estimating new routes and the accuracy of predicting passenger flow. Note that the number of passengers can also be expressed as a non-integer. Therefore, even if there is only one passenger before allocation, it is permissible to allocate a number of passengers less than one to multiple route candidates.

[0023] The greater the time difference between the boarding time of the candidate route and the boarding time of the estimated current route, the greater the route time difference may be. According to the above configuration, the time difference between routes is taken into consideration when expressing the magnitude of the time difference between routes as a numerical value called the route time difference. The larger the time difference between boarding times, the larger the route time difference. In this case, the smaller the time difference between boarding times between route candidates, the higher the allocation rate. The new route estimation unit can allocate passengers to multiple route candidates in accordance with passenger tendencies, improving the accuracy of estimating new routes and predicting passenger flow.

[0024] When the drop-off time of the route candidate is later than the drop-off time of the estimated current route, the route time difference may be larger as the time difference between the drop-off time of the route candidate and the drop-off time of the estimated current route is larger. According to the above configuration, when expressing the magnitude of the time difference between routes as a numerical value called the route time difference, the delay of the drop-off time of the candidate route relative to the drop-off time of the estimated current route is taken into consideration. The greater the delay, the greater the route time difference. In this case, the greater the delay in the drop-off time of a candidate route, the lower the allocation rate. The new route estimation unit can allocate passengers to multiple candidate routes in accordance with passenger tendencies, improving the accuracy of estimating new routes and predicting passenger flow.

[0025] Passengers with the same estimated current route may be counted in advance, and the new route estimation unit may simultaneously allocate multiple passengers counted as having the same estimated current route to multiple route candidates at an allocation rate according to the route time difference. According to the above configuration, when there are multiple passengers with the same estimated current route, a series of processes including extracting multiple route candidates based on the estimated current route, calculating route time differences corresponding to each route candidate, and allocating passengers according to the route time differences is performed once, so that the multiple passengers can be simultaneously allocated to the extracted multiple route candidates. Compared to performing the series of processes multiple times for each of multiple passengers with the same estimated current route, prediction results can be produced more quickly.

[0026] The earlier the boarding time of the route candidate is relative to the entry time used to estimate the estimated current route, the greater the route time difference may be. The later the disembarking time of the route candidate is relative to the exit time used to estimate the estimated current route, the greater the route time difference may be. According to the above configuration, when expressing the magnitude of the time difference between routes as a numerical value called the route time difference, the information on the entry time and exit time specific to each passenger is used again. Since the number of passengers is allocated based on the route time difference thus derived, the accuracy of passenger flow estimation is improved.

[0027] The train diagram simulation device may further include a passenger number estimation unit that estimates the number of passengers on each train in the new train diagram based on the new route estimated by the new route estimation unit. According to the above configuration, the accuracy of estimating the new route that each passenger will select instead of the estimated current route is improved, so the number of passengers on each train in the new train schedule can also be estimated with high accuracy.

[0028] A train diagram simulation device according to another embodiment of the present invention also includes a data acquisition unit and a new route estimation unit. The data acquisition unit acquires an estimated current route estimated based on current diagram data and data indicating the movement history of each passenger, and new diagram data. The current diagram data indicates the current train diagram. The estimated current route indicates which train each passenger will use to travel from the entrance station to the exit station in the current train diagram. The new diagram data indicates the new train diagram. The new route estimation unit estimates a new route indicating which train each passenger will use to travel from the entrance station to the exit station in the new train diagram, based on the estimated current route and the new diagram data.

[0029] "Data showing the movement history of each passenger" includes, for example, ticket gate passage data, location information acquired from terminals carried by passengers, image information acquired from cameras installed near or within station ticket gates, ticket reservation information by passengers, etc. Based on such data, it is possible to identify when and where a particular passenger departed and arrived, and a single trip from entry to exit.

[0030] Therefore, in this configuration, the passenger preferences reflected in the estimated current route are also reflected in the new route, and the predicted passenger flow when the new train schedule is applied reflects the actual preferences of each passenger. The train schedule simulation method of the present invention comprises obtaining an estimated current route, which is estimated based on current schedule data indicating the current train schedule and ticket gate passage data including the entrance records of each passenger and the corresponding exit records collected from automatic ticket gates during operation of the current train schedule, and new schedule data indicating the new train schedule; and estimating a new route, based on the estimated current route and the new schedule data, which indicates which train each passenger will use to travel from the entrance station to the exit station in the new train schedule.

[0031] A train diagram simulation program according to the present invention causes a computer to execute the above-described method. These methods and programs have the same or corresponding technical features as the above-mentioned devices, so that passenger preferences reflected in the estimated current route are also reflected in the new route, and the predicted passenger flow when the new train schedule is applied reflects the actual preferences of each passenger. [Effects of the Invention]

[0032] According to the present invention, the actual preferences of each passenger are taken into consideration when estimating the travel route of each passenger in a new train schedule, thereby improving the accuracy of passenger flow predictions. [Brief explanation of the drawings]

[0033] [Figure 1] 1 is a block diagram showing the configuration of a train diagram simulation device according to a first embodiment of the present invention and a system including the same. [Figure 2] FIG. 2 is a schematic diagram of timetable data. [Figure 3] FIG. 10 is a schematic diagram of ticket gate passage data. [Figure 4] 3 is a flowchart showing pre-processing in the train diagram simulation method according to the first embodiment of the present invention. [Figure 5] FIG. 10 is an explanatory diagram of the process of estimating an estimated current route when a passenger prefers the route that will get them to their destination the fastest. [Figure 6] FIG. 10 is an explanatory diagram of the estimation process for the estimated current route when a passenger prefers a route with fewer transfers and a higher probability of getting a seat. [Figure 7] FIG. 10 is an explanatory diagram of the process of estimating an estimated current route when a passenger prefers the route that will get them to their destination the fastest. [Figure 8] FIG. 10 is an explanatory diagram of the estimation process for the estimated current route when a passenger prefers a route with fewer transfers and a higher probability of getting a seat. [Figure 9] FIG. 10 is an explanatory diagram of a calculation process of an estimated current route. [Figure 10] 3 is a flowchart showing the main process in the train diagram simulation method according to the first embodiment of the present invention. [Figure 11] 10 is a flowchart showing a new route estimation process (passenger allocation process) in this process. [Figure 12] FIG. 10 is an explanatory diagram of a new route estimation process. [Figure 13] 13A to 13D are diagrams showing examples of output results. [Figure 14] FIG. 10 is an explanatory diagram of an estimated current route in the train diagram simulation device according to the second embodiment of the present invention. [Figure 15] 10 is a flowchart showing a part of a new route estimation process (passenger allocation process) in the main process in the train diagram simulation method according to the second embodiment of the present invention. [Figure 16] 10 is a flowchart showing a part of a new route estimation process (passenger allocation process) in the main process in the train diagram simulation method according to the second embodiment of the present invention. [Figure 17] FIG. 10 is an explanatory diagram of a new route estimation process. [Figure 18] FIG. 10 is an explanatory diagram of a new route estimation process. [Figure 19] FIG. 10 is an explanatory diagram of a new route estimation process. DETAILED DESCRIPTION OF THE INVENTION

[0034] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Note that the same or corresponding elements throughout the drawings are designated by the same reference numerals, and redundant explanations will be omitted. [First embodiment] (system) FIG. 1 shows a system 1 including a train diagram simulation device 10 according to an embodiment of the present invention. The train diagram simulation device 10 and the system 1 including the same are suitable for use in railway operators' train diagram revision operations. Railway operators periodically revise train diagrams to optimize passenger flow. Optimizing passenger flow includes reducing congestion on trains or in station premises. In the process of revising a train diagram, a new train diagram is created to replace the current one.

[0035] The train diagram simulation device 10 predicts passenger flow when a new train diagram is applied. This makes it possible to evaluate the merits of the new train diagram before it is actually applied. Therefore, it is possible to modify the new train diagram before the revision is actually implemented, for example, to further optimize passenger flow in specific time periods or sections. The train diagram simulation device 10 is realized, for example, in a server managed by a railway operator or an information terminal used by an employee of the railway operator who is involved in timetable revision work. The server or information terminal is a computer equipped with a CPU, memory, and an input / output interface.

[0036] The train diagram simulation program according to the present invention is installed in the computer and stored in its memory. The CPU reads the program from the memory and processes information according to the steps instructed by the program. This allows the computer to execute the train diagram simulation method according to the present invention and function as a train diagram simulation device.

[0037] The system 1 includes a train diagram simulation device 10, as well as an input device 2, a display device 3, and a data server 4, all of which are connected to the train diagram simulation device 10. The input device 2 receives input from an operator of data to be stored in the train diagram simulation device 10 (for example, current diagram data D1 and new diagram data D2) and a command to start a simulation. The display device 3 displays the predicted results of passenger flow obtained by the train diagram simulation device 10. The data server 4 stores ticket gate passage data D3. The ticket gate passage data D3 is generated based on entry records and exit records collected from automatic ticket gates 5 installed at each railway station.

[0038] (Diamond data) The current timetable data D1 is timetable data corresponding to the current train timetable. The new timetable data D2 is timetable data corresponding to the new train timetable and is the subject of the simulation. In this document, the term "timetable data" may be used without distinction between the current timetable data D1 and the new timetable data D2.

[0039] Timetable data is information that shows train timetables, that is, train operation plans. In general rail passenger transport, the daytime is "operation time" when passenger transport takes place, and the period from late at night to early morning is "track maintenance time" when passenger transport is suspended and track maintenance work is carried out, with operation time and track maintenance time repeating on a daily cycle. Timetable data is made up of operation data D4 (see Figure 2) for each train that operates within one operating time period.

[0040] Train schedules are created by direction, and train schedules for each direction are further created by day. The schedule data includes multiple types of data according to direction and day. In this document, the term "timetable data" may be used to describe multiple types of data without distinguishing between them. Examples of the types of timetables by direction include an inbound timetable and an outbound timetable. Examples of the types of timetables by day include a weekday timetable that applies to weekday operating hours and a holiday timetable that applies to holiday operating hours. Further examples include a seasonal timetable that applies only on weekends for a limited period, such as in autumn, and in which special trains are operated to transport vacationers, and an event timetable that applies only on days when events that are expected to attract large numbers of people are held, such as soccer games or horse races, and in which special trains are operated to transport event guests.

[0041] For ease of explanation, Figure 2 shows a schematic diagram of timetable data for a hypothetical P line. There are eight stations, A to H, on P line. Four-digit numbers indicate the hour and minute, downward arrows indicate passing stations, and blank spaces indicate non-operating sections. In the illustrated example, weekday operating hours in the current train timetable are set from 5:30 to 24:14. The operation data D4 includes corresponding train type data, formation data, stop station data, and time data. The type data indicates the type of train (for example, limited express, express, local, etc., and in the case of limited express, whether it is a paid or free train). The formation data includes information related to passenger capacity, such as the number of cars that make up the train. The stop station data includes information indicating the train's starting station, intermediate stop stations, and terminal station. The time data includes information indicating the departure time from the starting station, the arrival time and departure time at intermediate stop stations, and the arrival time at the terminal station.

[0042] (Ticket gate passing data) As shown schematically in FIG. 3, the ticket gate passage data D3 is a collection of multiple trip data D5. Each trip data D5 is a set of data including an entry record and a corresponding exit record. The entry record includes the entry station and the time of passing through the automatic ticket gate 5 installed at the entry station. The exit record includes the exit station and the time of passing through the automatic ticket gate 5 installed at the exit station. The entry record and exit record are acquired by an automatic ticket gate at different locations and at different times. The entry record and exit record are stored on the data server 4 in a state where they are linked to each other via ticket ID information assigned to each ticket used by the passenger to form trip data D5.

[0043] Trip data D5 indicates when a passenger entered and exited a station, identifying a single train trip from entry to exit. A single trip corresponds one-to-one with a single "passenger" who is the subject of the trip. Figure 3 shows an example of trip data indicating travel from Station A to Station D in the early morning, and trip data indicating travel from Station D to Station A in the evening. These two trip data contain the same ticket ID information, and therefore are recognized as representing round-trip travel by the same person using a commuter pass. These two trips are defined as different trips by different "passengers" because they differ in both direction and time of travel.

[0044] (Train schedule simulation device / method) Returning to Figure 1, the train diagram simulation device 10 of this embodiment has, in accordance with its functions, a memory unit 11, a data acquisition unit 12, a current route estimation unit 13, a counting unit 14, a new route estimation unit 15, a passenger number estimation unit 16, an output unit 17, and a diagram evaluation unit 20. The memory unit 11 stores various information necessary for predicting passenger flow. In this embodiment, the memory unit 11 is configured to store, as a mere example, current train schedule data D1 and new train schedule data D2 input by an operator via the input device 2. However, part of the ticket gate passage data D3 may be pre-stored in the memory unit 11, and part or all of the train schedule data may be saved in a storage device external to the train schedule simulation device 10.

[0045] Below, the operations of the data acquisition unit 12, current route estimation unit 13, aggregation unit 14, new route estimation unit 15, passenger number estimation unit 16, output unit 17, and timetable evaluation unit 20 will be explained in accordance with the procedure of the train timetable simulation method of the present invention. <Pretreatment> 4, in the train diagram simulation device 10 according to this embodiment, the current route estimation unit 13 and the counting unit 14 perform estimation processing and counting processing of the estimated current route D6 as preprocessing prior to predicting passenger flow for a new train diagram. For example, when the operator inputs the current train diagram data D1 and a command to perform preprocessing, the preprocessing starts.

[0046] The estimated current route D6 is information indicating which trains in the current train schedule each passenger used to complete a trip identified by one trip data D5. More specifically, the estimated current route D6 includes information indicating the entrance station, exit station, boarding time at the entrance station, disembarking time at the exit station, and one or more trains used to travel from the entrance station to the exit station. In a case where there is one or more transfers and multiple trains are used, the estimated current route D6 also includes information indicating each transfer station, disembarking time at each transfer station, boarding time at each transfer station, and waiting time at each transfer station.

[0047] In the pre-processing, the data acquisition unit 12 acquires the current train schedule data D1 and the ticket gate passage data D3 (S11). The estimated current route D6 is estimated for each timetable type. The data acquisition unit 12 acquires timetable data for each type that constitutes the current timetable data D1. The data acquisition unit 12 acquires one day's worth of ticket gate passage data D3 corresponding to each timetable type. For example, to estimate the estimated current route D6 based on weekday timetable data, ticket gate passage data D3 for one weekday is acquired, and to estimate the estimated current route D6 based on holiday timetable data, ticket gate passage data D3 for one holiday is acquired.

[0048] <<Estimated current route>> When the current route estimation unit 13 has finished estimating the inbound weekday timetable data, it then starts estimating the inbound holiday timetable data, and so on, performing estimation processing on all of the multiple timetable types while sequentially changing the estimation target. Therefore, the current route estimation unit 13 selects one timetable type to be estimated (S12).

[0049] The current route estimation unit 13 refers to the timetable data selected as the estimation target from the current timetable data D1 and the corresponding one-day ticket gate passage data D3 (S13), and estimates an estimated current route D6 from one trip data D5 in the ticket gate passage data D3 (S14). The process of estimating the estimated current route D6 is repeated until estimation of the estimated current route D6 is completed for all trip data D5 constituting the ticket gate passage data D3 (S15: N → S13).

[0050] Fig. 5 specifically illustrates the process of estimating an estimated current route D6-I corresponding to the trip data D5-I shown in Fig. 3. The current route estimation unit 13 sets a current route search period t1 based on the trip data D5-I. The current route search period t1 is set between the entry time and exit time of the trip data D5. More precisely, the current route search period t1 may be set taking into account the station premises travel time t2, which is individually set for each station. In this case, the start of the current route search period t1 is set to a time that is the station premises travel time t2 at the entrance station after the entry time, and the end of the current route search period t1 is set to a time that is the station premises travel time t2 at the exit station before the exit time. Station premises travel time t2 is the minimum time required for a passenger to travel between the ticket gate and the platform within the station, and is obtained empirically for each station. For example, in stations where the ticket gate floor and the platform floor are different, the station premises travel time t2 is set to be relatively long because there are stairs between the ticket gate and the platform.

[0051] Next, the current route estimation unit 13 uses a route search algorithm to refer to the current timetable data D1 and searches for a route that departs from the entrance station and arrives at the exit station within the current route search period t1. In other words, the current route estimation unit 13 searches for a route that satisfies both the condition that "the boarding time at the entrance station is after the start of the current route search period t1" and the condition that "the disembarking time at the exit station is before the end of the current route search period t1."

[0052] If only one route is found, that route is estimated as the estimated current route D6. If two or more routes are found, an appropriate one of the routes may be estimated as the estimated current route D6. Alternatively, one passenger may be assigned to multiple routes. 5, a passenger enters station A at 5:20 and exits station E at 5:55. If the travel time t2 within station A is, for example, 2 minutes and the travel time t2 within station E is, for example, 5 minutes, the current route estimation unit 13 searches for a route that allows departure from station A after 5:22 and arrival at station E before 5:50. In other words, the current route estimation unit 13 searches for a route that satisfies the conditions that the boarding time at the entrance station is after 5:22 and the disembarking time at the exit station is before 5:50.

[0053] In the case shown in Figure 5, there is only one route that can realize the trip specified by trip data D5-III. The route is "board train T01 departing at station A at 5:30, transfer from train T01 to train T02 at station D, and get off train T02 arriving at station E at 5:48." If a passenger misses train T01 at station A, even if they board the next train T03, it is impossible to arrive at station E before 5:50.

[0054] Therefore, the current route estimation unit 13 estimates this route as an estimated current route D6-I corresponding to the trip data D5-I. Although not shown, when similar processing is performed on the trip data D5-V, the current route estimation unit 13 estimates this route D6-I as an estimated current route corresponding to the trip data D5-V. In the case shown in FIG. 6, there are three routes D6-IIa, D6-IIb, and D6-IIc that can realize the trip specified by the trip data D5-II.

[0055] Route 1 D6-IIa allows the fastest arrival at the exit station. Passengers board train T01, the first train to arrive after entering the station, and disembark at station D, which is train T01's final destination. At station D, train T02 departs immediately after train T01 arrives, taking into account the connection with train T01. Passengers transfer to train T02 at station D and disembark from train T02 at station F. According to the second route D6-IIb, although it arrives later than the first route D6-IIa, no transfer is required. After entering the station, passengers see off the first arriving train T01, board the next train T03, and get off this train T03 at station F.

[0056] The third route D6-IIc is a combination of the above routes D6-IIa and D6-IIb. Passengers board the first train T01 that arrives after entering the station and disembark at station D, which is train T01's final destination. Passengers see off train T02 at station D, transfer to the next train T03, and disembark at station F. The current route estimation unit 13 calculates the sum of the time difference between the departure time and the entry time at the entry station and the time difference between the arrival time and the exit time at the exit station for each of the three retrieved routes D6-IIa, D6-IIb, and D6-IIc. The current route estimation unit 13 may estimate the route with the smallest sum of these time differences as the current route D6-II. The current route estimation unit 13 may allocate one passenger to the three retrieved routes D6-IIa, D6-IIb, and D6-IIc such that a larger number of passengers is allocated to a route with a smaller time difference.

[0057] In the case shown in FIG. 7, there is only one route that can realize the trip identified by the trip data D5-III. The route is "board train T03 departing at 5:49 at station B, transfer from train T03 to express train T04 at station D, and get off train T04 arriving at station H at 6:15." For example, if the passenger stays on local train T03 without transferring at station D, the arrival at station E will be after the departure time. Therefore, the current route estimation unit 13 estimates that this route is the estimated current route D6-III corresponding to the trip data D5-III.

[0058] In the case shown in FIG. 8, there are two routes D6-IVa and D6-IVb that can realize the trip specified by the trip data D5-IV. Route 1, D6-IVa, allows passengers to arrive at the exit station faster. Passengers board the first arriving local train, T03, and transfer to express train T04 at Station D. According to the second route D6-IVb, although it arrives later than the first route D6-IVa, no transfer is required. Passengers first board the same local train T03. At Station D, express train T04 arrives after local train T03 and departs before it. Passengers see off this express train T04 and continue on to Station H on the local train T03 they boarded at the entrance station.

[0059] In the same manner as described above, the current route estimation unit 13 estimates one route from the two searched routes D6-IVa, D6-IVb as the estimated current route D6-IV, or allocates one passenger to the two searched routes D6-IVa, D6-IVb. In this way, the estimated current route D6 contains information related to passenger preferences, such as whether the passenger prefers speed, or whether they prefer fewer transfers and a higher probability of getting a seat even if it means sacrificing some speed. The estimated results of the estimated current route are based on the trip data D5 actually measured by the automatic ticket gates 5, and reflect the actual preferences of passengers using the target route.

[0060] <<Summary>> When the estimation of the estimated current route D6 for all trip data D5 for the timetable type to be estimated is completed (S15: Y), the counting unit 14 counts passengers for whom the estimated current route D6 is the same (S16). "The estimated current route is the same" means that the entry station and exit station are the same, and the train used to travel from the entry station to the exit station is exactly the same. If a transfer is included, the transfer station and the number of transfers are also the same.

[0061] The upper part of FIG. 9 shows a portion of the estimated current route D6 before the counting process. The top three lines show the estimated current route D6, which is "Enter station A, travel to station D on train T01, transfer to train T02 at station D, and exit at station E." The counting unit 14 counts the number of passengers on the estimated current route (i.e., the number of passengers who used the estimated current route). In the illustrated example, there are three passengers.

[0062] The three lines below show an estimated current route D6, which states, "Enter Station A, travel to Station E on train T03, and exit at Station E." The estimated current route D6 is the same as the above route in terms of the entry station and exit station, but differs from the above route in terms of the train used (i.e., boarding time and disembarking time) and whether or not there is a transfer. Therefore, the estimated current route D6 is considered to be a route different from the above route. The counting unit 14 counts the number of passengers on the estimated current route D6 (i.e., the number of passengers who have used the estimated current route D6). In the illustrated example, there are three passengers.

[0063] Further down in the two lines, an estimated current route D6 is shown, which states, "Enter Station A, travel to Station F on train T03, and exit at Station F." The estimated current route D6 is the same as the above route in terms of the entry station, the train used, and whether or not a transfer is required, but is different from the above route in terms of the exit station. Therefore, the estimated current route D6 is considered to be a different route from the above route. The counting unit 14 counts the number of passengers on the estimated current route D6 (i.e., the number of passengers who have used the estimated current route D6). In the illustrated example, there are two passengers.

[0064] The lower part of Fig. 9 shows a portion of the estimated current route D6 after the aggregation process. The number of estimated current routes D6 after the aggregation process is less than the number of trip data D5 (in the illustrated example, the amount of data of 8 lines at the top is reduced to the amount of data of 3 lines at the bottom). This not only reduces the pressure on the capacity of the memory unit 11, but also contributes to speeding up the main process described below. In particular, when the population along the target line is large and the transportation volume per train is large, the aggregation process will significantly reduce the number of estimated current routes D6 compared to the number of trip data D5. If the line extends from the city center to the suburbs and there are extremely many users at the terminal station on the city center side during the morning rush hour, one of the conditions for determining the same route (the exit station) will be common to many passengers, so the number of estimated current routes D6 will be reduced even more significantly.

[0065] When the counting is completed, the data of the counted estimated current route D6 as shown in the lower part of FIG. 9 is stored in the storage unit 11 for the timetable type to be estimated (S17). The current route estimation unit 13 and the counting unit 14 repeat the same process while selecting each of the timetable types to be estimated one by one until the estimation process and counting process are completed for all timetable types (S18: N → S12). When the process for all timetable types is completed (S18: Y), the pre-processing ends.

[0066] <Main process (passenger flow prediction process)> 10 and 11, once the data of the estimated current route D6 that has been compiled is prepared, the train diagram simulation device 10 according to this embodiment becomes able to execute the process of predicting passenger flow in a new train diagram. For example, this process starts when an operator inputs new diagram data D2 and a command to execute this process. In this process, the data acquisition unit 12 acquires new timetable data D2 and data on the estimated current route D6 (S21). Next, the new route estimation unit 15 estimates a new route D7 corresponding to the estimated current route D6 (S30).

[0067] <<New route estimation>> In the process S30 of estimating the new route D7 shown in Fig. 11, the new route estimation unit 15 estimates a new route D7 indicating which train each passenger will use in the new train timetable to travel from the same entry station to the same exit station. In other words, when a new train timetable is newly applied in place of the current train timetable due to a timetable revision, the new route estimation unit 15 estimates, based on the estimated current route D6, the new route D7 that passengers will use instead of the estimated current route D6. The estimation of the new route D7 is performed for each timetable type.

[0068] <<<Extraction of route candidates>>> First, the new route estimation unit 15 selects a timetable type to be estimated (S31), and refers to the timetable data selected as the timetable data to be estimated from the new timetable data D2 and the corresponding data of the estimated current route D6 (S32). If upbound weekday timetable data is selected as the timetable data to be estimated, the new route estimation unit 15 refers to the data of the estimated current route D6 generated based on the same upbound weekday timetable data from the current timetable data D1.

[0069] Next, the new route estimation unit 15 extracts one estimated current route D6 as source data for the estimation process (S33). As described above, the estimated current route D6 includes information indicating the entrance station, exit station, boarding time at the entrance station, disembarking time at the exit station, and one or more trains used to travel from the entrance station to the exit station, and if there is one or more transfers and multiple trains are used, it also includes information indicating each transfer station, disembarking time at each transfer station, and boarding time at each transfer station.

[0070] Next, the new route estimation unit 15 sets a new route boarding time search period t3a and a new route disembarking time search period t3b based on the estimated current route D6 (S34). 12 is an explanatory diagram of the estimation process for new route D7. The boarding time search period t3a is set before or after the boarding time Top at the entrance station of estimated current route D6, so that the boarding time Top at the entrance station of estimated current route D6 falls within the boarding time search period t3a. As just one example, the boarding time search period t3a is set to 15 minutes before or after the boarding time Top at the entrance station of estimated current route D6, for a total of 30 minutes.

[0071] The same applies to the alighting time search period t3b. The alighting time search period t3b is set before or after the alighting time Tdp at the exit station of the estimated current route D6, so that the alighting time Tdp at the exit station of the estimated current route D6 falls within the alighting time search period t3b. As just one example, the alighting time search period t3b is set to 15 minutes before or after the alighting time Tdp at the exit station of the estimated current route D6, for a total of 30 minutes.

[0072] By appropriately setting the boarding time search period t3a and the disembarking time search period t3b in accordance with the operation cycle of the new train timetable, it becomes possible to reliably extract one or more route candidates in the processing described below. For example, if the new train timetable is designed to have a 15-minute pattern timetable, setting the period 15 minutes before and after as described above increases the reliability of extracting candidates. Next, the new route estimation unit 15 searches for a route whose boarding time falls within the boarding time search period t3a at the same entrance station as the estimated current route D6 and whose disembarking time falls within the disembarking time search period t3b at the same exit station as the estimated current route D6 (S35). That is, the new route estimation unit 15 searches for a route whose boarding time Ton falls within a predetermined first time range (e.g., 15 minutes before and after) based on the boarding time Top of the estimated current route D6 and whose disembarking time Tdn falls within a predetermined second time range (e.g., 15 minutes before and after) from the disembarking time Tdp of the estimated current route D6 (see the "New Route Example" section in FIG. 12).

[0073] Next, the new route estimation unit 15 excludes routes that have a greater number of transfers than the estimated current route D6 from the routes extracted as satisfying the above search conditions (S36). Transfers involve physical strain on the passenger, such as getting on and off trains, walking and waiting in the station, etc. It is considered that passengers do not like to have more transfers than the estimated current route. Therefore, routes that increase the number of transfers are excluded from the estimation targets (see the "Exclusion" section in Figure 12). The route remaining after the exclusion is set as a route candidate for the new route D7. If there is only one route candidate (S37: Y), the new route estimation unit 15 estimates the route candidate as the new route D7 (S38), and assigns all passengers associated with the estimated current route D6 as passengers for this one new route D7 (S39).

[0074] <<<Route time difference between route candidates>>> If multiple route candidates are extracted (S37: N), the new route estimation unit 15 calculates the route time difference Diff for each route candidate (S40). The route time difference Diff is a parameter that quantitatively indicates the magnitude of the time difference between the route candidate and the estimated current route D6. As an example, the route time difference Diff is derived from the following equation (1):

[0075] Diff=│Ton-Top│+max(0,(Tdn-Tdp))……(1) Here, Ton is the boarding time at the entrance station of each route candidate of new route D7. Top is the boarding time at the entrance station of estimated current route D6. Tdn is the disembarking time at the exit station of each route candidate of new route D7. Tdp is the disembarking time at the exit station of estimated current route D6. In this embodiment, the route time difference Diff is the sum of the boarding difference value based on the time difference between the boarding times Ton and Top at the entrance station and the disembarking difference value based on the time difference between the disembarking times Tdn and Tdp at the exit station.

[0076] The boarding difference value is the absolute value of the time difference between the boarding times Ton,Top at the entrance station. The larger the time difference between the boarding times Ton,Top at the entrance station, the larger the boarding difference value and the larger the route time difference Diff. By using the absolute value, whether the boarding time Ton of the route candidate is earlier or later than the estimated current route D6, the route time difference Diff becomes larger according to the time difference.

[0077] The disembarking differential value is the larger of two values: 0 and the subtraction value obtained by subtracting the disembarking time Tdp of the estimated current route D6 from the disembarking time Tdn of the route candidate. The later the disembarking time Tdn at the exit station is compared to the estimated current route D6, the larger the disembarking differential value and the larger the route time difference Diff. If the disembarking time Tdn is earlier than the estimated current route D6, a time difference itself occurs, but the subtraction value becomes a negative value and the disembarking differential value becomes 0. Therefore, the route time difference Diff does not change. If the disembarking time Tdn is earlier, the new route estimation unit 15 ignores the time difference between the disembarking times Tdn and Tdp when evaluating the magnitude of the time difference between the routes. However, an earlier disembarking time Tdn does not necessarily mean that the time difference between the routes is smaller.

[0078] Here, it is thought that passengers will not like the time difference between the estimated current route D6 and the new route D7 to change significantly even if the train schedule is revised. An increase in the time difference will lead to changes in passenger lifestyles. This tendency is thought to be particularly likely to appear among passengers, such as regular passengers, for whom travel within set sections and at set times is part of their daily routine. The route time difference Diff quantitatively indicates the time difference between these routes. The larger the route time difference Diff between the route candidates, the lower the probability that the route candidate will be selected as new route D7. In this way, the route time difference Diff is treated as a numerical value corresponding to the probability that a passenger will select the route candidate as new route D7.

[0079] In particular, it is thought that each passenger is likely to choose a route with as small a time difference as possible between the boarding times Ton and Top at the entrance station. This tendency is also thought to be particularly likely to appear among regular passengers, especially those traveling in the morning. In this embodiment, the route time difference Diff becomes large whether the boarding time Ton of the route candidate is earlier or later than the estimated current route D6. In light of passengers' propensity for boarding times, the route time difference Diff has a higher correlation with the probability that the route candidate will be selected as the new route D7.

[0080] Similarly, it is considered that each passenger is likely to select a route with as small a time difference as possible between the disembarking times Tdn and Tdp at the exit station. In this case, it is considered that each passenger is more likely to select an earlier disembarking time Tdn than a later one, even if the same time difference occurs from the disembarking time Tdp of the estimated current route. Travel often involves business at the destination. Regardless of whether there is a timetable change or not, it may not be desirable to delay the disembarking time Tdn at the exit station. In this embodiment, the route time difference Diff becomes large only when the drop-off time Tdn of the route candidate is later than the estimated current route D6. In light of passengers' propensity for drop-off times, the route time difference Diff has a higher correlation with the probability that the route candidate will be selected as the new route D7.

[0081] <<<Passenger Allocation>>> After the route time difference Diff of each extracted route candidate is calculated, the new route estimation unit 15 next determines whether or not there is a route candidate whose route time difference Diff is zero (S41). If there is a route candidate whose route time difference Diff is zero (S41: Y), the new route estimation unit 15 estimates only that route candidate as a new route D7 (S38), and assigns all passengers associated with the estimated current route D6 as passengers on this new route D7 (S39).

[0082] In accordance with the route candidate extraction process and equation (1), in route candidates for which the route time difference Diff is zero, the number of transfers is less than or equal to the number of transfers on the estimated current route D6, the boarding time at the entry station is the same as that on the estimated current route D6, and the disembarking time at the exit station is the same as or earlier than that on the estimated current route D6. As shown in the middle part of Figure 12, for example, if the new train timetable also contains a route that is exactly the same as the estimated current route D6, the route time difference Diff for this route will be zero. Also, for example, if express trains depart at regular intervals from terminal stations in the current train timetable, but the timetable revision upgrades the type of some express trains and reduces the number of intermediate stops for the upgraded trains compared to the current timetable, the route that uses this train will have an earlier disembarkation time Tdn while keeping the boarding time Ton the same. In this case, the route time difference Diff will be zero.

[0083] In this way, if a route for which the route time difference Diff is zero exists in the new train timetable, it is considered that passengers will not be motivated to choose another route as the new route D7. The new route estimation unit 15 estimates a route candidate for which the route time difference Diff is zero as the new route D7. Since the new route D7 is estimated so as to suit the passenger's tendencies, the estimation accuracy is high.

[0084] If there is no route candidate with a route time difference Diff of zero (S41: N), the new route estimation unit 15 calculates the ratio of passenger allocation for each route candidate according to the route time difference Diff (hereinafter referred to as "allocation rate") (S42), and allocates passengers corresponding to the estimated current route D6 as passengers to multiple route candidates according to the allocation rate (S43). The smaller the route time difference Diff of a route candidate, the higher the allocation rate of that route candidate.

[0085] Referring to the lower part of FIG. 12, for example, assume that 30 passengers are associated with one estimated current route D6. Based on this estimated current route D6, assume that there are three route candidates D7(a), D7(b), and D7(c) whose route time difference Diff is not zero. All three route candidates D7(a), D7(b), and D7(c) are treated as new routes D7. The new route estimation unit 15 estimates that these 30 passengers will select one of the three route candidates D7(a), D7(b), and D7(c) as the new route D7.

[0086] Let the allocation rate of the first route candidate D7(a) be r(a), the allocation rate of the second route candidate D7(b) be r(b), and the allocation rate of the third route candidate D7(c) be r(c). The sum of the three allocation rates r(a), r(b), and r(c) is 100%. It can be said that a 100% allocation rate r is set for the route candidate with the above-mentioned route time difference Diff being zero, regardless of the existence of other route candidates. Assume Diff(a) < Diff(b) < Diff(c). In this case, r(a) > r(b) > r(c). The largest number of passengers is allocated to the first route candidate D7(a) with the smallest route time difference, and the smallest number of passengers is allocated to the third route candidate D7(c) with the largest route time difference.

[0087] In this embodiment, the ratio of the allocation rate r between route candidates is the inverse ratio of the route time difference Diff. r(a):r(b):r(c) is equal to 1 / Diff(a):1 / Diff(b):1 / Diff(c). Suppose the route time difference Diff(a) is 2, the route time difference Diff(b) is 5, and the route time difference Diff(c) is 10. In this case, r(a):r(b):r(c) is 1 / 2:1 / 5:1 / 10 = 5:2:1. The allocation rate r(a) is 62.5%, the allocation rate r(b) is 25%, and the allocation rate r(c) is 12.5%. Therefore, out of 30 passengers, 18.75 passengers are allocated to the first route candidate D7(a), 7.5 passengers are allocated to the second route candidate D7(b), and 3.75 passengers are allocated to the third route candidate D7(c).

[0088] In this way, the new routes D7 of a plurality of passengers associated with the estimated current route D6 are estimated simultaneously. Note that one passenger corresponds to one trip data D5, and passengers are counted as integer values at the aggregation processing stage. However, it is allowed to represent the number of passengers as a non-integer, and non-integer numbers of passengers may be allocated as passengers on the trains of the new route D7. If it is necessary to round the number of passengers to an integer, it is better to round at a later stage to improve the estimation accuracy.

[0089] The process of estimating new routes D7 is repeated until the estimation of new routes D7 and the allocation of passengers are completed for all estimated current routes D6 (S44: N → S33). When the estimation of new routes D7 for all estimated current routes D6 is completed (S44: Y), the same process is repeated while selecting each schedule type to be estimated one by one until the estimation process and the counting process are completed for all schedule types (S45: N → S31). When the process for all schedule types is completed (S45: Y), the process returns to the main routine (see Figure 10).

[0090] <<Estimated number of passengers>> The passenger number estimation unit 16 calculates the number of passengers for each running section of each train based on the new route D7 estimated by the new route estimation unit 15 and the number of passengers allocated to the new route D7 (S22), for all of the operation data D4 constituting the new timetable data D2. A "running section" is a section between two adjacent stations among the starting station, intermediate stops, and terminal station of the train. In the example of FIG. 2, seven running sections are set for the local train T03, and two running sections are set for the express train T04.

[0091] The number of passengers on a certain section of a train is the sum of the number of passengers using the route that includes that section. For example, if new route D7 includes train T01, which departs from station A and terminates at station D, there are seven new routes D7: a route that travels between A and B, a route that travels between AC, a route that travels between AD, a route that travels between BC, a route that travels between BD, and a route that travels between CD. The number of passengers on section AB is the sum of the number of passengers on the three routes D7 between A and B, between AC, and between AD. The number of passengers on section BC is the sum of the number of passengers on the four new routes D7 between AC, between AD, between BC, and between BD. The number of passengers on section CD is the sum of the number of passengers on the three new routes D7 between AD, between BD, and CD. In this way, the number of passengers on each train's travel section can be calculated using the results of the estimation process for the new route D7 described above.

[0092] <<Diamond Evaluation>> The schedule evaluation unit 20 evaluates the simulation target based on the estimated number of passengers. The schedule evaluation unit 20 includes an in-car congestion evaluation unit 21 and an in-station congestion evaluation unit 22.

[0093] The in-car congestion evaluation unit 21 estimates the in-car congestion rate for each running section by dividing the number of passengers in each running section by the capacity of the train (S23) with reference to the operation data D4 for each train. The in-car congestion evaluation unit 21 determines to which numerical range, defined by a plurality of predetermined thresholds, the estimated in-car congestion rate belongs. The in-car congestion evaluation unit 21 evaluates the degree of congestion reduction achievement in stages according to the numerical range to which the in-car congestion rate belongs (S24).

[0094] The station congestion evaluation unit 22 estimates the number of people remaining in the station for each specified time period at each station (S25). Based on the estimation results for the new route D7, it is possible to tally up each transfer station and the number of passengers boarding and alighting there. By considering this number of transfer passengers, the number of passengers entering the station, and the number of passengers leaving the station, it is possible to estimate the number of passengers remaining in the station. The station congestion evaluation unit 22 compares the estimated number of people remaining with a specified threshold determined by the size of the station, etc., and evaluates the degree of achievement of the reduction in station congestion in stages (S26).

[0095] <<Output>> The output unit 17 outputs the evaluation result by the schedule evaluation unit 20 to the display device 3 (S27). By referring to the evaluation result, the operator can easily determine whether the simulation target has achieved a given objective (such as alleviating congestion or improving transportation efficiency). 13(A) to 13(D) show examples of output data. As shown in FIG. 13(A), the output unit 17 can output the congestion degree for each running section. As shown in FIG. 13(B), the output unit 17 can output the number of passengers for each running section, which is the basis for estimating the congestion degree. As shown in FIG. 13(C), the output unit 17 can output the number of entering passengers and the number of exiting passengers at each stop of each train. As shown in FIG. 13(D), the output unit 17 can output the number of passengers who get off to transfer and the number of passengers who get on by transferring at each stop of each train.

[0096] (Action and effect) According to the train diagram simulation device 10 of this embodiment configured as described above, the new route estimation unit 15 estimates a new route D7 that will be used in place of the estimated current route D6 when the current train diagram is revised to a new train diagram, based on the estimated current route D6. Passenger preferences reflected in the estimated current route D6 are also reflected in the new route D7. This estimation is performed for each passenger (i.e., for each trip). Therefore, the predicted results of passenger flow when the new train diagram is applied reflect the actual preferences of each passenger, improving the accuracy of passenger flow prediction.

[0097] The new route estimation unit 15 estimates the new route D7 based on the boarding time Top and disembarking time Tdp of the estimated current route D6. In particular, the new route estimation unit 15 estimates the new route D7 so that the time difference between the boarding time Ton of the new route D7 and the boarding time Top of the estimated current route D6 is small. The new route estimation unit 15 also estimates the new route D7 so that the disembarking time Tdn of the new route D7 is earlier than the disembarking time Tdp of the estimated current route D6. Furthermore, the new route estimation unit 15 estimates the new route D7 so that the number of transfers on the new route D7 is equal to or less than the number of transfers on the estimated current route D6. In this way, the new route estimation unit 15 estimates the new route D7 so as to suit the passenger's tendencies, thereby improving the estimation accuracy of the new route D7.

[0098] If a route identical to the estimated current route D6 exists in the new train timetable, the new route estimation unit 15 estimates the identical route as a new route D7. Since the new route estimation unit 15 estimates the new route D7 so as to conform to passenger tendencies, the estimation accuracy of the new route D7 is improved. If there is no route identical to the estimated current route D6 in the new train timetable, the new route estimation unit 15 extracts multiple route candidates as candidates for the new route D7. This improves the accuracy of passenger flow prediction compared to when one route is speculated as the new route D7.

[0099] The new route estimation unit 15 calculates a route time difference Diff for each of the extracted multiple route candidates, which indicates the magnitude of the time difference from the estimated current route D6, and allocates passengers to the multiple route candidates according to the calculated route time difference Diff. The smaller the route time difference Diff of a route candidate, the higher the allocation rate r of that route candidate. The new route estimation unit 15 can allocate passengers to the multiple route candidates in accordance with the passenger's tendencies, improving the estimation accuracy of the new route D7 and the prediction accuracy of passenger flow.

[0100] The larger the time difference between the boarding time Ton of the route candidate and the boarding time Top of the estimated current route D6, the larger the route time difference Diff, and the smaller the time difference between the boarding times Ton and Top of the route candidate, the higher the allocation rate r. The new route estimation unit 15 can allocate passengers to multiple route candidates in accordance with passenger tendencies, improving the estimation accuracy of the new route D7 and the prediction accuracy of passenger flow.

[0101] When the drop-off time Tdn of the route candidate is later than the drop-off time Tdp of the estimated current route D6, the greater the time difference between the drop-off time Tdn of the route candidate and the drop-off time Tdp of the estimated current route D6, the larger the route time difference Diff and the lower the allocation rate r. The new route estimation unit 15 can allocate passengers to multiple route candidates in accordance with passenger tendencies, improving the estimation accuracy of the new route D7 and the prediction accuracy of passenger flow.

[0102] The new route estimation unit 15 simultaneously allocates multiple passengers who have been counted as having the same estimated current route D6 to multiple route candidates at an allocation rate r according to the route time difference Diff. By performing a series of processes for allocation once, multiple passengers are simultaneously allocated to multiple route candidates. Compared to performing a series of processes multiple times for each of multiple passengers who have the same estimated current route D6, prediction results can be produced more quickly.

[0103] (Modification of this embodiment) In the first embodiment, the path time difference Diff is calculated by equation (1). However, equation (1) can be modified as appropriate. For example, any of the following equations (2) to (4) may be employed. Diff=│Ton-Top│+│Tdn-Tdp│……(2) Diff=│Ton-Top│ m +max(0,(Tdn-Tdp)) n ...(3) Diff=│Ton-Top│ m +│Tdn-Tdp│ n ……(4) As shown in equation (2), the absolute value of the time difference may be used as the drop-off difference value, and the route time difference Diff may also become larger when the drop-off time Tdn becomes earlier than the current time.

[0104] Furthermore, as shown in equations (3) and (4), the boarding differential value and the disembarking differential value may be powers of the absolute value of the time difference. The exponents m and n may be the same or different, but are equal to or greater than 1. The larger the differential value, the greater the rate of increase in the route time difference Diff, and the more the allocation rate decreases. Note that it is sufficient that either one of the exponents m and n is 1, and it is sufficient that at least one of the boarding differential value and the disembarking differential value is essentially a power (the boarding differential value or the disembarking differential value may be the absolute value of the time difference itself).

[0105] Second Embodiment Next, a second embodiment of the present invention will be described, focusing on the differences from the first embodiment. This embodiment differs from the first embodiment (see FIG. 11) in the estimation process for new route D7. The estimation process according to this embodiment differs from the first embodiment in, for example, using trip data D5 to derive the route time difference Diff, searching for route candidates for new route D7 based on train schedules rather than time widths, and changing the estimation method for new route D7 depending on the time of day.

[0106] In the first embodiment, the route time difference Diff is the sum of the boarding difference value and the disembarking difference value, where the boarding difference value is based on the time difference between the boarding times Ton and Top at the entry station, and the disembarking difference value is based on the time difference between the disembarking times Tdn and Tdp at the exit station. In contrast, in the present embodiment, the route time difference Diff is derived from the following formula (5): Diff=(Ton-Tenter)+(Texit-Tdn)……(5) Here, Tenter is the entry time in the trip data D5, and Texit is the exit time in the trip data D5. In this embodiment, when estimating a new route D7 corresponding to a certain estimated current route D6, the trip data D5 used to estimate the estimated current route D6 is referenced again.

[0107] In the first embodiment, the route time difference Diff is derived from the boarding time Top and disembarking time Tdp of the estimated current route D6. Therefore, it is possible to count passengers using the same route and simultaneously assign the counted passengers to the route candidate of the new route D7. In contrast, in this embodiment, the route information is derived from the entry time Tenter and the exit time Texit, which are information specific to each passenger. Therefore, the aggregation process as in the first embodiment is not performed, and the estimation process of the new route D7 is performed for each trip.

[0108] In this embodiment, the boarding difference value indicates the delay of the boarding time Ton at the entrance station of the new route D7 relative to the entrance time Tenter in the trip data D5. The greater the delay, the greater the boarding difference value and the greater the route time difference Diff. The disembarking difference value indicates the advancement of the disembarking time Tdn at the exit station of the new route D7 relative to the departure time Tarr in the trip data D5. The greater the advancement, the greater the disembarking difference value and the greater the route time difference Diff.

[0109] 14 shows an estimated current route D6 according to the second embodiment. The estimation method for the estimated current route D6 is the same as that in the first embodiment, but in this embodiment, the entry time Tenter and the exit time Tarr of the trip data D5, which is the original data used to estimate the estimated current route D6, are added to the information indicating the estimated current route D6. The three trips shown in FIG. 14 are the same as the three trips shown in the top three rows of FIG. 6, but because their entry times and exit times are different from each other, there is no need to perform aggregation processing in preprocessing.

[0110] 15 and 16 show the new route estimation process S130 in this process according to the second embodiment. First, as in the first embodiment, the new route estimation unit 15 selects a timetable type to be estimated (S31), and refers to the timetable data selected as the timetable data to be estimated from the new timetable data D2 and the corresponding data of the estimated current route D6 (S32). Next, the new route estimation unit 15 extracts one estimated current route D6 as source data for the estimation process (S133). The new route estimation unit 15 extracts route candidates for the new route D7 by referring to the entry time Tenter and the exit time Tarr added to the estimated current route D6 (S136).

[0111] As shown in FIG. 17, when extracting route candidates, the new route estimation unit 15 first extracts a predetermined number k of trains as candidates for trains that passengers board at the entrance station and a predetermined number k of trains as candidates for trains that passengers disembark at the exit station. The candidates for boarding trains are the predetermined number k of trains that depart from the entrance station after the entrance time Tenter, in order of earliest departure time. The candidates for disembarking trains are the predetermined number k of trains that arrive at the exit station before the departure time Tarr, in order of latest arrival time. The predetermined number k is set, for example, to 3 to 7 (5 in the illustrated example).

[0112] Next, the new route estimation unit 15 extracts a route that satisfies the above-mentioned boarding train conditions and disembarking train conditions. If such a route can be extracted, the new route estimation unit 15 excludes from the extracted routes any route whose number of transfers exceeds the number of transfers on the estimated current route D6. That is, among the routes that satisfy the boarding train conditions and the disembarking train conditions, only those with the number of transfers equal to or less than the number of transfers on the estimated current route D6 are ultimately extracted as route candidates for the new route D7. If speed is improved in the new train timetable, such routes will exist. In the example shown in Figure 17, it is assumed that the number of transfers on the current estimated route D6 is 0, and routes with one or more transfers are excluded.

[0113] In this embodiment, the subsequent estimation method changes depending on whether or not route candidates exist. If one or more candidates exist (S137: Y), the route time difference Diff for each route candidate is calculated (S140). Although the method for deriving the route time difference Diff differs from that of the first embodiment, the subsequent processes S138, S139, and S141 to S143 are the same as the processes S38, S39, and S41 to S43 of the first embodiment. Note that in this process, since there is one passenger, when allocating passengers to multiple route candidates, the number of passengers allocated to each route candidate will be less than one.

[0114] If there is no candidate (S137: N), the new route estimation unit 15 determines whether the entry time is in the morning (S151), as shown in Fig. 16. "Morning" refers to a time period when congestion occurs due to demand for commuting to work or school, for example, the time period between 7:00 and 9:00 AM. If the entry time is in the morning (S151: Y), the new route estimation unit 15 extracts one route from the new train timetable as a route candidate (S152). In this extraction, as shown in FIG. 18, the new route estimation unit 15 extracts the route with the latest boarding time Ton at the entry station from among routes in which the disembarking time Tdn is before the departure time Tarr and the number of transfers is less than or equal to the number of transfers of the current estimated route D6. Note that, since this is a second-best route search when no route candidate has been extracted, it is permissible to extract a route in which the boarding time Ton at the entry station is earlier than the entry time Tenter.

[0115] If the entry time is not in the morning (S151: N), the new route estimation unit 15 extracts one route from the new train timetable as a route candidate (S153). In this extraction, as shown in Fig. 19, the new route estimation unit 15 extracts the route with the earliest disembarking time Tdn at the exit station from among routes whose boarding time Ton is later than the entry time Tenter and whose number of transfers is less than or equal to the number of transfers of the current estimated route D6. Note that, since this is a second-best route search when no route candidate has been extracted, it is permissible to extract a route whose disembarking time Ton at the exit station is later than the departure time Tarr.

[0116] When one route candidate is extracted in this manner, as shown in Figure 15, the new route estimation unit 15 estimates the single route candidate as a new route D7 (S138), in the same way as when the route time difference Diff is zero or when the number of route candidates extracted in the extraction process S136 is one, and assigns one passenger on the estimated current route D6 as one passenger using the new route D7 (S139).

[0117] The process of estimating new routes D7 is repeated until the estimation of new routes D7 and the allocation of passengers are completed for all estimated current routes D6 (S144: N → S33). When the estimation of new routes D7 for all estimated current routes D6 is completed (S144: Y), the same process is repeated while selecting each schedule type to be estimated one by one until the estimation process and the counting process are completed for all schedule types (S45: N → S31). When the process for all schedule types is completed (S45: Y), the process returns to the main routine (see Figure 10).

[0118] In the second embodiment described above, the trip data D5 specific to each passenger is again used to derive the route time difference Diff, and the new route D7 is estimated based on this route time difference Diff. This makes it easier to reflect the route preferences and behavior patterns of each individual passenger in the new route D7. Furthermore, if no route candidates are extracted under the predetermined conditions, the estimation method for the new route D7 is changed depending on the time of day. This allows passenger preferences that change depending on the time of day to be reflected in the new route D7, improving the accuracy of passenger flow estimation.

[0119] In this case, in the morning, the route where the disembarkation time is before the departure time and the boarding time Ton is the latest is extracted. Passengers are thought to prefer starting their morning activities later rather than earlier, and it is thought that they do not want to arrive at their destination late due to daytime business. Since routes are extracted taking into account such passenger preferences and situations, the accuracy of passenger flow estimation is improved.

[0120] For times other than the morning, the system extracts a route that has a boarding time after the entrance time and the earliest disembarking time. Passengers are likely to prefer to arrive at their destination (e.g., home) earlier when traveling during the day or at night (e.g., on the way home). By extracting routes that take such passenger preferences into consideration, the accuracy of passenger flow estimation is improved.

[0121] [Modification] The embodiments of the present invention have been described above, but the above configurations can be appropriately changed, added, and / or deleted within the scope of the present invention. (A) In the above embodiment, the pre-processing is performed by the train diagram simulation device 10 that performs the main processing. However, the pre-processing may be performed by a device separate from the device that performs the main processing. The train diagram simulation device 10 only needs to have the function of performing the main processing, and be configured so that the data of the estimated current route D6 that has been calculated by the pre-processing can be referenced in the main processing. In this case, the storage unit 11 does not necessarily need to store the current diagram data D1. Furthermore, only the calculation process of the pre-processing may be performed by the device that performs the main processing.

[0122] (B) In the above embodiment, when multiple route candidates exist for estimating new route D7, passengers are allocated to the multiple route candidates at an allocation rate r according to the route time difference Diff of each route candidate. However, the allocation rate r may depend on a parameter other than the route time difference Diff. For example, if the route candidates include a route that uses a paid express train, the allocation rate corresponding to this route may be corrected to decrease from the allocation rate determined according to the route time difference by processing such as multiplying it by a weighting coefficient determined by whether or not a seat charge is included.

[0123] (C) In the above embodiment, ticket gate passage data is used as data indicating the passenger's movement history. However, the estimated movement path can also be estimated using data other than ticket gate passage data. For example, instead of ticket gate passage data, location information acquired from a terminal carried by the passenger, image information acquired from a camera installed near a station ticket gate or within the station, ticket reservation information by the passenger, etc. may be used. [Explanation of symbols]

[0124] 5 Automatic ticket gates 10 Train schedule simulation device 12 Data Acquisition Section 13 Current route estimation unit 14. Counting Unit 15 New Route Estimation Unit 16 Passenger Number Estimation Department D1 Current timetable data D2 New Diagram Data D3 Ticket gate passage data D6 Estimated current route D7 New Route Diff Route time difference Top Estimated boarding time at the entrance station of the current route Ton Boarding time at the entrance station of the proposed new route Tdp Estimated time of disembarkation at the exit station on the current route Tdn: The time of disembarkation at the departure station of the proposed new route

Claims

1. a data acquisition unit that acquires an estimated current route that indicates which train each passenger used to travel from an entrance station to an exit station in the current train schedule, which route is estimated based on current train schedule data that indicates a current train schedule and ticket gate passage data that includes the entrance records of each passenger and the corresponding exit records collected from automatic ticket gates during operation of the current train schedule, and new train schedule data that indicates a new train schedule; a new route estimation unit that estimates a new route indicating which train each of the passengers will use to travel from the entry station to the exit station in the new train schedule based on the estimated current route and the new train schedule data; A train schedule simulation device comprising:

2. The information representing the estimated current route or the new route includes a boarding time indicating a departure time from the entrance station of a train that each passenger is estimated to board at the entrance station, a disembarking time indicating an arrival time at the exit station of a train that each passenger is estimated to disembark at, and a number of transfers indicating the number of train changes required to move from the entrance station to the exit station. The train diagram simulation device according to claim 1.

3. the new route estimation unit estimates the new route based on the boarding time and the disembarking time of the estimated current route; The train diagram simulation device according to claim 2.

4. the new route estimation unit estimates the new route so as to reduce a time difference between the boarding time of the new route and the boarding time of the estimated current route; 4. The train diagram simulation device according to claim 2 or 3.

5. the new route estimation unit estimates the new route so as to reduce a time difference between the alighting time on the new route and the alighting time on the estimated current route; The train diagram simulation device according to any one of claims 2 to 4.

6. the new route estimation unit estimates the new route such that the drop-off time on the new route is earlier than the drop-off time on the estimated current route; The train diagram simulation device according to claim 5.

7. the new route estimation unit estimates the new route such that the number of transfers on the new route is equal to or less than the number of transfers on the estimated current route; The train diagram simulation device according to any one of claims 2 to 6.

8. If a route identical to the estimated current route exists in the new train diagram, the new route estimation unit estimates the identical route as the new route. The train diagram simulation device according to any one of claims 1 to 7.

9. the new route estimation unit extracts a plurality of route candidates as candidates for the new route when the new train timetable does not include a route identical to the estimated current route; The train diagram simulation device according to claim 1.

10. the new route estimation unit extracts, as the route candidate, a route whose boarding time falls within a predetermined first time range based on the boarding time of the estimated current route and whose disembarking time falls within a predetermined second time range from the disembarking time of the estimated current route; The train diagram simulation device according to claim 9.

11. the new route estimation unit calculates a route time difference representing a time difference between each of the extracted plurality of route candidates and the estimated current route, and allocates the passengers to the plurality of route candidates according to the calculated route time difference; The smaller the route time difference of the route candidate, the higher the allocation rate of the passengers to the route candidate. The train diagram simulation device according to claim 9 or 10.

12. The larger the time difference between the boarding time of the route candidate and the boarding time of the estimated current route, the larger the route time difference. The train diagram simulation device according to claim 11.

13. When the drop-off time of the route candidate is later than the drop-off time of the estimated current route, the route time difference increases as the time difference between the drop-off time of the route candidate and the drop-off time of the estimated current route increases. The train diagram simulation device according to claim 11 or 12.

14. Passengers with the same estimated current route are counted in advance, the new route estimation unit simultaneously allocates a plurality of passengers counted as having the same estimated current route to the plurality of route candidates at the allocation rate according to the route time difference; The train diagram simulation device according to any one of claims 11 to 13.

15. The earlier the boarding time of the route candidate is compared to the entry time used in estimating the estimated current route, the larger the route time difference is, and the later the disembarking time of the route candidate is compared to the exit time used in estimating the estimated current route, the larger the route time difference is. The train diagram simulation device according to claim 11.

16. The train information processing device further includes a passenger number estimation unit that estimates the number of passengers of each train in the new train timetable based on the new route estimated by the new route estimation unit. The train diagram simulation device according to any one of claims 1 to 14.

17. a data acquisition unit that acquires an estimated current route that is estimated based on current train schedule data that indicates the current train schedule and data that indicates the movement history of each passenger, and that indicates which train each passenger used to move from an entry station to an exit station in the current train schedule, and new train schedule data that indicates a new train schedule; a new route estimation unit that estimates a new route indicating which train each of the passengers will use to travel from the entry station to the exit station in the new train schedule based on the estimated current route and the new train schedule data; A train schedule simulation device comprising:

18. A data acquisition unit of the train timetable simulation device acquires an estimated current route that indicates which train each passenger used to travel from an entrance station to an exit station in the current train timetable, and new timetable data that indicates a new train timetable, which is estimated based on current timetable data that indicates the current train timetable and ticket gate passage data that includes the entrance records of each passenger and the corresponding exit records collected from automatic ticket gates during the operation of the current train timetable, A new route estimation unit of the train diagram simulation device estimates a new route indicating which train each of the passengers will use in the new train diagram to travel from the entry station to the exit station, based on the estimated current route and the new train diagram data. A train schedule simulation method comprising:

19. A method for simulating a train diagram according to claim 18, Train schedule simulation program.

Images