Information processing apparatus, information processing method, and program

Abstract

To allow for presenting a prediction result that contributes to an improvement in vehicle occupancy rate.SOLUTION: An information preprocessing device is provided, comprising a display control unit configured to control a display unit to display a map including a search result of a recommended route obtained on the basis of a score of each of multiple routes based on passenger demand prediction data for commercial vehicles. The technique is applicable, for example, to an information processing device or the like for displaying a passenger demand prediction result for taxis.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present technology relates to an information processing apparatus, an information processing method, and a program, and more particularly to an information processing apparatus, an information processing method, and a program capable of presenting prediction results contributing to an increase in the occupancy rate.

Background Art

[0002] In the taxi industry, efforts to predict the demand for taxis and conduct more effective business operations have been active (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a system for predicting the demand for taxis, it is important what prediction results are presented in order to increase the occupancy rate.

[0005] The present technology has been made in view of such a situation, and is capable of presenting prediction results contributing to an increase in the occupancy rate.

Means for Solving the Problems

[0006] An information processing apparatus according to one aspect of the present technology is based on ride demand prediction data of business vehicles 、 a plurality of routes that each in the path through which score the total score obtained by summing up and 、 based on the operation data of each of the business vehicles including empty vehicle information, among the plurality of routes a display control unit is provided that controls a display unit to display a map including a search result of a recommended route.

[0007] In an information processing method according to one aspect of the present technology, an information processing apparatus controls a display unit to display a map including a search result of a recommended route based on ride demand prediction data of a business vehicle and operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes 、 a plurality of routes that each in the path through which score the total score obtained by summing up and 、 operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes among the plurality of routes operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes

[0008] A program according to one aspect of the present technology causes a computer to control a display unit to display a map including a search result of a recommended route based on ride demand prediction data of a business vehicle and operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes 、 a plurality of routes that each in the path through which score the total score obtained by summing up and 、 operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes among the plurality of routes operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes

[0009] In one aspect of the present technology, a map including a search result of a recommended route is displayed based on ride demand prediction data of a business vehicle and operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes 、 a plurality of routes that each in the path through which score the total score obtained by summing up and 、 operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes among the plurality of routes operation data of each of the business vehicles including empty vehicle information for each of a plurality of routes

[0010] Note that the program can be transmitted via a transmission medium or provided by being recorded on a recording medium

[0011] Note that an information processing apparatus according to one aspect of the present technology can be realized by causing a computer to execute a program

[0012] Also, a program to be executed by a computer to realize an information processing apparatus according to one aspect of the present technology can be transmitted via a transmission medium or provided by being recorded on a recording medium

[0013] The information processing apparatus may be an independent apparatus or an internal block constituting one apparatus.

Effect of the Invention

[0014] According to one aspect of the present technology, it is possible to present a prediction result that contributes to an improvement in the boarding rate.

[0015] Note that the effects described here are not necessarily limited, and may be any of the effects described in the present disclosure.

Brief Description of the Drawings

[0016]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

Figure 30

Figure 31

Figure 32

Figure 33

Figure 34

Figure 35

Figure 36

Figure 37

Figure 38

Figure 39

Figure 40

Figure 41

Figure 42

Figure 43

Figure 44

Figure 45

Embodiments for Carrying Out the Invention

[0017] Hereinafter, embodiments for carrying out the present technology (hereinafter referred to as embodiments) will be described. The description will be made in the following order. 1. Configuration example of the prediction system 2. Screen examples of the demand prediction application 3. Block diagram 4. Real vehicle sequence data generation process 5. Learning prediction process 6. Unknown area cluster classification process 7. Combined display of area AR 8. Display of demand direction and frequency 9. Display of pinpoint prediction 10. Display of waiting time prediction 11. Display of length prediction 12. Display of riding distance prediction 13. Display of fare prediction 14. Learning of boarding location 15. Learning of alighting location 16. Learning of boarding location 17. Riding demand guide by sound 18. Recommended route presentation process 19. Display of prohibited boarding area guidance 20. Display of attachment location 21. Train time display 22. Reverse boarding point display 23. Display of demand prediction classification for vehicle allocation / running / waiting 24. Display of fare prediction 25. Display of real - time empty vehicle number 26. Display of daily business evaluation 27. Display of additional information considering distance and direction 28. Display of information according to traveling direction 29. Example of computer configuration

[0018] <1. Configuration example of prediction system> Figure 1 shows a configuration example of an embodiment of a prediction system to which this technology is applied.

[0019] The prediction system 1 in Figure 1 is composed of a plurality of taxis 11 and a server (information processing device) 12, and is a system that predicts the demand for rides in the business area of the taxis 11 based on the data acquired from the taxis 11.

[0020] The taxi 11 is a business vehicle that travels in a predetermined business area and picks up passengers. The taxi 11 is equipped with a fare meter 21, a vehicle management device 22, and a terminal device 23.

[0021] The fare meter 21 accepts operations of "occupied" and "unoccupied" by the driver. "Occupied" represents the state of driving with passengers on board, and "unoccupied" represents the state of driving without passengers on board. When in the "occupied" state, the fare meter 21 calculates the fare (passenger fare) according to at least one of the driving time or distance, and displays it on a predetermined display unit.

[0022] The vehicle management device 22 generates vehicle dynamic log data that records in time series at a predetermined time interval the position (route) where the taxi 11 has traveled, the status of "occupied" or "unoccupied", etc., and transmits it to the server 12 via a predetermined network. The status of "occupied" or "unoccupied" is obtained from the fare meter 21.

[0023] The terminal device 23 is composed of an information processing device such as a smartphone or a tablet terminal, for example. An application program (hereinafter, also simply referred to as a demand prediction app) that displays a ride demand prediction on a display using the ride demand prediction data transmitted from the server 12 is stored in the terminal device 23.

[0024] The demand prediction app is activated and executed on the terminal device 23 by the driver's operation. The demand prediction app receives the ride demand prediction data transmitted from the server 12 via a predetermined network, and based on the received ride demand prediction data, displays on the display a prediction result of predicting the ride demand on the map. A specific display example of the prediction result of predicting the ride demand will be described later with reference to FIG. 2 and the like.

[0025] The server 12 obtains vehicle dynamic log data from a plurality of taxis 11 via the network. Then, the server 12 generates ride demand prediction data using the obtained large number of vehicle dynamic log data, and transmits it to each of the plurality of taxis 11 via the network.

[0026] The network connecting the server 12, the vehicle management device 22, and the terminal device 23 is composed of, for example, a mobile communication network such as a so-called 3G line or 4G line, the Internet, a public telephone line network, a satellite communication network, and the like.

[0027] The driver of the taxi 11 drives the taxi 11 to obtain passengers while referring to the predicted boarding demand displayed on the display of the terminal device 23 by the demand prediction application.

[0028] <2. Screen example of demand prediction application> FIG. 2 shows an example of a demand prediction screen displayed by the demand prediction application on the terminal device 23.

[0029] On the demand prediction screen of FIG. 2, a map 41 is displayed, and a current location mark 61, zoom buttons 62, demand prediction meshes 63, a setting button 64, etc. are superimposed on the map 41.

[0030] In addition, on the demand prediction screen, a prediction time setting area 42 is provided in an area different from the display area of the map 41, and the prediction time setting area 42 includes a prediction time display 71 and prediction time change buttons 72A and 72B.

[0031] The current location mark 61 represents the current location of the taxi 11. The zoom buttons 62 are operated when enlarging or reducing the scale of the map 41.

[0032] The demand prediction meshes 63 are configured by arranging a plurality of areas AR in a matrix. The area AR represents one area obtained by dividing the demand prediction mesh 63 into a grid. In the example of FIG. 2, 28 areas AR of 4x7 are arranged in a part of the area on the map 41, but the area AR may be superimposed on all areas on the map 41.

[0033] Each area AR of the prediction mesh 63 is displayed with a color or density corresponding to the degree of ride demand based on the ride demand prediction data transmitted from the server 12. For example, in FIG. 2, an area AR with a high density represents an area AR with a high ride demand, and an area AR with a low density represents an area AR with a low ride demand.

[0034] The setting button 64 is operated when making various settings regarding the display of the demand prediction screen, such as selecting items that can be displayed on the demand prediction screen and the display order. Details of each item that can be displayed on the demand prediction screen will be described later.

[0035] The prediction time display 71 in the prediction time setting area 42 displays the corresponding time of the ride demand prediction being displayed by the demand prediction mesh 63. That is, the demand prediction for the time displayed on the prediction time display 71 is displayed on the demand prediction mesh 63. When the prediction time display 71 is tapped, it is reset to the current time. The prediction time change buttons 72A and 72B are operated when advancing or rewinding the prediction time of the prediction time display 71 in a predetermined unit (for example, 10 minutes).

[0036] As described above, the ride demand prediction app of the terminal device 23 receives the ride demand prediction data transmitted from the server 12, and based on the received ride demand prediction data, displays the demand prediction mesh 63 that predicts the ride demand on the map 41 on the display as the prediction result.

[0037] In the example of FIG. 2, each area AR of the demand prediction mesh 63 is displayed with different colors and densities according to the degree of ride demand. However, as shown in FIG. 13 described later, the prediction results of the number of rides can also be displayed together.

[0038] <3. Block Diagram> Next, the detailed configurations of each device mounted on the taxi 11 and the server 12 will be described.

[0039] FIG. 3 is a block diagram showing a configuration example of the server 12, the fare meter 21, the vehicle management device 22, and the terminal device 23.

[0040] The fare meter 21 receives the operation of "occupied car" or "empty car" by the driver, and displays the status of "occupied car" or "empty car" and the fare (freight) on a predetermined display unit. The fare meter 21 supplies the status of "occupied car" or "empty car" to the vehicle management device 22.

[0041] The vehicle management device 22 includes a position detection unit 101, a speed detection unit 102, a control unit 103, a storage unit 104, and a communication unit 105.

[0042] The position detection unit 101 is composed of, for example, a GPS (Global Positioning System) receiver or the like, receives the positioning signal broadcast by the positioning satellite, and detects the current position of the taxi 11. In addition, the position detection unit 101 includes a gyro sensor, a geomagnetic sensor, etc., and detects the traveling direction of the taxi 11.

[0043] The speed detection unit 102 is composed of a speed sensor, an acceleration sensor, etc., and detects the moving speed of the taxi 11. Note that the speed detection unit 102 may detect the moving speed of the taxi 11 by acquiring a measurement value from a speed sensor that detects the rotational speed of the wheels of the taxi 11.

[0044] The control unit 103 is composed of, for example, a CPU (Central Processing Unit), a RAM (Random Access Memory), etc., reads out the operation control program stored in the storage unit 104, and controls the operation of the entire vehicle management device 22 according to the operation control program. Specifically, the control unit 103 acquires data from each of the fare meter 21, the position detection unit 101, and the speed detection unit 102 at regular time intervals, generates vehicle dynamic log data, and stores it in the storage unit 104. In addition, the control unit 103 transmits the vehicle dynamic log data stored in the storage unit 104 to the server 12 via the communication unit 105 regularly or irregularly at a preset timing.

[0045] The memory unit 104 is composed of, for example, a hard disk, a ROM (Read Only Memory), a RAM, and an NVRAM (Non-Volatile RAM), etc., and stores vehicle dynamic log data. The communication unit 105 performs predetermined communication with the server 12 according to the control of the control unit 103. The communication unit 105 is composed of a network interface that performs network communication via a predetermined network.

[0046] The server 12 includes a control unit 121, a memory unit 122, and a communication unit 123.

[0047] The control unit 121 is composed of, for example, a CPU, a RAM, etc., reads out the operation control program stored in the memory unit 122, and controls the operation of the entire server 12 according to the operation control program.

[0048] Functionally, the control unit 121 includes at least a data generation unit 131, a learning unit 132, and a prediction unit 133, and predicts the ride demand for each area AR on the map 41 by machine learning. As the machine learning method, for example, any method such as the k-means method, a self-organizing map (SOM), a neural network, an HMM (Hidden Markov Model), etc. can be selected.

[0049] The data generation unit 131 causes the memory unit 122 to store the vehicle dynamic log data acquired from each of the vehicle management devices 22 of a plurality of taxis 11 via the communication unit 123.

[0050] FIG. 4 shows an example of the vehicle dynamic log data generated by the vehicle management device 22 of the taxi 11 and transmitted to the server 12.

[0051] The vehicle management device 22 generates and accumulates vehicle dynamic log data at predetermined time intervals (for example, at 1-minute intervals).

[0052] As shown in FIG. 4, the items generated as vehicle dynamic log data include a company ID for identifying the company to which taxi 11 belongs, a wireless ID for identifying the vehicle of taxi 11, a crew ID for identifying the driver riding in taxi 11, a status time representing the generation time of the status, the latitude and longitude which are the location information of taxi 11, the direction and speed indicating the traveling speed and the traveling direction of taxi 11, and the status of "occupied" or "empty".

[0053] The data generation unit 131 generates in-vehicle data which is data related to the occupied vehicle from the vehicle dynamic log data stored in the storage unit 122.

[0054] FIG. 5 shows an example of the generation of in-vehicle data.

[0055] The in-vehicle data is data obtained by extracting information related to the boarding of taxi 11 from the vehicle dynamic log data, and is generated from the information of the boarding change point where the status changes from "empty" to "occupied" and the alighting change point where the status changes from "occupied" to "empty" in the vehicle dynamic log data.

[0056] The in-vehicle data includes, for example, as shown in FIG. 5, items such as ID, boarding time, departure location, arrival location, boarding duration, boarding distance, and fare.

[0057] The ID is data obtained by combining the company ID, wireless ID, and crew ID of the vehicle dynamic log data.

[0058] For the boarding time, the time between the status time of "empty" and the status time of "occupied" at the boarding change point is calculated and recorded.

[0059] For the departure location, the latitude and longitude between the latitude and longitude of "empty" and the latitude and longitude of "occupied" at the boarding change point are calculated and recorded.

[0060] For the arrival location, the latitude and longitude between the latitude and longitude of "empty" and the latitude and longitude of "occupied" at the alighting change point are calculated and recorded.

[0061] For the boarding time, the time (unit: minutes, for example) from the boarding time to the time between the "empty vehicle" status time and the "occupied vehicle" status time at the alighting change point is calculated and recorded.

[0062] For the boarding distance, the distance from the departure point to the arrival point (unit: km, for example) is calculated and recorded.

[0063] The fare is calculated and recorded according to the regulations of taxi fares based on the boarding time and the boarding distance.

[0064] Note that the calculation method of each item of the occupied vehicle data is not limited to the method described above, and other methods may also be used. For example, each of the above items may be calculated from the first and last vehicle dynamic log data with the status of "occupied vehicle". Also, the information on the fare and the boarding distance may be obtained from the vehicle management device 22 as part of the vehicle dynamic log data, rather than being calculated from the positions of the boarding change point and the alighting change point.

[0065] Based on a large number of occupied vehicle data generated from the vehicle dynamic log data of the vehicle management devices 22 of a large number of taxis 11, the data generation unit 131 generates real vehicle sequence data, which is time-series data representing the number of boardings in a predetermined time unit (10 minutes), for each area AR. For example, the data generation unit 131 generates real vehicle sequence data, which is time-series data obtained by counting the number of boardings every 10 minutes, for each area AR.

[0066] FIG. 6 shows an example of real vehicle sequence data for three areas AR, namely area 1223, area 1224, and area 1225, among a plurality of areas AR obtained by dividing the business area of the taxi 11.

[0067] The horizontal axis of the in-vehicle sequence data represents the date and time, and the vertical axis represents the number of passengers. The in-vehicle sequence data shown in Fig. 6 is data for 8 days, but the creation period of the in-vehicle sequence data can be set to any period, such as one week, one month, one year, etc. For example, if the creation period of the in-vehicle sequence data is set to one week, fluctuations due to the day of the week can be captured. If it is set to a long period such as several months or one year, seasonal fluctuations such as the end of the year and the beginning of the year, Golden Week, and summer vacation can also be captured in addition to fluctuations due to the day of the week.

[0068] Taking the example of the in-vehicle sequence data of Area 1223, for example, the number of in-vehicle data whose boarding time is included between 10:00 and 10:10 on March 21, 2017 and whose departure point is located within Area 1223 is counted as the number of passengers. The count result becomes the in-vehicle sequence data of Area 1223 from 10:00 to 10:10 on March 21, 2017. The same process is calculated for the entire period of the acquired in-vehicle data, and the in-vehicle sequence data of Area 1223 is generated.

[0069] Returning to Fig. 3, the learning unit 132 generates a predictor for predicting the boarding demand by learning using a large number of long-term in-vehicle sequence data generated based on the in-vehicle data acquired from a large number of vehicle management devices 22 of taxis 11.

[0070] The prediction unit 133 predicts the boarding demand at a predetermined time or time zone using the predictor generated by the learning unit 132. The prediction result of the prediction unit 133 is transmitted to the terminal device 23 as boarding demand prediction data.

[0071] The storage unit 122 stores the vehicle dynamic log data acquired from each of the vehicle management devices 22 and the in-vehicle sequence data generated from the vehicle dynamic log data. The in-vehicle data, which is intermediate data for generating the in-vehicle sequence data from the vehicle dynamic log data, may also be stored in the storage unit 122.

[0072] The communication unit 123 performs predetermined communication with the vehicle management device 22 and the terminal device 23 in accordance with the control of the control unit 121. The communication unit 123 is composed of a network interface that performs network communication via a predetermined network.

[0073] The terminal device 23 includes a control unit 141, an operation unit 142, a display unit 143, a communication unit 144, a speaker 145, and a microphone 146.

[0074] The control unit 141 is composed of, for example, a CPU, a RAM, etc., and controls the operation of the entire terminal device 23 according to an operation control program stored in a storage unit (not shown). For example, the control unit 141 executes a demand prediction application based on the operation of the driver who is the user. Then, the control unit 141 also functions as a display control unit that controls the display unit 143, and causes the display unit 143 to display the execution result of the demand prediction application, for example, the demand prediction screen in FIG. 2.

[0075] The operation unit 142 is composed of a plurality of operation buttons provided on the terminal device 23, a touch panel superimposed on the display unit 143, etc., receives the operation of the user, and supplies an operation signal corresponding to the received operation to the control unit 141.

[0076] The display unit 143 is composed of, for example, an LCD (Liquid Crystal Display), etc., and displays predetermined information such as the demand prediction screen in FIG. 2.

[0077] The communication unit 144 performs predetermined communication with the server 12 in accordance with the control of the control unit 141. The communication unit 144 is composed of a network interface that performs network communication via a predetermined network.

[0078] The speaker 145 outputs sounds such as electronic sounds, sound effects, and message voices. The microphone 146 detects the voice uttered by the user or collects ambient sounds.

[0079] Server 12, fare meter 21, vehicle management device 22, and terminal device 23 are configured as described above.

[0080] Hereinafter, the details of the processes executed by each of server 12, vehicle management device 22, and terminal device 23 will be described.

[0081] <4. Real vehicle sequence data generation process> First, with reference to the flowchart of FIG. 7, the real vehicle sequence data generation process by server 12 will be described. This process can be executed at a predetermined timing such as periodically or irregularly.

[0082] First, in step S1, data generation unit 131 of server 12 acquires (receives) vehicle movement log data transmitted via the network from each of the vehicle management devices 22 of a plurality of taxis 11. Each vehicle management device 22 can transmit the vehicle movement log data to server 12 individually at an arbitrary timing, and it is not necessary for them to be simultaneous.

[0083] In step S2, data generation unit 131 generates real vehicle data from the acquired vehicle movement log data. The real vehicle data includes, for example, data calculated from the items of the vehicle movement log data such as boarding time and departure point, and external data added on the server 12 side such as fare. As external data, other than that, for example, date-related information related to dates such as day of the week, weekdays or holidays, event information related to events held on the data acquisition date in the corresponding area AR, weather information, etc. can be provided. By adding external data as real vehicle data, for example, it is possible to learn and predict the boarding demand for each situation according to the day of the week, the presence or absence of events, the weather, etc.

[0084] In step S3, data generation unit 131 generates real vehicle sequence data for each area AR based on a large number of real vehicle data generated from the vehicle management devices 22 of a large number of taxis 11, stores it in storage unit 122, and ends the real vehicle sequence data generation process.

[0085] <5. Learning and Prediction Processing> Next, with reference to the flowchart of FIG. 8, the learning and prediction processing for learning and predicting the ride demand using the actual vehicle sequence data for each generated area AR will be described. This processing can also be executed, for example, at a predetermined timing such as regularly or irregularly.

[0086] First, in step S21, the learning unit 132 of the server 12 extracts a representative area from among a plurality of areas AR obtained by dividing the business area of the taxi 11. The learning unit 132 selects a predetermined number of areas AR from among the plurality of areas AR in advance and designates them as representative areas. The representative areas may be determined randomly, or, for example, a user with knowledge, such as an area AR in the city center and an area AR in the suburbs, an area AR close to a station and an area AR far from a station, or an area AR with many stations and an area AR with few stations, may select them according to a predetermined criterion.

[0087] In step S22, the learning unit 132 performs two-stage clustering using the actual vehicle sequence data of each area AR extracted as a representative area. More specifically, the learning unit 132 executes a first clustering for clustering the plurality of extracted areas AR using a first parameter, and executes a second clustering for clustering the plurality of extracted areas AR using a second parameter.

[0088] For example, the learning unit 132 executes the first clustering using, as the first parameter, the average and variance of the number of rides per unit time (e.g., one day) within the area AR, and executes the second clustering using, as the second parameter, the waveform of the average number of rides per unit time (e.g., one day) within the area AR. Note that, for example, the k-means method or the like can be used as the clustering method.

[0089] Figures 9 and 10 show an example of the result of the first clustering in which a plurality of areas AR, which are representative areas, are clustered using the average and variance of the number of passengers as parameters.

[0090] Figure 9 shows the distribution of a plurality of areas AR extracted as representative areas, with the average on the horizontal axis and the variance on the vertical axis.

[0091] Figure 10 is a diagram showing the actual vehicle sequence data of a plurality of areas AR that are representative areas, for each cluster. The horizontal axis of Figure 10 represents time (from 0:00 to 24:00), and the vertical axis represents the number of passengers.

[0092] Since the actual vehicle sequence data basically shows similar characteristics for each time period (morning, noon, night, etc.) of a day, the clustering is performed using data obtained by dividing the actual vehicle sequence data into basic units (one day).

[0093] In Figures 9 and 10, a plurality of areas AR (and their actual vehicle sequence data) extracted as representative areas are classified into six clusters.

[0094] Figure 11 shows an example of the two-stage clustering result that summarizes the clustering result of the first clustering and the clustering result of the second clustering.

[0095] In Figure 11, the horizontal direction (column unit) represents the clustering result of the first stage, and the vertical direction (row unit) represents the clustering result of the second stage. The horizontal axis and the vertical axis of each graph arranged in a matrix are the same as those in Figure 10.

[0096] In FIG. 11, column numbers 1, 2, 3, ··· are the clustering results by the first clustering, and a plurality of vertically arranged areas AR are a set (area AR group) of areas AR where the average and variance of the number of passengers are similar. On the other hand, row numbers A, B, C, ··· are the clustering results by the second clustering, and are the results of further clustering each area AR group that is the clustering result of the first clustering with areas AR having similar average passenger number waveforms. The numbers in each matrix-shaped graph represent the number of area ARs classified into that cluster. For example, the number "468" in the graph of cluster D-2 with row number D and column number 2 represents that 468 area ARs among the representative areas are classified into cluster D-2. The average and variance of the number of passengers per unit time used as the first parameter represent the magnitude of the number of passengers per unit time and the magnitude of the change in the number of passengers within the unit time, and the waveform of the average number of passengers per unit time used as the second parameter represents the tendency of the change in the number of passengers over time within the unit time.

[0097] Note that the second-stage clustering may be performed individually for each result of the first-stage clustering, or may be performed on the entire plurality of area ARs extracted as representative areas separately from the result of the first-stage clustering.

[0098] In this embodiment, for example, the business area of taxi 11 is divided into 4,400 area ARs, and further, among the 4,400, 2,200 area ARs, which is half, are extracted as representative meshes, and two-stage clustering is performed on the 2,200 area ARs, so that they are classified into 44 clusters.

[0099] Next, in step S23 of FIG. 8, the learning unit 132 adjusts the learning parameters of the predictor represented by the learning rate and the like for each cluster using the actual vehicle sequence data belonging to the cluster, and proceeds to step S24.

[0100] In step S24, the learning unit 132 uses the adjusted learning parameters and the on-vehicle sequence data of one or more areas AR belonging to the cluster to train, for each cluster, a predictor that predicts the boarding demand, and proceeds to step S25.

[0101] In step S25, the prediction unit 133 uses the predictor generated by the learning unit 132 to predict the boarding demand at a predetermined time in a predetermined area AR. For example, when predicting the boarding demand for the area AR belonging to the cluster C-4, the boarding demand at a predetermined time is predicted using the predictor for the cluster C-4.

[0102] The learning process in steps S21 to S24 and the prediction process in step S25 may be executed as continuous processes, or the prediction process in step S25 may be executed at a timing different from that of the processes in steps S21 to S24.

[0103] For example, the process in step S25 is executed following the process in step S24, and the boarding demand at a predetermined time for each area AR constituting the business area of the taxi 11 is calculated and stored in the storage unit 122. Then, in response to a request from the terminal device 23, the predicted demand data stored in the storage unit 122 is transmitted to the terminal device 23 as boarding demand prediction data.

[0104] Alternatively, at the timing when prediction data for the boarding demand at a predetermined time for one or more areas AR is requested from the terminal device 23, the process in step S25 is executed, and the processing result in step S25 is transmitted to the terminal device 23 as boarding demand prediction data.

[0105] As shown in FIG. 2, the demand prediction application of the terminal device 23 that has received the boarding demand prediction data displays a demand prediction mesh 63 with changed colors and densities according to the number of boardings in each area AR that is the prediction result.

[0106] According to the above learning and prediction process, among the 4,400 area ARs that make up the business area, each of the 2,200 area ARs extracted as representative areas is classified into a predetermined cluster, and the boarding demand can be predicted according to the classification result.

[0107] On the other hand, for the remaining 2,200 area ARs (hereinafter also referred to as unknown area ARs) that were not extracted as representative areas, at this stage, it is unknown which cluster they will be classified into, and the boarding demand cannot be predicted.

[0108] <6. Unknown Area Cluster Classification Process> Therefore, next, a process for predicting the boarding demand of unknown area ARs will be described.

[0109] Referring to the flowchart of FIG. 12, an unknown area cluster classification process for determining the cluster to which the unknown area AR belongs will be described. This process can be executed at a predetermined timing, such as regularly or irregularly.

[0110] First, in step S41, the learning unit 132 of the server 12 learns the characteristics (average, variance, shape) of the actual vehicle sequence data of each cluster classified in the learning and prediction process. In other words, the relationship between the actual vehicle sequence data of the 2,200 area ARs extracted as representative areas and the clusters is learned by the learning device.

[0111] In step S42, the prediction unit 133 of the server 12 inputs the actual vehicle sequence data of the unknown area AR into a classifier using the parameters obtained in the learning of step S41, and determines the cluster of the unknown area AR.

[0112] As described above, according to the unknown area cluster classification process, clustering of unknown area ARs other than the representative areas can be executed using a classifier generated by learning the relationship between the clustering result of the representative areas and the actual vehicle sequence data.

[0113] When the cluster of the unknown area AR can be discriminated, the prediction process of step S25 described above is executed using the predictor of the discriminated cluster, whereby the ride demand of the unknown area AR can be predicted.

[0114] Therefore, by executing both the learning prediction process of FIG. 8 and the unknown area cluster classification process of FIG. 12, it is possible to predict the ride demand of all 4400 areas AR that make up the business area of the taxi 11.

[0115] In the learning prediction process of FIG. 8, by performing the extraction of the representative area which is the process of step S21, the number of areas AR to be learned, in other words, the data volume of the actual vehicle sequence data can be reduced, so that the calculation load can be reduced and the cost and time required for ride demand prediction can be reduced.

[0116] Also, in step S22, by performing two-stage clustering using the actual vehicle sequence data of each area AR extracted as the representative area, the number of learners can be reduced, and the cost and time required for ride demand prediction can be reduced. Specifically, when two-stage clustering is not performed, learners for the number of areas AR (2200) extracted as the representative area are required, but by performing two-stage clustering and being able to classify into a predetermined number of clusters, the number of learners required for learning can be set to the number of clusters (44).

[0117] For the learning device of each cluster, it is possible to use all the actual vehicle sequence data of area AR classified into that cluster. That is, for example, when learning the ride demand prediction of area 1223, generally, learning is performed using only the actual vehicle sequence data obtained in that area 1223. In contrast, in this technology, when area 1223 is classified into cluster D-2, for example, and there are 468 area ARs belonging to cluster D-2, learning can be performed using the actual vehicle sequence data of 468 area ARs including area ARs other than area 1223. Therefore, for one learning device, learning can be performed with a data volume larger than the data volume that can be obtained in one area AR, so the prediction accuracy can be improved.

[0118] And for the unknown area AR that is not extracted as the representative area in the learning prediction process, through the unknown area cluster classification process, the cluster of the unknown area AR can be discriminated, and the ride demand of the unknown area AR can be predicted using the predictor of the discriminated cluster.

[0119] Note that in steps S23 and S24 described above, the adjustment of learning parameters and the learning of the predictor were performed using only the actual vehicle sequence data of each area AR extracted as the representative area. However, after the clusters of all the unknown area ARs included in the business area are discriminated, the actual vehicle sequence data of the unknown area AR may also be added to perform the adjustment of learning parameters and the learning of the predictor.

[0120] Therefore, according to the prediction system 1 in FIG. 1, learning and prediction can be performed more efficiently. Also, the prediction accuracy can be improved with a small data volume.

[0121] In the above-described learning prediction process and unknown area cluster classification process, regardless of weekdays, weekdays, or holidays, etc., cluster classification and learning were performed using the actual vehicle sequence data generated from all the vehicle movement log data obtained from the vehicle management devices 22 of a plurality of taxis 11.

[0122] However, the actual vehicle sequence data may be divided into categories such as day of the week, weekdays, holidays, or weather, and clustering classification and learning may be performed for each category. Thereby, the passenger demand can be predicted for each predetermined condition such as day of the week, weekdays, holidays, weather, presence or absence of events, etc., and the prediction result can be displayed on the display.

[0123] <7. Combined display of area AR> Hereinafter, various display examples in which the demand prediction application of the terminal device 23 displays the prediction result of the passenger demand on the display will be described.

[0124] FIG. 13 shows a first display example of a demand prediction screen displayed by the demand prediction application.

[0125] In the demand prediction screen shown in FIG. 2, the demand prediction mesh 63 was configured by arranging area ARs of the same rectangular size in a matrix. Also, the number of passengers in each area AR, which is the prediction result, was not displayed on the screen.

[0126] On the other hand, in the demand prediction mesh 63 of FIG. 13, the number of passengers in each area AR, which is the prediction result, is displayed within the area AR.

[0127] In addition, for a plurality of area ARs where the number of passengers in a plurality of adjacent area ARs is equal to or less than a predetermined threshold, the demand prediction application combines them into one area AR and displays the number of passengers. In the first display example of FIG. 13, a plurality of area ARs where the number of passengers in a plurality of adjacent area ARs is 10 or less are combined and displayed as one area AR. Specifically, a 2×2 area AR where the number of passengers when displayed in the same rectangular size is "4", "2", "2", and "1" is combined into one area AR and displayed as "9". Of course, depending on the number of adjacent passengers, there may be cases where they are not combined even if they are 10 or less.

[0128] When the predicted number of passengers is small, such as 0, 1, 2, etc., it is difficult to meet the demand. Therefore, the demand prediction app can display the demand prediction in units of area AR where the number of passengers is a certain value or more. This can improve the accuracy of the prediction and provide more useful information to the driver.

[0129] Note that the number of passengers displayed as the prediction result may be a value with a certain width, such as "10-13".

[0130] <8. Display of demand direction and frequency> FIG. 14 shows a second display example of the demand prediction screen displayed by the demand prediction app.

[0131] In FIG. 14, the display of colors and densities according to the degree of riding demand in each area AR is omitted.

[0132] FIG. 14 shows a display example that further displays detailed prediction results for the area AR (hereinafter referred to as the target area AR) that the driver pays attention to among the area ARs of the demand prediction mesh 63 superimposed on the map 41.

[0133] When the driver performs an operation to specify the target area AR, such as tapping (touching) a predetermined area AR from among the area ARs of the demand prediction mesh 63 superimposed on the map 41, the demand prediction app performs the display as shown in FIG. 14 for the specified target area AR.

[0134] In FIG. 14, a target area frame 211 with a width thicker than that of other area ARs is displayed for the target area AR specified by the driver. And arrows 212-1 to 212-8 are displayed outward from the target area frame 211. When not particularly distinguishing each of the arrows 212-1 to 212-8, they are simply referred to as arrows 212.

[0135] The direction of arrow 212 represents the moving direction of passengers boarding in the area of interest AR, and the length of arrow 212 represents the average moving distance of passengers boarding in the area of interest AR and moving in the direction of arrow 212. Also, the width of arrow 212 (the thickness in the direction perpendicular to the direction of the arrow) represents the boarding ratio in the direction indicated by that arrow 212 with respect to all directions.

[0136] Therefore, in the example of FIG. 14, among the passengers boarding in the area of interest AR, it shows that as for the boarding ratio, there are many passengers moving in the direction of arrow 212-3, and the passengers moving in the direction of arrow 212-4 have a long moving distance. Also, for example, among the passengers boarding in the area of interest AR, there are few passengers moving to arrow 212-2 or arrow 212-6, indicating that their moving distances are also short.

[0137] When determining an area AR for performing, for example, a so-called "cruise" (driving taxi 11 while looking for passengers), the driver can set a predetermined area AR of the demand prediction mesh 63 as the area of interest AR and display arrow 212 to search for an area AR where there are many passengers in the same direction as the direction the driver will return, etc.

[0138] The moving direction of passengers in each area AR can be predicted by learning including the information on the direction (travel direction) of the vehicle movement log data.

[0139] Note that the number of arrows 212 to be displayed, in other words, the number of predictions of the moving direction of passengers, may be a number other than 8 shown in FIG. 14. Also, the ratio of passengers moving in the direction of arrow 212 in all directions may be represented by a method other than representing it by the width of the arrow, for example, by differences in color or numerical notation.

[0140] <9. Display of pinpoint prediction> FIG. 15 shows a third display example of the demand prediction screen displayed by the demand prediction application.

[0141] FIG. 15 also shows a display example for displaying more detailed prediction results when the driver selects a predetermined area AR as the area of interest AR.

[0142] Among the areas AR obtained by dividing the business area into predetermined units, there are places where the boarding location is fixed, such as in front of a station or a taxi stand in front of a hotel, and the number of boardings may be relatively large compared to others.

[0143] When such a boarding location with a large number of boardings exists within the target area AR, the demand prediction application can pinpoint and predict and display the boarding location with a large number of boardings and the number of boardings at that boarding location separately from the number of boardings in the entire target area AR. Hereinafter, the boarding location with a large number of boardings identified within the target area AR is referred to as the pinpoint boarding location.

[0144] In FIG. 15, a pinpoint boarding location mark 221 representing the pinpoint boarding location is displayed at a predetermined position within the target area AR, and a boarding number display 222 for displaying the predicted number of boardings at the pinpoint boarding location mark 221 is displayed. In FIG. 15, "43" displayed within the target area frame 211 is the number of boardings for the entire target area AR. Among them, "29" in the boarding number display 222 indicates the number of boardings at the pinpoint boarding location "Shinagawa Station Takatori-guchi Taxi Stand" of the pinpoint boarding location mark 221. In this way, in addition to the number of boardings in the target area AR, by displaying the pinpoint boarding location and the predicted number of boardings there, the actual vehicle occupancy rate can be increased.

[0145] The pinpoint boarding location can be estimated by learning using actual vehicle data, rather than investigating one by one the places where the boarding location is fixed within the area AR.

[0146] Specifically, as shown by the black circles on the left side of FIG. 16, the past boarding positions of passengers can be grasped from the information on the departure points of the actual vehicle data. By learning the past boarding positions of passengers, as shown on the right side of FIG. 16, the estimated values of the boarding positions indicated by the black circles and the probabilities (likelihoods) of those boarding positions are calculated. The probability of the boarding position is represented by a number in the range of 0 to 1 and is displayed near the boarding position in FIG. 16. The demand prediction application can, for example, display the estimated values of the boarding positions where the probability of the boarding position is equal to or greater than a predetermined threshold (e.g., 0.8) as pinpoint boarding positions within the target area AR.

[0147] <10. Display of waiting time prediction> FIG. 17 shows a fourth display example of the demand prediction screen displayed by the demand prediction application.

[0148] FIG. 17 also shows a display example in which more detailed prediction results are displayed when the driver selects a predetermined area AR as the target area AR.

[0149] At locations where the boarding position is determined and the number of passengers boarding is large, such as in front of stations or in taxi stands in front of hotels, there is a method of obtaining passengers in a state where taxis 11 are waiting in line to pick up passengers at the boarding position, which is so-called "queue waiting". The disadvantage of queue waiting is, for example, that when a long line of taxis 11 forms at the taxi stand, it takes a long time from the time of lining up at the end of the long line of taxis 11 until the passengers are picked up.

[0150] Therefore, when queue waiting is being carried out at a boarding position (pinpoint boarding position) where the number of passengers boarding is large, the demand prediction application can display the time required for queue waiting, in other words, the time required to wait at the boarding position until the passengers are picked up.

[0151] Specifically, as shown in FIG. 17, when the pinpoint boarding position mark 221 in the area of interest AR is a place where passengers wait, the demand prediction app displays a pick-up start button 223 within the boarding number display 222 at the pinpoint boarding position mark 221. When the pick-up start button 223 is tapped (touched), the demand prediction app displays a pick-up display 224 indicating the time required for pick-up (pick-up time) when pick-up is performed. In the example of FIG. 17, "20 minutes" is displayed as the pick-up time.

[0152] For example, at the pinpoint boarding position, the driver can confirm the pick-up time and select a pick-up location by displaying the pick-up display 224. The pick-up time displayed in the pick-up display 224 may be a value with a certain width, such as "15 minutes - 20 minutes".

[0153] In the vehicle dynamics log data, since it is possible to detect the boarding change point where the status changes from "empty vehicle" to "occupied vehicle" and the state in which the taxi 11 is moving slowly slightly before that boarding change point, the pick-up operation of the taxi 11 can be detected. For example, driving at a speed equal to or lower than a predetermined speed (5 km / h or less) within a predetermined period or within a predetermined distance before the time of the boarding change point can be detected as a pick-up operation. Therefore, by learning the pick-up operation, the pick-up time at a predetermined boarding position can be predicted.

[0154] <11. Display of Longitude Prediction> FIG. 18 shows a fifth display example of the demand prediction screen displayed by the demand prediction app.

[0155] FIG. 18 also shows a display example in which more detailed prediction results are displayed when the driver selects a predetermined area AR as the area of interest AR.

[0156] Among the areas AR obtained by dividing the business area into predetermined units, there are areas AR where, for example, when the destination is Haneda Airport or Narita Airport, the proportion of rides with a long driving distance (equal to or longer than a predetermined distance) is high, and there are also pick-up locations. It is preferable that the driver can determine the possibility of having a long-distance passenger.

[0157] Therefore, as shown in FIG. 18, the demand prediction application can perform a long display 241 that displays the proportion of long-distance passengers in the area AR of interest separately from the number of rides in the entire area AR of interest.

[0158] In the long display 241, the proportion (ratio) of rides with a long driving distance among all the rides in the area AR of interest is displayed as the long degree. Also, in the long display 241, the driving distance is divided into a plurality of sections, and the proportion of rides for each divided section is displayed as a long section in a bar graph. The "All" bar graph shown in FIG. 18 indicates the proportion of rides for each section in the entire business area, and the "This" bar graph indicates the proportion of rides for each section in the area AR of interest.

[0159] As shown in FIG. 18, the long display 241 may display the long degree and long sections for the area AR of interest, or may be displayed along with the pinpoint pick-up location mark 221 to display the long degree and long sections for the pinpoint pick-up location.

[0160] Although the bar graph in the long display 241 of FIG. 18 shows the proportion of rides for each divided section (driving distance) by dividing the driving distance into a plurality of sections as the long section, it may also show the proportion of rides for each divided section (fare) by dividing the fare into a plurality of sections.

[0161] Also, the long display 241 may predict the ride demand for each time period and weather, and display the long degree and long sections specialized for a predetermined time period and weather.

[0162] The long degree and long sections can be predicted by learning including the items of driving distance and fare in the actual vehicle data.

[0163] <Display of Riding Distance Prediction> FIG. 19 shows a sixth display example of a demand prediction screen displayed by a demand prediction application.

[0164] FIG. 19 also shows a display example in which when a driver selects a predetermined area AR as a target area AR, a more detailed prediction result is displayed.

[0165] As shown in FIG. 19, when a predetermined area AR is selected as the target area AR, the demand prediction application can perform a riding distance display 251 that displays the average riding distance of passengers boarding in the target area AR and its confidence interval. The confidence interval represents the range in which the population mean (population average) is included with a predetermined confidence level.

[0166] In the riding distance display 251, it is displayed that the average riding distance of rides in the target area AR is "2.4 km", and for example, the confidence interval of the average riding distance at a 95% confidence level is "from 1.1 km to 3.7 km". The confidence level of the confidence interval is not limited to 95% and can be arbitrarily set such as 99%.

[0167] In this way, by displaying the average riding distance of the target area AR and its confidence interval, a driver can, for example, search for an area AR with a riding distance suitable for the remaining working hours, or search for an area AR with a long riding distance as a "flow-through" route.

[0168] As shown in FIG. 19, the riding distance display 251 may be displayed for the target area AR, or may be displayed in association with the pinpoint boarding position mark 221 to display the average riding distance and confidence interval for the pinpoint boarding position.

[0169] Alternatively, instead of the average riding distance and confidence interval, the average riding fare (passenger fare) and confidence interval may be shown.

[0170] Alternatively, instead of the average travel distance and the confidence interval, the average travel time and the confidence interval may be shown.

[0171] Also, the travel distance display 251 may predict the travel demand for each time zone and weather condition, and display the average travel distance and the confidence interval, the average fare and the confidence interval, or the average travel time and the confidence interval specialized for a predetermined time zone or weather condition.

[0172] <13. Display of Fare Prediction> Since the fare of taxi 11 is not determined until one rides it, there are also users who refrain from using it. The demand prediction app has a function of predicting and displaying the fare from the current location and the destination.

[0173] Figure 20 shows an example of a fare prediction screen displayed by the demand prediction app.

[0174] The demand prediction app causes the display to show the time and fare required for the movement to the destination as prediction results, and also causes the display to show the time and fare required for the movement for each divided unit obtained by dividing the movement route to the destination into predetermined units as prediction results.

[0175] In the fare prediction screen of Figure 20, the individual display 261 shows the time and fare required for the movement for each divided unit. The destination display 262 shows the time and fare required for the movement to the destination.

[0176] When learning the fare and travel time, if the actual vehicle data shown in Fig. 5 is used as it is, it is difficult to learn because data with the same departure and arrival points is required. Therefore, the control unit 121 learns the time and fare required for movement for each division unit from the vehicle movement log data. Then, the control unit 121 calculates the time and fare required for movement to the destination by obtaining the sum of the time and fare of each division unit included from the departure point to the arrival point. The division unit can be, for example, a unit divided using at least one or a plurality of units such as a predetermined distance, a predetermined time, or a unit delimited by a road section (block), a unit delimited by a signal or an intersection, etc.

[0177] The travel time and fare for each division unit, as well as the travel time and fare to the destination, may be displayed with a predetermined width, for example, as "5 minutes - 10 minutes", "300 yen - 500 yen".

[0178] With the various displays described with reference to Figs. 13 to 20, the driver of the taxi 11 can conduct more efficient business. That is, the demand prediction app can present prediction results that contribute to an increase in the occupancy rate.

[0179] The various displays described with reference to Figs. 13 to 20 can be appropriately set, such as turning the display on or off and the order of display, on the setting screen displayed on the display by the driver operating the setting button 64 of the demand prediction app.

[0180] <14. Learning of boarding position> Next, the learning and prediction other than the boarding demand performed by the server 12 will be described.

[0181] Fig. 21 is a diagram for explaining the learning of the boarding position with respect to a building.

[0182] For example, a user (customer) of taxi 11 arranges for taxi 11 using a carpooling app 272 executed on a terminal such as a smartphone from a predetermined position within building 271, and boards taxi 11 at a predetermined position 273 such as a taxi stand in building 271. In this case, the carpooling app 272 acquires the position information of the user at the timing when taxi 11 is arranged from the GPS receiver within the terminal, and transmits it to server 12 as the position information at the time of carpooling request. Also, the boarding position information, which is the position information of the user at the timing when the user boards taxi 11, can be obtained from the vehicle movement log data transmitted from the vehicle management device 22 of taxi 11.

[0183] Server 12 learns the relationship between the position information at the time of carpooling request and the position information at the time of boarding. Thereby, when the user arranges for taxi 11 from a predetermined position within building 271, the server 12 can learn the boarding position in building 271, i.e., where the driver should drive taxi 11 to in building 271. Server 12 stores the learning result in storage unit 122 as a boarding position list. The demand prediction app can display the learned boarding position in building 271 on map 41. Also, when the user designates building 271 as the destination, the driver can use the learned boarding position in building 271 as the drop-off position.

[0184] Also, server 12 can analogize the boarding position of a building other than building 271 where taxi 11 is actually arranged, from the learned relationship between the position information at the time of carpooling request and the position information at the time of boarding, and display it on map 41.

[0185] <15. Learning of Drop-off Position> FIG. 21 is a diagram for explaining the learning of the drop-off position for a building.

[0186] For example, the user gets off the taxi 11 at a predetermined position 274 and moves to a predetermined building 271 which is the destination. The getting-off position information, which is the position information of the user at the timing when the user gets off the taxi 11, can be obtained from the vehicle movement log data transmitted from the vehicle management device 22 of the taxi 11. Also, the dispatch app 272 obtains the position information of the building 271 where the user moves after getting off the taxi 11 from the GPS receiver in the terminal, and transmits it to the server 12 as the post-movement position information.

[0187] The server 12 learns the relationship between the getting-off position information and the post-movement position information. Thereby, when the user designates the building 271 as the destination, the driver can learn where to drop off the user, that is, the drop-off position of the building 271. The server 12 stores the learning result in the storage unit 122 as a drop-off position list. The demand prediction app can display the learned drop-off position of the building 271 on the map 41. Also, when the user arranges the taxi 11 using the dispatch app 272 from a predetermined position inside the building 271, the driver can also utilize the learned drop-off position of the building 271 as the pick-up position.

[0188] <16. Learning of Pick-up Position> In the example described with reference to FIGS. 21 and 22, it has been described that the learned pick-up position is also displayed as the drop-off position, and the learned drop-off position is also displayed as the pick-up position. Generally, the pick-up position and the drop-off position for a building are often places such as a taxi pool, a vehicle gathering point, or an entrance, and they often coincide or are close to each other.

[0189] Therefore, as shown in FIG. 23, the server 12 learns the pick-up time position information and the drop-off time position information, learns the optimal pick-up position for the building 271, and stores it in the storage unit 122 as a pick-up position list. The demand prediction app can display the learned pick-up position of the building 271 on the map 41.

[0190] <17. Ride Demand Guide by Sound> Next, an explanation will be given of the voice guidance of the predicted ride demand results executed by the demand prediction app.

[0191] During driving, for safety reasons, the driver of Taxi 11 cannot view the demand prediction screen displayed by the demand prediction app. Therefore, in addition to displaying the predicted results of the ride demand predicted on the map on the display, the demand prediction app has a function to notify the driver of the predicted results of the ride demand also by voice.

[0192] Since the demand prediction app cannot notify all the predicted results displayed on the demand prediction screen by voice, it notifies by voice the predicted results corresponding to the current location of Taxi 11 (hereinafter also referred to as the vehicle location) and the traveling direction of Taxi 11.

[0193] In addition, the demand prediction app switches the display method of the demand prediction and the voice notification method according to the scale of the map 41 on which the demand prediction is displayed.

[0194] Below, the demand prediction display and voice output when the scale of the map 41 on which the demand prediction is displayed is at a high magnification, in other words, in a wide-area map display, and the demand prediction display and voice output when the scale of the map 41 is at a low magnification, in other words, in a detailed map display, will be explained in order.

[0195] Note that the voice notification of the demand prediction app includes both notification by non-verbal sounds (also called effect sounds or electronic sounds) such as "pi, pi, pi", "pohn", "ping pong", and notification by voice (messages) of words and sentences. Below, for simplicity, the effect sounds will be simply expressed as sounds, and the voice output by words will be expressed as voices for explanation.

[0196] <Example of voice notification in wide-area map display> FIG. 24 shows an example of a demand prediction screen when the scale of the map 41 on which the demand prediction is displayed is at a high magnification, in other words, in a wide-area map display.

[0197] In the demand prediction screen of FIG. 24, among the areas AR of the demand prediction mesh 63 superimposed on the map 41, it is assumed that the area 411 is the area AR predicted to be a place with high boarding demand. In FIG. 24, the display of colors and densities according to the degree of boarding demand for each area AR other than the area 411 is omitted.

[0198] The taxi 11 is traveling at the location of the own vehicle position mark 421, and there is an area 411 predicted to be a place with high boarding demand in the traveling direction of Route 1.

[0199] The demand prediction application detects that there is an area 411 with high boarding demand in the traveling direction and notifies the driver by sound or voice that there is an area 411 with high boarding demand in the traveling direction. For example, the demand prediction application outputs "There is a place with high demand nearby in the traveling direction." by voice and outputs three consecutive sounds of "Pipipi".

[0200] Also, when the taxi 11 is traveling on Route 1 and is at the location of the own vehicle position mark 422, the demand prediction application outputs, for example, "High demand in the traveling direction." by voice and outputs four consecutive sounds of "Pipipipi".

[0201] In this way, when there is an area 411 which is a place with high boarding demand in the traveling direction, the demand prediction application notifies the driver by sound or voice at a predetermined timing. The notification timing can be performed, for example, in units of the area AR of the demand prediction mesh 63. In this case, every time the area AR where the traveling taxi 11 is located is changed, a notification by sound or voice is output. Alternatively, the demand prediction application may notify at preset time intervals (for example, every 1 minute) or at preset distance intervals (for example, every 1 km). The notification timing can be changed on the setting screen displayed by operating the setting button 64.

[0202] The demand prediction app changes the content of the voice message and the type of sound output according to the distance (closeness) to the area AR with high detected ride-hailing demand in the traveling direction.

[0203] In the example described above, when the vehicle position mark 421 of the vehicle position is far from the high-demand area, the demand prediction app outputs the voice "There is a place with high demand nearby in the traveling direction." and outputs three consecutive "pipipi" sounds. In a closer place, the demand prediction app outputs the voice "High demand in the traveling direction." and outputs four consecutive "pipipipi" sounds. That is, in the voice output, the message is changed from "There is a place with high demand nearby in the traveling direction." to "High demand in the traveling direction." In the sound output, the three consecutive "pipipi" sounds are changed to four consecutive "pipipipi" sounds, and the number of consecutive outputs of the "pi" sound increases as it gets closer to the high-demand area. In addition to increasing the number of consecutive times of the sound effect, for example, the length of the sound may be changed like "pi", "pii", "piii", or the volume may be gradually increased. Also, a combination of at least two of the number of sounds, length, or volume may be used.

[0204] Also, for example, when there is an intersection in the traveling direction and the high-demand area is in the direction of turning right at the intersection, the demand prediction app can output voice like "High demand in the right-turn direction at the intersection."

[0205] The degree of ride-hailing demand for which the demand prediction app notifies the high-demand area by sound or voice can also be changed on the setting screen displayed by operating the setting button 64. For example, as shown in Figure 2, each area AR of the demand prediction mesh 63 is displayed with a color and density corresponding to the degree of ride-hailing demand. When the degree of ride-hailing demand distinguished by color and density on the demand prediction screen in Figure 2 is in five levels from level 1 to level 5, it can be set to notify when the area AR of level 5 with the highest degree of ride-hailing demand exists in the traveling direction, or to notify when the area AR of level 4 or higher exists in the traveling direction.

[0206] The driver can also set on the setting screen whether to guide the high-demand area only by sound, only by voice, or by both sound and voice. The demand prediction application has a notification function (notification unit) that notifies the driver of a predetermined area AR, which is a place with high boarding demand, by at least one of sound or voice.

[0207] In addition, in FIG. 18, when there is an area AR where the ratio of rides with a long riding distance is high or a riding position, the long display 241 that displays the ratio (rate) of rides with a long riding distance as the longness was described.

[0208] When the demand prediction application is displaying the demand prediction screen of the wide-area map display on the display, it can also notify the driver of the area AR or riding position with a high longness by sound or voice.

[0209] For example, when an area AR with a longness equal to or greater than a predetermined value exists in the traveling direction, the demand prediction application outputs "There is a long area nearby." by voice and also outputs a sound of "Boon", which is a different type of sound from the notification of the high-demand area.

[0210] <Example of sound notification in detailed map display> FIG. 25 shows an example of the demand prediction screen in the case of a detailed map display, that is, when the scale of the map 41 for displaying the demand prediction is at a low magnification.

[0211] In the detailed map display of FIG. 25, the taxi 11 is traveling at the location of the own vehicle position mark 441.

[0212] In the detailed map display, as a result of learning using the actual vehicle sequence data, within the display area displayed on the display, the riding positions where the boarding demand exceeds a predetermined level are each displayed as circular demand points 451. Note that in order to prevent the figure from becoming complicated, in FIG. 25, some of the symbols of the demand points 451 are omitted.

[0213] For example, among the boarding positions learned within the display area displayed on the display, the top 30 locations with high boarding demand are displayed as demand points 451. However, even if a predetermined boarding position is included in the top 30, those selected due to the small total number of boarding positions are excluded. Therefore, the boarding positions displayed as demand points 451 are those where the predicted boarding demand is at least a predetermined first level ThA or higher and are included in the top 30 boarding positions.

[0214] In the example of FIG. 25, as demand points 451, there are a demand point 451A displayed as a thick circle, a demand point 451B displayed as a thin circle, and a demand point 451C displayed in a pattern different from that of demand points 451A and 451B.

[0215] The difference in the density between demand point 451A and demand point 451B represents the degree of predicted boarding demand. That is, when the predicted boarding demand is at or above a second level ThB indicating a high boarding demand, it is displayed as a demand point 451A with a high density, and when it is above the first level ThA and less than the second level ThB, it is displayed as a demand point 451B with a low density. In this way, by changing the density of demand point 451 according to the magnitude of the boarding demand, for example, when demand point 451 is displayed alone without overlapping other demand points 451, the driver can recognize the difference in boarding demand due to the difference in density. Also, when demand point 451 is displayed overlapping other demand points 451, since the density of demand point 451 appears high, the driver can also be made to recognize the location where the boarding demand is concentrated as a location with a high boarding demand. Note that instead of distinguishing with two types of densities of demand point 451A and demand point 451B, the density may be changed continuously according to the degree of boarding demand.

[0216] The demand point 451C displayed in a different pattern represents the demand point 451 among the top 30 demand points selected with high boarding demand, where the longitude is at or above a predetermined value. Thereby, the driver can recognize the demand point 451 with a high longitude in the detailed map display.

[0217] In FIG. 25, due to drawing constraints, the high-demand point 451C and the other demand points 451A and 451B are displayed with different patterns. However, on a display capable of color display, the demand point 451C and the demand points 451A and 451B can be distinguished by changing the color.

[0218] In the above example, up to the top 30 boarding positions within the display area of the detailed map display are displayed as the demand points 451. However, the number of demand points 451 to be displayed can be changed on the setting screen. For example, on the setting screen, it is possible to select from among the top 30, the top 20, and the top 10.

[0219] Also, in the above example, up to the top 30 boarding positions within the display area of the detailed map display are selected and displayed. However, up to the top 30 may be selected and displayed in units of the area AR of the demand prediction mesh 63. That is, the extraction unit for extracting the demand points 451 can be set as appropriate.

[0220] Next, the guidance by sound and voice in the detailed map display as shown in FIG. 25 will be described.

[0221] The taxi 11 is traveling from the location of the own vehicle position mark 441 in a predetermined direction (for example, the direction of Shinagawa Station). When a demand point 451 exists within a predetermined distance in the traveling direction, the demand prediction application outputs "High demand in the traveling direction." by voice. Also, every time the taxi 11 passes a demand point 451, the demand prediction application outputs a "beep" sound. One sound is output for each demand point 451.

[0222] In the detailed map display, since a sound is output when passing through the demand point 451, on a road where many demand points 451 are displayed, the sound of "pi" is continuously generated. As a result, the driver can recognize that the road currently being traveled is a road with high passenger demand, contributing to the improvement of the recognition of areas AR and roads with high passenger demand, and ultimately contributing to the improvement of the boarding rate. According to the degree of passenger demand at each demand point 451, the volume of the "pi" sound may be changed and output.

[0223] Although different from the example in FIG. 25, when there is no demand point 451 in the traveling direction of the taxi 11 and many demand points 451 exist in the direction opposite to the traveling direction, the demand prediction application outputs a voice such as "There are high-demand points in the opposite direction." Also, when there is an intersection in the traveling direction and the road where many demand points 451 exist is the direction for the taxi to turn right at the intersection, the demand prediction application outputs a voice such as "Intersection right-turn direction, high demand."

[0224] FIG. 26 shows other display examples of the detailed map display.

[0225] In the detailed map display of FIG. 26, some of the demand points 451 shown in the detailed map display of FIG. 25 are replaced by demand points 451D to 451F.

[0226] As shown in FIG. 4, the traveling direction of the taxi 11 is also recorded in the vehicle dynamics log data, so the traveling direction of the taxi 11 that boarded at the demand point 451 is also learned. The traveling direction of the taxi 11 at the demand point 451 is learned in 8 directions, similar to the arrows 212-1 to 212-8 in FIG. 14, for example. When there is a traveling direction that accounts for a ratio of 50% or more among the 8 traveling directions of the taxi 11 that boarded at each demand point 451, the demand prediction application can display that direction (hereinafter referred to as the dominant direction).

[0227] Demand point 451D represents demand point 451 indicating the dominant direction. Demand point 451E is displayed when there are a predetermined number or more of demand points 451D having the same dominant direction within a predetermined range. That is, demand point 451E represents a set of a predetermined number or more of demand points 451D.

[0228] Demand point 451D is displayed, for example, as a mountain-shaped (V-shaped) symbol as shown in FIG. 26, and the direction pointed by the corner represents the dominant direction. Demand point 451E is represented by a symbol obtained by further enlarging the symbol of demand point 451D. Note that demand point 451D and demand point 451E may also be displayed using other symbols that can indicate a direction, such as an arrow symbol.

[0229] Demand point 451F represents a demand point 451 having a dominant direction and a high length. In other words, when demand point 451D having a dominant direction is a demand point 451 with a high length like demand point 451C in FIG. 25, by changing the pattern or color for display like demand point 451F, it is also simultaneously expressed that it is a demand point 451 with a high length. When demand point 451E also has a high length, similarly, it is displayed by changing the pattern or color.

[0230] When there is a dominant direction at each demand point 451 in the detailed map display, by displaying that dominant direction, it is possible to assist the driver in determining the driving direction when, for example, performing a so-called "flow", and it is possible to contribute to an improvement in the boarding rate. Also, at intersections where there can be various directions, by referring to the dominant direction around the intersection, the driver can determine the driving direction with a high boarding demand.

[0231] The notification method by sound and voice in the detailed map display of FIG. 26 is the same as that in FIG. 25, so the description is omitted.

[0232] Even in the detailed map display, whether to guide the demand point 451 only by sound, only by voice, or by both sound and voice is changed according to the setting value on the setting screen. The demand prediction app has a notification function (notification unit) that notifies the driver of a predetermined demand point 451, which is a place with high boarding demand, by at least one of sound or voice.

[0233] As described above, the demand prediction app can contribute to improving the boarding rate by notifying the driver of the prediction result of the boarding demand by sound or voice with respect to the traveling direction of the taxi 11.

[0234] When the demand prediction screen is in the wide-area map display (Figure 24) and when it is in the detailed map display (Figures 25 and 26), the display method of the boarding demand is also different, and accordingly, the notification methods of sound and voice are also different.

[0235] The switching between the wide-area map display and the detailed map display can be performed, for example, when the driver operates the zoom button 62 or changes the setting on the setting screen. Alternatively, instead of the touch panel operation, the demand prediction app may recognize (voice recognition) the voice instructions of "wide-area display" or "detailed display" issued by the driver and perform the switching. Furthermore, the demand prediction app may perform the switching automatically. For example, when it is determined that the own vehicle position of the taxi 11 has entered the high-demand area guided by the wide-area map display, the demand prediction app can change to the detailed map display, and when it is determined that the taxi has exited the high-demand area, it can change to the wide-area map display.

[0236] <Voice guidance control process> The above-mentioned guidance on the prediction of the boarding demand by sound and voice is required when the taxi 11 is empty without passengers and is not required when there are passengers. Also, the notification by sound and voice is noise for the passengers. Therefore, the demand prediction app can obtain the status of "occupied" or "empty" detected by the fare meter 21 and control the on / off of the guidance by sound and voice in conjunction with the status.

[0237] Figure 27 is a flowchart of the voice guidance control process for controlling voice guidance.

[0238] First, in step S51, the demand prediction application determines whether the status of taxi 11 (vehicle) has been changed to "actual vehicle".

[0239] If it is determined in step S51 that the status of taxi 11 has been changed to "actual vehicle", the process proceeds to step S52, and the demand prediction application controls to turn off the voice guidance. After step S52, the process returns to step S51.

[0240] On the other hand, if it is determined in step S51 that the status of taxi 11 has not been changed to "actual vehicle", the process proceeds to step S53, and the demand prediction application determines whether the status of taxi 11 has been changed to "available".

[0241] If it is determined in step S53 that the status of taxi 11 has been changed to "available", the process proceeds to step S54, and the demand prediction application controls to turn on the voice guidance. After step S54, the process returns to step S51.

[0242] On the other hand, if it is determined in step S53 that the status of taxi 11 has not been changed to "available", the process also returns to step S51.

[0243] The voice guidance control process in Figure 27 starts when the demand prediction application is launched or when the voice guidance is set to on in the setting screen, and is repeated until the demand prediction application is terminated.

[0244] As described above, the demand prediction application can control the on / off of the voice guidance in conjunction with the "actual vehicle" or "available" status detected by the fare meter 21. By controlling in conjunction with the "actual vehicle" or "available" status, the voice guidance can be automatically executed (without the driver's operation) only when the driver needs it.

[0245] Note that the voice guidance control process in Fig. 27 is an example of a notification control process for voice guidance that outputs the boarding demand in the traveling direction as a voice message. However, notifications by sound, and notifications by both sound and voice are controlled in the same way. The status of "occupied" or "empty" may be directly obtained by the demand prediction application from the fare meter 21, or may be indirectly obtained via the vehicle management device 22.

[0246] <Other voice outputs> In addition to notifying the above-mentioned boarding demand by sound and voice, the demand prediction application can also notify the driver of the following information by voice.

[0247] For example, the demand prediction application can obtain the train operation stop information and operation restart information from the server 12 in real time and output them as voice (messages). As a result, the driver can quickly move to an area where the boarding demand increases according to the train operation information.

[0248] For example, the demand prediction application can obtain the end timing and start timing of an event held in the business area of the taxi 11 from the server 12 in real time and output them as voice (messages). As a result, the driver can quickly move to an area where the boarding demand increases according to the end and start of the event.

[0249] For example, the demand prediction application can obtain the weather information in the business area of the taxi 11, particularly information regarding sudden weather changes, from the server 12 in real time and output them as voice (messages). As a result, the driver can quickly move to an area where the boarding demand increases according to the weather changes.

[0250] The server 12 obtains train operation information, event information, weather information, etc. from the servers of partner information providing companies and transmits them to the demand prediction applications of each terminal device 23.

[0251] <18. Recommended route presentation process> In the above-described embodiment, an example has been described in which the driver himself / herself determines the moving direction of the taxi 11 based on the display of the ride demand prediction and the voice output displayed on the display of the terminal device 23 by the demand prediction application, and drives the taxi 11.

[0252] Hereinafter, a recommended route presentation function will be described in which the demand prediction application searches for a route with a high ride demand based on the ride demand prediction data transmitted from the server 12 and presents a recommended route.

[0253] The demand prediction application has a navigation function (navigation processing unit) that searches for a route to a destination set by an arbitrary method based on the position of the own vehicle. Further, in addition to the functions of a general navigation system (hereinafter referred to as a car navigation system) mounted on the vehicle, the demand prediction application has a function of searching for a route to the destination in consideration of a place predicted to be a place with a high ride demand. Note that the destination may be a predetermined pinpoint location or a target area such as one or more areas AR divided by the demand prediction mesh 63. Hereinafter, a place predicted to be a place with a high ride demand will be described as a boarding point. The boarding point is the same as the place referred to as the boarding position or demand point in the above description.

[0254] The recommended route presentation function of the prediction app is used when the taxi 11 is empty, that is, when there are no passengers, during what is called "circulating". By using the recommended route presentation function, passengers can be picked up in a short time, and the empty vehicle time and the driving distance in the empty vehicle state can be shortened. Also, new drivers who have been in the taxi business for a short period of time do not have a defined preferred area (region) and lack knowledge of places with high passenger demand, so it is particularly useful for them. Even for veteran drivers who have been in the taxi business for a long time or drivers who achieve above-average sales, there are preferred areas and non-preferred areas. So, when driving in a non-preferred area, passengers can be picked up in a short time, and the empty vehicle time and the driving distance in the empty vehicle state can be shortened. By setting the destination during route search to a preferred area or location, it becomes possible to return to the desired area or location while picking up passengers through a high-demand route (pick-up point). By repeatedly passing through high-demand routes (pick-up points) using the recommended route presentation function, non-preferred areas can be overcome and preferred areas can be expanded.

[0255] Preferred areas may be pre-registered (set) for the driver in the settings screen or the like, or the demand prediction app may determine (automatically register) them using the driver's vehicle movement log data and actual vehicle data. Preferred locations can be, for example, the driver's business office or pinpoint pick-up locations that the driver frequently uses, such as locations where "waiting for passengers" is carried out, such as taxi stands that the driver frequently uses.

[0256] When the demand prediction app determines an area of expertise using data, the demand prediction app creates, for example, a heatmap with the cumulative time in the Z-axis direction for business areas distributed in the XY plane as shown in FIG. 28, and can extract an area with a time value equal to or greater than a predetermined threshold value, for example, a time value corresponding to 70% of the total time, as the driver's area of expertise. The area of expertise may be all one or more areas extracted as being equal to or greater than the threshold value, or may be one area with the maximum cumulative time, or multiple areas with the top cumulative times. The display unit of the area of expertise may be the extracted area as it is, or, in order to make it easier for the driver to recognize, the city, town, or village with the largest area among the extracted areas, or the city, town, or village where the point with the maximum cumulative time exists may be used as the display unit.

[0257] When the original data for calculating the cumulative time in the Z-axis direction is vehicle motion log data, an area where the driver has moved for a long time can be set as the area of expertise. Also, when the original data for calculating the cumulative time in the Z-axis direction uses only vehicle motion log data with a status of "actual vehicle", an area where the driver has carried passengers for a long time can be set as the area of expertise. When the original data for calculating the cumulative time in the Z-axis direction is actual vehicle data, an area where the driver has carried passengers in many places can be set as the area of expertise.

[0258] FIG. 29 shows an example of the display of a recommended route presentation screen in a state where the recommended route presentation function is executed in the demand prediction app.

[0259] The recommended route presentation screen in FIG. 29 is displayed, for example, after a recommended route is searched when the driver taps a recommended route search button or the like displayed on the screen of a demand prediction screen such as FIG. 2.

[0260] On the recommended route presentation screen of FIG. 29, a map 41 with a demand prediction mesh 63 superimposed thereon is displayed. Each area AR of the demand prediction mesh 63 is classified and displayed by color or density according to the degree of boarding demand, as described with reference to FIG. 2. However, in the example of FIG. 29, for the sake of clarity of the figure, the display by color or density is omitted. The same applies to FIGS. 32 to 44 described later in that the display by color or density according to the degree of boarding demand is omitted.

[0261] On the map 41, a route search result 501 of the recommended route is displayed. Also, on the map 41, a detail button 511 operated when reducing the scale of the map 41 to a low magnification, a wide area button 512 operated when increasing the scale of the map 41 to a high magnification, a full screen display button 513 operated when switching to full screen display, etc. are also displayed.

[0262] Below the map 41, a predicted time display section 502 is arranged. In the predicted time display section 502, the predicted time of demand prediction when the recommended route is searched is displayed, and similar to the predicted time setting area 42 of FIG. 2, the predicted time of demand prediction can be changed.

[0263] On the right side of the map 41 and the predicted time display section 502, a recommended route information presentation section 503 is arranged. In the recommended route information presentation section 503, the characters "Recommended Route" indicating that it is a recommended route presentation screen are displayed at the uppermost stage.

[0264] In the recommended route information presentation section 503, "<Destination Mode>" is displayed, indicating that the route search result 501 displayed on the map 41 was searched in the "Destination Mode" among a plurality of search modes. The demand prediction application has three search modes: "Destination Mode", "Pick Up Immediately Mode", and "Nearby Boarding Point Mode" as route search modes considering demand points.

[0265] "Destination Mode" is a mode in which a destination is set, and a route passing through demand points is searched from among multiple routes to the destination. The destination may be set by the driver, or the driver's preferred area may be used as the destination (target area). For example, "Destination Mode" is suitable when the current location of Taxi 11 is outside the preferred area and the driver wants to pick up passengers while returning to the preferred area.

[0266] "Immediate Pickup Mode" is a mode in which there is no specific destination, and a route passing through demand points is searched from the current location. "Immediate Pickup Mode" can be used when the current location of Taxi 11 is within the preferred area.

[0267] "Nearby Pickup Point Mode" is a mode in which there is no specific destination, and a route passing through demand points with a high evaluation value is searched in the vicinity of the current location. "Nearby Pickup Point Mode" can be used whether the current location of Taxi 11 is within the preferred area or outside the preferred area.

[0268] Below the "<Destination Mode>" of the Recommended Route Information Display Unit 503, "Return to the area 'Setagaya Ward'" indicating that the destination of the route search in "Destination Mode" is Setagaya Ward is displayed.

[0269] Furthermore, in the recommended route information presentation unit 503, information such as "recommendation level: 98 points", "number of boarding points: 108 locations", and "total demand prediction level: 6 levels" is displayed. "Recommendation level: 98 points" indicates that the recommendation level of the route search result 501 displayed on the map 41 is 98 points. The recommendation level may be a score out of 100 points, or it may be a score display where the larger the numerical value, the higher the recommendation level. "Number of boarding points: 108 locations" indicates that the route of the route search result 501 displayed on the map 41 is a route that passes through 108 boarding points. In the demand prediction screen of FIG. 2, an example was described in which the degree of boarding demand for each area AR distinguished by color and density is displayed in five levels from level 1 to level 5. "Total demand prediction level: 6 levels" indicates that when the levels of the boarding demand for one or more areas AR through which the route search result 501 displayed on the map 41 passes are totaled, it becomes 6 levels.

[0270] Referring to the flowchart of FIG. 30, a recommended route presentation process for searching for a recommended route considering demand points and presenting it to the driver will be described.

[0271] This process is started, for example, by tapping a recommended route search button or the like displayed on the screen of the demand prediction screen such as FIG. 2. Note that as the boarding demand prediction data used in the recommended route presentation process, data periodically transmitted from the server 12 can be used, or it can be obtained by requesting the server 12 as necessary.

[0272] First, in step S71, the demand prediction application determines whether a search mode is specified. The search mode can be specified, for example, by selecting any one of the search mode buttons of "destination mode", "pick-up immediately mode", or "nearby boarding point mode" on the mode selection screen displayed after pressing the route search button. Alternatively, it may be preset such that the "destination mode" is selected when the vehicle position is outside the preferred area, and the "pick-up immediately mode" or "nearby boarding point mode" is selected when the vehicle position is within the preferred area.

[0273] The process of step S71 is repeated until it is determined that the search mode is specified. When it is determined that the search mode is specified, the process proceeds to step S72.

[0274] Then, in step S72, the demand prediction application determines whether the specified search mode is any of "destination mode", "pick-up immediately mode", or "nearby pick-up point mode".

[0275] In step S72, if it is determined that the specified search mode is "destination mode", the process proceeds to step S73, and the processes of steps S73 to S75 are executed. If it is determined that the specified search mode is "pick-up immediately mode", the process proceeds to step S76, and the processes of steps S76 and S77 are executed. If it is determined that the specified search mode is "nearby pick-up point mode", the process proceeds to step S78, and the processes of steps S78 to S80 are executed.

[0276] In step S73 when it is determined that the specified search mode is "destination mode", the demand prediction application sets the destination to the preferred area. The destination is set to the pre-registered preferred area by default, but it can also be changed by the driver's operation.

[0277] In step S74, the demand prediction application uses a search path algorithm such as Dijkstra's algorithm or A* algorithm to search for a plurality (predetermined number) of routes based on the vehicle's position and the destination. This process is the same as that of a general car navigation function.

[0278] In step S75, the demand prediction application calculates the total score SUMscore of each searched route.

[0279] The total score SUMscore of the route can be calculated, for example, by calculating the score Sc = the ride demand level [1 - 5] of area AR × the boarding point [0, 1] × the direction consistency [cosθ] for each area AR through which the searched route passes, and then summing up the scores Sc of each area AR from the vehicle position to the destination. The ride demand level [1 - 5] of area AR represents the degree of ride demand in that area AR and takes any value from level 1 to 5. The boarding point [0, 1] is "1" if there is a boarding point in that area AR and "0" if there is none. The direction consistency [cosθ] is the angle (cosθ) of the direction from the vehicle position to area AR with respect to the direction from the vehicle position to the destination, and the more the directions match, the closer the value is to 1. Regarding the demand prediction used in the above-mentioned score Sc calculation formula, the predicted time of area AR is sequentially updated according to the distance from the vehicle position, for example, the number of area ARs passed through. For example, assuming that one area AR is 500m × 500m and runs for 10 minutes at 30 km / h, it advances 5 km in 10 minutes. So, for every 10 area movements, the predicted time of area AR is advanced by 10 minutes, and the demand prediction of the route is updated. Note that the boarding point [0, 1] in the above-mentioned score Sc calculation formula may also be the number of boarding points existing in that area AR, etc.

[0280] The total score SUMscore of the route may be calculated by calculating the score Sc for each boarding point existing on the route, rather than in the above-mentioned area AR unit. Also, in some paths of the searched route, when the direction is opposite to the destination, the addition of the score Sc may not be performed, or the subtraction of the score Sc may be performed.

[0281] The total score SUMscore of the route is not limited to the above example, and any calculation method can be adopted such that the total score SUMscore of a route with a high demand prediction level and many boarding points becomes large.

[0282] In addition, other perspectives may be added to the total score SUMscore of the route to calculate the score Sc or the total score SUMscore. For example, a coefficient or score Sc corresponding to the length of the pick-up point may be added so that the higher the length, the larger the total score SUMscore or the score Sc. Alternatively, a coefficient or score Sc may be added according to whether the route passed through in the searched route is a main road or a narrow road. From the starting point and the arrival point of the past riding history (actual vehicle data), the moving direction of the passengers when boarding at each pick-up point can be identified. The closer the driver's destination is to the moving direction of the pick-up point, the larger the total score SUMscore or the score Sc may be made.

[0283] After calculating the total score SUMscore of each route in step S75, the process proceeds to step S81 described later.

[0284] On the other hand, in step S76 when it is determined that the specified search mode is the "immediate pick-up mode", the demand prediction application determines a plurality of routes by depth-limited search, which is a graph theory, for a predetermined route search area. More specifically, in the graph theory, nodes are intersections and edges are roads (routes) between intersections, and a plurality of routes are determined by depth-limited search within the route search area. The route search area may be the preferred area when the vehicle position is within the preferred area, may be within a radius of several kilometers from the vehicle position, or may be a predetermined number of areas AR centered on the vehicle position.

[0285] In step S77, the demand prediction application calculates the total score SUMscore of each searched route. The demand prediction application calculates the total score SUMscore, for example, by calculating and summing the scores Sc for each pick-up point existing on the searched route, for example, with the score Sc = the ride demand level [1-5] of the area AR including the pick-up point.

[0286] The total score SUMscore of the route can adopt any calculation method such that, not limited to the above example, the total score SUMscore of the route with a high demand prediction level and many boarding points becomes larger. Addition or subtraction of the reverse score Sc, update of the prediction time of demand prediction, etc. can be performed in the same way as the calculation in the "destination mode".

[0287] After calculating the total score SUMscore of each route in step S77, the process proceeds to step S81 described later.

[0288] On the other hand, in step S78 when it is determined that the specified search mode is the "nearby boarding point mode", the demand prediction application extracts the boarding points within a certain distance from the vehicle position and assigns the score Sc to each of the extracted boarding points.

[0289] The score Sc of each boarding point in the "nearby boarding point mode" can be assigned, for example, as follows.

[0290] For example, the demand prediction application assigns a larger score Sc to each boarding point as the date and time of boarding is closer to the current date and time. The demand prediction application assigns a larger score Sc to each boarding point as the number of boarding times (boarding count) is larger. The demand prediction application assigns a larger score Sc to each boarding point as the total boarding time (the sum of the boarding times of each boarding trip) or the average boarding time is larger.

[0291] The demand prediction application assigns a score Sc to each boarding point such that the places that can be reached by going straight or turning left from the vehicle position are high, and the places that can be reached by turning right are low.

[0292] For example, as shown in FIG. 31, taxi 11 is traveling at the location of the own vehicle position mark 505. Assume that there are a pick-up point HS1 that can be reached straight ahead, a pick-up point HS2 that can be reached by turning left, and a pick-up point HS3 that can be reached by turning right as pick-up points. In Japan where traffic keeps to the left, the straight-ahead pick-up point HS1 and the left-turn pick-up point HS2 are relatively easy to reach, but the right-turn pick-up point HS3 that requires crossing the oncoming lane depends a lot on the signal timing and the timing of vehicles traveling in the oncoming lane, and it often takes a long time to reach. Therefore, the demand prediction app can assign a higher score Sc to the pick-up points HS1 and HS2 than to the pick-up point HS3.

[0293] Note that on roads in foreign countries where traffic keeps to the right, the score Sc is assigned such that locations that can be reached by going straight or turning right are given a high score, and locations that can be reached by turning left are given a low score. That is, pick-up points for going straight and turning in a direction that does not require crossing the oncoming lane are given a high score, and pick-up points for turning in a direction that requires crossing the oncoming lane are given a low score, and the score Sc is assigned.

[0294] The demand prediction app assigns a score Sc that gets higher the closer the location is to the own vehicle position at each pick-up point.

[0295] The demand prediction app assigns a score Sc that gets higher the larger the ratio (e.g., the number of pick-ups at that pick-up point / the total number of pick-ups in area AR) that the pick-up point occupies within the area AR where the pick-up point exists at each pick-up point.

[0296] When a destination is set, the demand prediction app assigns a score Sc that gets higher the greater the degree of match with the direction of the destination at each pick-up point.

[0297] The sum of the scores Sc assigned to each pick-up point as described above becomes the final score Sc for each pick-up point. Note that the above is an example of an example of assigning the score Sc to each pick-up point, and the score Sc may be assigned based on other assignment criteria.

[0298] In step S79, the demand prediction app refers to the score Sc assigned to each boarding point, and searches for a plurality of routes using a predetermined number of boarding points with high scores Sc as waypoints. More specifically, first, the score Sc assigned to each boarding point is referred to, and a predetermined number of boarding points with high scores Sc are extracted. Then, a plurality (predetermined number) of routes are searched using a search route algorithm so as to pass through the extracted boarding points.

[0299] In step S80, the demand prediction app calculates the total score SUMscore of each searched route. The demand prediction app calculates the total score SUMscore, for example, by adding up the scores Sc of each boarding point existing on the searched route. Alternatively, the number of boarding points existing on the route may be used as the total score SUMscore.

[0300] After calculating the total score SUMscore of each route in step S80, the process proceeds to step S81.

[0301] In step S81, among the plurality of routes calculated in any of the search modes of "destination mode", "pick up immediately mode", or "nearby boarding point mode", the route with the highest total score SUMscore is displayed on the display as the recommended route. The recommended route presentation screen shown in FIG. 29 is an example where the search mode is displayed in "destination mode".

[0302] The recommended route presentation process of FIG. 30 is executed as described above. After the recommended route is presented, route guidance is started in the same manner as a general car navigation system. That is, in addition to the route display on the display, the driver is instructed about the route by voice guidance such as "turn right at the next intersection". In the route display on the display, the symbols, colors, patterns, etc. of the boarding points displayed on the route may be changed according to the value (size) of the score Sc of the boarding point, like the demand point 451 shown in FIG. 26.

[0303] When the driver performs an operation to stop the recommended route guidance by touch panel operation or the like, the recommended route guidance ends. In addition, when the demand prediction application of the terminal device 23 acquires the status of "occupied vehicle", "empty vehicle", or "picking up passenger" from the fare meter 21 or the vehicle management device 22 and becomes a status other than "empty vehicle", that is, a status of "occupied vehicle" or "picking up passenger", it may be set to end automatically without the driver's operation.

[0304] <19. Prohibited boarding area guidance display> The driver of taxi 11 needs to pay attention to the prohibited boarding area where it is prohibited to pick up passengers outside the taxi stand. If a driver who is not familiar with the prohibited boarding area, such as a novice driver, picks up a passenger in the prohibited boarding area, there are severe penalties. For example, in the Kanto region, prohibited boarding areas are set in Ginza and Shinbashi. In the Kansai region, prohibited boarding areas are set in Kitashinchi and Minami.

[0305] The demand prediction application can display the prohibited boarding area on the demand prediction screen.

[0306] Figure 32 shows an example of a demand prediction screen displaying the prohibited boarding area.

[0307] In Figure 32, the parts corresponding to Figure 29 and the like are labeled with the same reference numerals, and the description of those parts is omitted as appropriate.

[0308] On the demand prediction screen of Figure 32, on the map 41 with the demand prediction mesh 63 superimposed, a detail button 511, a wide area button 512, and a full screen display button 513 are displayed. Also, on the map 41, a current location button 514 for switching the display of the map 41 to a display based on the vehicle's own position is also displayed.

[0309] Furthermore, in the area corresponding to the no-boarding area on the map 41, a no-boarding area display 521 is shown. Also, in the vicinity of the no-boarding area display 521, a detailed display 522 for displaying detailed information regarding the no-boarding area is also shown. The detailed display 522 includes the text "No-boarding area from 22:00 to 1:00" indicating the time when the no-boarding area is applicable, and an [OFF] button for erasing the no-boarding area display 521. Since the no-boarding area display 521 is superimposed on the map 41, when it is difficult to see the display indicating the boarding point for demand prediction, or for a driver who does not need the display of the no-boarding area, the [OFF] button can be operated to erase the no-boarding area display 521. The demand prediction application erases the no-boarding area display 521 and the detailed display 522 when it is outside the time when the no-boarding area is applicable.

[0310] Also, on the demand prediction screen of FIG. 32, a predicted time display section 502 and an additional information display section 531 are provided in an area different from the display area of the map 41. In the additional information display section 531, for example, train operation information, event information, weather information, etc. are displayed.

[0311] Information regarding the no-boarding area may be stored in advance in the demand prediction application (terminal device 23), or may be acquired from the server 12 or the servers of other information providing companies, etc.

[0312] <20. Location display for picking up passengers> In the display of the predicted waiting time for picking up passengers shown in FIG. 17, it was explained that there is a method of lining up in the taxi queue at the taxi stand called "picking up passengers" and waiting for the taxi 11 to obtain passengers. The place where "picking up passengers" is performed (hereinafter referred to as the picking-up location) is, for example, a taxi stand in front of a station or a hotel, or in front of the entrance of a predetermined office building. There are places where the taxi companies that can use the picking-up location are limited in such picking-up locations. Taxis 11 of other taxi companies cannot use the picking-up locations limited to a predetermined taxi company. The demand prediction application has a function of displaying that it is a picking-up location dedicated to a predetermined taxi company.

[0313] FIG. 33 shows an example of a demand prediction screen that displays the pick-up locations for each taxi company.

[0314] On the demand prediction screen of FIG. 33, on the map 41 overlaid with the demand prediction mesh 63, a detail button 511, a wide area button 512, a full screen display button 513, and a current location button 514 are displayed.

[0315] Also, at each of the predetermined pick-up points 541A and 541B on the map 41, pick-up location displays 551A and 551B indicating that these pick-up points are pick-up locations are displayed. The pick-up location display 551A is represented by a solid-line circular graphic, and the pick-up location display 551B is represented by a dashed-line circular graphic, and the display methods of the pick-up location display 551A and the pick-up location display 551B are different. This difference in the display method represents that the taxi companies that can use the pick-up location are different. When not particularly distinguishing between the pick-up location display 551A and the pick-up location display 551B, it is simply referred to as the pick-up location display 551.

[0316] The driver can display the detailed information 553 of the pick-up location by selecting (tapping) a predetermined pick-up location display 551. In the example of FIG. 33, the detailed information 553 regarding the pick-up location display 551A of the pick-up point 541A is displayed.

[0317] The detailed information 553 displays information such as the name of the pick-up location, whether the taxi 11 can use the pick-up location, the time zone when passengers can be picked up, the name of the taxi company that can use the pick-up location, and other available pick-up locations displayed on the map 41 of the demand prediction screen.

[0318] In the detailed information 553 of the pick-up location display 551A in Fig. 33, the pick-up location name is "Hotel Shinagawa". Regarding whether taxi 11 can be used, "× (not available)" indicates that it is a pick-up location where it cannot be used. "All day" is indicated as the time period during which passengers can be taken. "Km Taxi Exclusive" is indicated as the name of the taxi company that can use the pick-up location. On the map 41 of the demand prediction screen, another available pick-up location is displayed, that is, the pick-up location name of the boarding point 541B of the pick-up location display 551B is "Osaki Think Building".

[0319] Regarding the taxi company of the driver, in the login screen and settings screen when the demand prediction app is launched, by registering (inputting) the company ID that identifies the company and the crew ID that identifies the driver on taxi 11, the demand prediction app can identify them.

[0320] In the screen example of Fig. 33, the pick-up location displays 551A and 551B are represented by circular graphics surrounding the boarding point, but the display method for indicating the pick-up location is not limited to this. The pick-up location display 551 may change the display (color and symbol) according to whether taxi 11 can be used, or only display the pick-up location display 551 for the pick-up locations where taxi 11 can be used.

[0321] The information about the pick-up location may be stored in advance in the demand prediction app, or may be obtained from server 12 or the servers of other information providing companies. Since the vehicle movement log data includes the company ID that identifies the company to which taxi 11 belongs, server 12 can not only obtain the known pick-up location information, but also estimate the boarding point from the history of the vehicle movement log data, and determine whether the boarding point is a pick-up location. If it is a pick-up location, it can identify which taxi company can be used. As described above, since the pick-up waiting operation of taxi 11 can be detected to identify whether the boarding point is a pick-up location, after the pick-up waiting operation, the boarding change point where the status changes from "available" to "occupied" can be set as the boarding point of the pick-up location.

[0322] The location display 551 of the prediction app can prevent the driver from going to unnecessary locations, enabling efficient business operations.

[0323] <21. Train schedule display> During the operating hours of a certain railway line, at the station where the last train (so-called last train) departs, the demand for taxis 11 increases among those who miss the train. Also, at the station where the last train arrives, it is often the case that other transfer lines and buses have ended operation by that time, increasing the demand for taxis 11. Alternatively, on railway lines with few train operations (long operation intervals), people who get off the train at a certain station often use taxis 11. Therefore, if the driver can be informed of train time information such as the last train time and arrival time of the train, the driver can go to the taxi stand in front of the station based on that train time information and acquire passengers.

[0324] The demand prediction app has a function of displaying predetermined train times such as the last train time of stations on the railway line shown on the map 41 on the demand prediction screen.

[0325] Figure 34 shows an example of a demand prediction screen displaying the train time of the last train.

[0326] On the map 41 of the demand prediction screen in Figure 34, there are the Shinagawa Station of Keihin Kyuko Electric Railway Co., Ltd., the Shinagawa Station of East Japan Railway Company (JR East Japan), and the Takarada Station of the Toei Asakusa Line, and the train times of the last trains of each are displayed.

[0327] Specifically, the time display 571 shows "Shinagawa Station 0:23 to Kanazawa Bunko", indicating that the departure time of the last train from the Shinagawa Station of Keihin Kyuko Electric Railway Co., Ltd. to Kanazawa Bunko is 0:23.

[0328] The time display 572 shows "Shinagawa Station 0:46 to Oosaki", indicating that the departure time of the last train from the Shinagawa Station of East Japan Railway Company (JR East Japan) to Oosaki is 0:46.

[0329] The time display 573 shows "Takatori Station 0:31, heading for Nishi-Magome", indicating that the departure time of the last train heading for Nishi-Magome at Takatori Station on the Toei Asakusa Line is 0:31.

[0330] On the right side of the map 41 on the demand prediction screen, a list display section 581 that displays the train time information shown on the map 41 in a list is shown.

[0331] In the list display section 581, the same information as the time displays 571 to 573 is listed in a predetermined order such as in ascending or descending order of train time, or in ascending or descending order of the distance from the vehicle position to the station. The sort button 582 is operated when changing the order displayed in the list display section 581, such as from ascending order of train time to descending order, from ascending order of the distance from the vehicle position to descending order, or from train time order to distance order from the vehicle position. The train time may be either the arrival time or the departure time.

[0332] By having the demand prediction application have a function to display predetermined train times such as the last train time, the chance of acquiring passengers such as those who missed the last train or those who got off at the last train increases. The train time display function can be turned on or off by setting. Instead of displaying the train times of all the stations shown on the map 41 of the demand prediction screen, it may be limited to stations with a large number of users (users exceeding a certain value), terminal stations where multiple lines converge, stations that are the starting or ending points, etc. Instead of the stations shown on the map 41, stations within a certain distance from the vehicle position (for example, within a radius of 2.5 km) may be displayed. Or, only the stations in the traveling direction of the vehicle may be displayed. How to display the train times of stations under what conditions may also be able to be set on the setting screen.

[0333] As described above, on railway lines with few train operation frequencies (long operation intervals), not limited to the last train or the first train, all train times may be displayed.

[0334] The demand prediction app can acquire train schedule data from server 12 together with or as part of the ride demand prediction data and display it on the demand prediction screen. The demand prediction app may be linked to the current location and current time of taxi 11 and, for example, display it when it is a predetermined time before the train schedule, or a train schedule display button may be provided on the demand prediction screen and displayed based on the driver's operation.

[0335] <22. Reverse pick-up point display> When a passenger uses taxi 11 with Haneda Airport or Narita Airport as the destination, it can be expected that the riding distance will be long. Therefore, a driver may wish to pick up passengers heading to a specific destination. Since the actual vehicle data records the departure point and the arrival point, by collecting the actual vehicle data where the arrival point is a specific location, the pick-up point (departure point) where the destination is a specific location can be analyzed.

[0336] Therefore, the demand prediction app has a reverse pick-up point display function that allows the driver to specify a specific location as the destination and only displays the pick-up points where the specified location was actually used by passengers as the destination in the past actual vehicle data. The reverse pick-up point represents a pick-up point where the destination is limited to a specific location. Examples of locations that can be specified as the destination include Haneda Airport, Narita Airport, Tokyo Disney Resort (registered trademark) (hereinafter referred to as TDR), etc.

[0337] Figure 35 shows an example of a demand prediction screen on which the reverse pick-up point display function is executed.

[0338] The demand prediction screen in Figure 35 is displayed, for example, when the driver operates a reverse pick-up point display button or the like displayed on the demand prediction screen.

[0339] In the demand prediction screen of Figure 35, a reverse pick-up point display section 601 is arranged adjacent to (on the right side of) the map 41 overlaid with the demand prediction mesh 63. Below the map 41, a predicted time display section 602 is arranged.

[0340] On the reverse boarding point display unit 601, a list of destinations of passengers for displaying demand prediction and an execution button 611 for executing the demand prediction display are shown. In the example of FIG. 35, three destinations, Haneda Airport, Narita Airport, and TDR, are shown. When displaying the demand prediction of passengers heading for Haneda Airport, execution button 611A is touched (selected); when displaying the demand prediction of passengers heading for Narita Airport, execution button 611B is touched; and when displaying the demand prediction of passengers heading for TDR, execution button 611C is touched.

[0341] On the prediction time display unit 602, the prediction time of the demand prediction is displayed, and similar to the prediction time setting area 42 in FIG. 2, the prediction time of the demand prediction can be changed.

[0342] FIG. 36 shows an example of a demand prediction screen for passengers heading for Haneda Airport when execution button 611A is touched. By operating the detail button 511 or the wide area button 512, the magnification of the map 41 on the demand prediction screen can be changed. The example in FIG. 36 is an example where the scale of the map 41 of the demand prediction first displayed by execution button 611A is set to a high magnification (wide area map). However, the scale of the map 41 of the demand prediction first displayed by execution button 611A can also be set to the same magnification as the map 41 on the demand prediction screen in FIG. 35 at the time of execution. On the demand prediction screen in FIG. 36, the probability corresponding to the passengers going to that destination may be displayed for each boarding point and destination.

[0343] The demand prediction of passengers with a predetermined location as the destination can be classified according to conditions such as time zone, weather, and day of the week (weekday, day before holiday, holiday), and the reverse boarding points that match the conditions at the time of demand prediction execution can be displayed.

[0344] One or more destinations displayed on the reverse pick-up point display unit 601 may be preset locations such as Haneda Airport, Narita Airport, and TDR shown in FIG. 35, or may be locations that are frequently traveled from the current location according to the current location of the taxi 11. In addition to the examples described above, other theme parks, concert venues, event venues, stadiums, etc. can be set as destinations.

[0345] <23. Classification display of demand prediction for vehicle dispatching / cruising / customer waiting at stand> When a user takes the taxi 11, there are three ways to acquire the taxi 11: "trip request", "cruising vehicle", and "customer waiting at stand". "Trip request" is a method of arranging the taxi 11 through the call center or app of the taxi company and having the taxi 11 come to a predetermined location. "Cruising vehicle" is a method of catching a taxi 11 that is moving empty. "Customer waiting at stand" is a method of moving to the pick-up location and taking the taxi 11 that is waiting for customers. When the demand prediction app displays demand prediction on the demand prediction screen, it can display the demand prediction while distinguishing the differences in the boarding methods of "trip request", "cruising vehicle", and "customer waiting at stand". That is, the demand prediction app has a function of displaying the demand prediction while distinguishing the differences in the boarding methods of "trip request", "cruising vehicle", and "customer waiting at stand".

[0346] FIG. 37 shows an example of a demand prediction screen that displays demand prediction while distinguishing the differences in the boarding methods of "trip request", "cruising vehicle", and "customer waiting at stand".

[0347] The demand prediction screen in FIG. 37 is provided with a map 41 on which a demand prediction mesh 63 is superimposed, a prediction time display unit 502 capable of designating the prediction time of the demand prediction, and an additional information display unit 531.

[0348] On the map 41 of the prediction screen, in addition to the detail button 511, the wide area button 512, the full screen display button 513, and the current location button 514, a pick-up display button 641, a streaming display button 642, and a dispatched vehicle display button 643 are provided. Also, pick-up points 645 with different display methods such as colors, patterns, and mark shapes are displayed on the map 41 so as to be able to distinguish the differences in the boarding methods of "dispatched vehicle", "streaming", and "pick-up waiting".

[0349] The pick-up display button 641 is operated when displaying the pick-up point 645 by the boarding method of "pick-up waiting" on the map 41. The streaming display button 642 is operated when displaying the pick-up point 645 by the boarding method of "streaming" on the map 41. The dispatched vehicle display button 643 is operated when displaying the pick-up point 645 by the boarding method of "dispatched vehicle" on the map 41. The pick-up display button 641, the streaming display button 642, and the dispatched vehicle display button 643 are toggle buttons, and each time they are operated, the display of the pick-up point 645 can be turned on and off in units of the specified boarding method. Also, any combination of "dispatched vehicle", "streaming", and "pick-up waiting" is possible. For example, when both "dispatched vehicle" and "streaming" are on, both pick-up points 645 of the boarding methods of "dispatched vehicle" and "streaming" are displayed on the map 41.

[0350] The demand prediction of the pick-up point due to the difference in the boarding methods of "dispatched vehicle", "streaming", or "pick-up waiting" can be obtained by predicting the boarding demand for each difference in the boarding methods of "dispatched vehicle", "streaming", or "pick-up waiting". As for the actual vehicle data for "dispatched vehicle", the actual vehicle data of "actual vehicle" immediately after the status of the taxi 11 becomes "picking up passengers" may be collected. As for the actual vehicle data for "pick-up waiting", the actual vehicle data of "actual vehicle" after the pick-up waiting operation may be collected. The actual vehicle data for "streaming" can be the actual vehicle data other than that for "dispatched vehicle" and "pick-up waiting".

[0351] By displaying the boarding demand (pick-up point) while distinguishing the differences in the boarding methods of "dispatched vehicle", "streaming", or "pick-up waiting", the boarding demand according to the driver's business style can be presented to the driver.

[0352] <Display of Fare Prediction> In FIG. 18, an example of performing a long display 241 for displaying the ratio of long-distance passengers in the target area AR when a predetermined area AR is selected as the target area AR was described.

[0353] Also, in FIG. 19, an example of performing a riding distance display 251 for displaying the average riding distance of passengers boarding in the target area AR and its confidence interval when a predetermined area AR is selected as the target area AR was described.

[0354] Also, in the description of FIG. 19, it was explained that the demand prediction application may show the average fare (fare) and the confidence interval instead of the average riding distance and the confidence interval, and that it is possible to predict the boarding demand for each time zone and weather.

[0355] FIG. 38 shows a display example when the average fare (fare) of passengers boarding in the target area AR and its confidence interval are displayed.

[0356] As shown in FIG. 38, the demand prediction application can perform a fare display 711 for displaying the average fare of passengers boarding in the target area AR and its confidence interval separately from the number of boardings in the entire target area AR.

[0357] In the fare display 711, it is displayed that the average fare for boarding in the target area AR is "2,400 yen", and for example, the confidence interval of the average fare at a confidence level of 70% is "from 1,110 yen to 3,700 yen". The confidence level of the confidence interval is not limited to 70% and can be arbitrarily set such as 80%.

[0358] Also, the fare display 711 indicates that the predicted value of the displayed fare is based on actual vehicle data specialized for the time zone of "18:00 - 18:30", weekdays, rain, and October.

[0359] In this way, by displaying the average fare and its confidence interval for the area of interest AR, a driver can, for example, search for an area AR where a high fare can be expected.

[0360] As shown in FIG. 38, the fare display 711 may be displayed for the area of interest AR, or may be displayed in association with the pinpoint boarding position mark 221 (FIG. 15) to display the average fare and the confidence interval for the pinpoint boarding position.

[0361] In the above description, as described with reference to FIG. 2, each area AR of the demand prediction mesh 63 is displayed separately by color or density according to the degree of boarding demand. However, it may be displayed with the color or density changed according to the expected fare. In this case, a driver can, for example, select a route where a high fare can be expected and perform "flow" driving.

[0362] <25. Display of real-time available taxi numbers> The demand prediction application predicts and displays the boarding demand at a predetermined time (time zone). For example, if 10 boarding demands are predicted in the area of interest AR, but there are 20 taxis 11 that want to pick up passengers there, then the 10 taxis 11 cannot acquire passengers. That is, whether passengers can be acquired also depends on the relationship between demand and supply.

[0363] A taxi company manages in real time the current location of each taxi 11 in operation and its status such as "occupied", "available" or "awaiting passengers" at a dispatching center or the like. By combining the operation data including the current location and status of each taxi 11, which is acquired in real time, with the boarding demand prediction, a driver can conduct efficient business taking into account the above-described relationship between demand and supply. Note that real time includes a certain time lag (for example, about several minutes) for collecting the current location and status information of each taxi 11 in operation and delivering the operation data to the demand prediction application.

[0364] Figure 39 shows an example of a demand prediction screen that displays real-time available taxi information based on operation data.

[0365] The demand prediction screen in Figure 39 includes a map 41 overlaid with a demand prediction mesh 63, a prediction time display section 502 where the prediction time of demand prediction can be specified, and an area information display section 741.

[0366] For each area AR divided by the demand prediction mesh 63 on the map 41, the available taxi information 742 of the taxis 11 existing in that area AR is displayed in real time. The available taxi information 742 of the taxis 11 represents the number of taxis 11 that are currently moving within that area AR and have the status of "available".

[0367] Among the areas AR divided by the demand prediction mesh 63, a highlighted area frame 211 is displayed in the highlighted area AR specified by the driver. In the area information display section 741, the detailed information of the highlighted area frame 211 is displayed as area information. The area information display section 741 displays, for example, the predicted number of boarding demands in that area AR and the long rate of passengers in that area AR. In the example of Figure 39, it shows that there are currently 5 "available" taxis 11 in the highlighted area AR where the highlighted area frame 211 is displayed, while the predicted number of boarding demands is 0.

[0368] Figure 40 shows a display example of a demand prediction screen showing real-time available taxi information when the scale of the map 41 is at a low magnification, in other words, a detailed map display.

[0369] Figure 39 showed a display example of a demand prediction screen showing real-time available taxi information when the scale of the map 41 was at a high magnification, in other words, a wide-area map display. In the case of a wide-area map display, as shown in Figure 39, the number of "available" taxis 11 is displayed as available taxi information in units of the area AR of the demand prediction mesh 63.

[0370] On the other hand, in the case of detailed map display, as shown in FIG. 40, an icon 751 indicating the presence of the taxi 11 with the sign "Available" is displayed at the position where the taxi 11 with the sign "Available" exists.

[0371] In addition, at the pick-up points in the waiting areas, etc., the number of taxis waiting for passengers at the pick-up points in the waiting areas can also be displayed.

[0372] By having the demand prediction application equipped with a function to display real-time available taxi information based on operation data, the driver can select an area AR with a high probability of acquiring passengers and operate in that area, thereby increasing the probability of acquiring passengers.

[0373] Also, when the demand prediction application can acquire operation data, in the above-described recommended route presentation process, it is possible to search for and present a recommended route including real-time available taxi information in the operation data. That is, when selecting a predetermined route as part of the recommended route, the demand prediction application selects a route where the predicted number of passengers is equal to or more than the number of taxis with the sign "Available" as the recommended route, or assigns a large score Sc. Further, if no taxi 11 with the sign "Available" has passed through a predetermined route for a certain period from the present to a certain time before, the demand prediction application may include a process of selecting it as the recommended route because there may be a demand for passengers.

[0374] <Display of the business evaluation on the 26.1st> The demand prediction application can be equipped with a function to output business evaluation information for the driver to evaluate the business of that day after the end of the day's business. The evaluation of the business can be based on excellent drivers, for example, drivers with a high average daily sales. By using excellent drivers as the reference drivers for evaluation, it is possible to provide information to the driver for increasing sales.

[0375] FIG. 41 shows an example of an evaluation screen for outputting the business evaluation information for one day. This evaluation screen is displayed, for example, at the timing when an operation to end the day's business is performed in the demand prediction application.

[0376] On the evaluation screen of FIG. 41, a title display 811 is displayed at the upper part thereof. In the example of FIG. 41, it is displayed as "10 / 11 Duty Score", indicating that it is the evaluation information for the business day of October 11th.

[0377] In addition, on the evaluation screen, an evaluation score display 812, a radar chart 813, a business revenue graph 814, and a route history display button 815 are provided.

[0378] The evaluation score display 812 shows the overall daily evaluation of the driver as a value based on a reference driver being 100 points. By referring to this overall evaluation value, it is possible to confirm how close the driver is to the reference driver.

[0379] The radar chart 813 shows the evaluation results obtained by dividing the overall daily evaluation value of the driver into multiple items. In the example of FIG. 41, it is classified into five items: business revenue, actual vehicle rate, empty vehicle time, number of business trips, and business scope. Business revenue represents the evaluation value from the perspective of business revenue (sales) per actual driving time. The actual vehicle rate represents the evaluation value from the perspective of "actual vehicle" driving time / total driving time. The empty vehicle time represents the evaluation value from the perspective of "empty vehicle" time / total driving time. The number of business trips represents the evaluation value from the perspective of the number of times of carrying passengers. The business scope represents the evaluation value from the perspective of the size of the area traveled.

[0380] The business revenue graph 814 shows the change in the daily business revenue with the horizontal axis representing the daily business hours (business times) and the vertical axis representing the business revenue. The solid line 821 displayed within the business revenue graph 814 represents the actual sales of the driver. On the other hand, the dashed line 822 displayed within the business revenue graph 814 represents the ideal virtual sales based on the driver's actual driving route and the operation data of other taxis 11, etc. For example, if there is the driver's actual driving route and the operation data of other taxis 11, at an intersection, if the driver actually went straight but hypothetically turned left, cases where it is predicted that passengers could have been obtained can be analyzed. By analyzing such assumptions for the actual driving route, the ideal business revenue that could have been obtained with a slight difference in the driving route can be predicted. Such ideal business revenue is displayed as the dashed line 822. Also, at the points (locations) on the dashed line 822 where there was a possibility of increasing the business revenue, comments such as "if turning at the 7th chome of Higashi Ginza" and "if turning at the 5th chome of Shinbashi" are displayed.

[0381] The route history display button 815 is a function that displays the driver's actual daily driving history on a map.

[0382] Figure 42 shows an example of the driving history screen displayed when the route history display button 815 is operated.

[0383] On the driving history screen, as shown in Figure 42, from the start to the end of the day's business, the route that the taxi 11 has traveled, the status of "occupied vehicle", "empty vehicle", or "awaiting passengers", the locations where passengers got on and off and the times, etc. are displayed. In the example of Figure 42, the map 41 is not shown, but actually, it is superimposed and displayed on the map 41. Also, in the example of Figure 42, only a part of the daily driving route is shown, but the driver can confirm all or part of the daily driving route by changing the display magnification of the map 41.

[0384] The operation history screen in FIG. 42 may be displayed on the same screen as the evaluation screen in FIG. 41. By referring to the operation history screen in FIG. 42 in combination with the points that could potentially increase the operating revenue in the operating revenue graph 814 of the evaluation screen in FIG. 41, it can be used as a reference for the next business operation.

[0385] <27. Display of additional information considering distance and direction> When displaying additional information such as train operation information, event information, and weather information, as shown in the example of FIG. 32, an additional information display section 531 may be provided in an area different from the display area of the map 41 and displayed. However, considering distance and direction, the additional information may also be displayed on the map 41.

[0386] FIG. 43 shows an example of a demand prediction screen on which additional information is displayed considering distance and direction.

[0387] In the example of FIG. 43, additional information 831 and additional information 832 are displayed on the map 41.

[0388] Additional information 831 is information notifying that an operation suspension has occurred at Tamachi Station on the Yamanote Line. Additional information 831 is displayed at a position corresponding to the direction of Tamachi Station based on the own vehicle position mark 505 indicating the current location of the taxi 11. In the example of FIG. 43, since Tamachi Station is outside the display area of the map 41 in the demand prediction screen, only additional information 831 is displayed at a position corresponding to the direction of Tamachi Station. However, if Tamachi Station is on the map 41 in the demand prediction screen, symbols such as × or △ indicating that an operation suspension has occurred, together with additional information 831, are displayed on the part of Tamachi Station on the map 41. Also, only a symbol may be displayed as additional information 831, and detailed information may be displayed when the symbol is tapped (selected).

[0389] Additional information 832 is information notifying that a train delay has occurred at Togoshi Station on the Toei Asakusa Line. Additional information 832 is displayed at a position corresponding to the direction of Togoshi Station based on the own vehicle position mark 505 indicating the current location of the taxi 11.

[0390] The direction based on the own-vehicle position mark 505 may be a detailed angle such as in 1-degree units, or an angle converged within a predetermined range such as 4 directions or 8 directions.

[0391] Similarly for the distance, when the distance to the position regarding the additional information is far from the current location, it is displayed at a position far from the own-vehicle position mark 505, and when it is close, it is displayed at a position close to the own-vehicle position mark 505.

[0392] For example, when the additional information is information regarding a train delay, the station calculated to be affected can be used as the position regarding the additional information, and the direction and distance from the own-vehicle position mark 505 can be calculated.

[0393] For example, when the additional information is information regarding an event, the location where the event is held can be used as the position regarding the additional information, and the direction and distance from the own-vehicle position mark 505 can be calculated.

[0394] For example, when the additional information is information regarding the weather such as a guerrilla heavy rain, the location where the meteorological phenomenon occurs can be used as the position regarding the additional information, and the direction and distance from the own-vehicle position mark 505 can be calculated.

[0395] As described above, by displaying the additional information on the map 41 according to the distance and direction of the additional information and presenting it to the driver, the driver can intuitively understand the additional information.

[0396] Note that in the demand prediction screen of FIG. 43, the display of the boarding point 645 (similarly for the demand point 451) displayed on the map 41 and the display of the prediction time display section 502 are omitted.

[0397] <28. Display of Information According to Travel Direction> Prediction information such as the pick-up point to be predicted, additional information such as event information and train delay information shown in FIG. 43, etc., information in the traveling direction of the taxi 11 is important, but information in the opposite direction of the traveling direction is not so important. The same applies to the route information in FIG. 41.

[0398] Therefore, when the demand prediction application displays the map 41 on the demand prediction screen, it can display the information in the traveling direction with a larger amount of information to be displayed than the information in the opposite direction of the traveling direction.

[0399] A in FIG. 44 shows an example of display in the head-up mode where the traveling direction of the taxi 11 is set to the upper side (upper edge side) of the screen.

[0400] In the head-up mode, for the entire area of the map 41, the own vehicle position mark 505 is arranged such that, in the left-right direction, the right area R1 and the left area L1 are the same or substantially the same, and in the up-down direction, the upper area U1 is larger than the lower area D1.

[0401] B in FIG. 44 shows an example of display in the north-up mode where the north direction is set to the upper side (upper edge side) of the screen regardless of the traveling direction of the taxi 11.

[0402] In the north-up mode, the distribution of the right area R1 and the left area L1, and the distribution of the upper area U1 and the lower area D1 differ depending on the traveling direction of the taxi 11. B in FIG. 44 shows an example of display when the traveling direction of the taxi 11 is in the northeast direction. In this case, in the left-right direction, the right area R1 is larger than the left area L1, and in the up-down direction, the upper area U1 is larger than the lower area D1.

[0403] Although other illustrations are omitted, for example, when the traveling direction of the taxi 11 is in the southwest direction, in the left-right direction, the left area L1 is larger than the right area R1, and in the up-down direction, the lower area D1 is larger than the upper area U1.

[0404] As described above, by displaying the information on the traveling direction in such a manner that the amount of information to be displayed is larger than that of the information in the reverse direction of the traveling direction, useful information can be displayed for the driver.

[0405] <29. Computer Configuration Example> The above-described series of processes can be executed by hardware or by software. When the series of processes is executed by software, the program constituting the software is installed in a computer. Here, the computer includes a microcomputer incorporated in dedicated hardware, or a general-purpose personal computer or the like that can execute various functions by installing various programs.

[0406] FIG. 45 is a block diagram showing a configuration example of the hardware of a computer when each process executed by the server 12, the vehicle management device 22, or the terminal device 23 is executed by the computer according to a program.

[0407] In the computer, a CPU (Central Processing Unit) 301, a ROM (Read Only Memory) 302, and a RAM (Random Access Memory) 303 are interconnected by a bus 304.

[0408] Further connected to the bus 304 is an input / output interface 305. Connected to the input / output interface 305 are an input unit 306, an output unit 307, a storage unit 308, a communication unit 309, and a drive 310.

[0409] The input unit 306 includes operation buttons, a keyboard, a mouse, a microphone, a touch panel, input terminals, and the like. The output unit 307 includes a display, a speaker, output terminals, and the like. The storage unit 308 includes a hard disk, a RAM disk, a non-volatile memory, and the like. The communication unit 309 includes a network interface and the like. The drive 310 drives a removable recording medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory.

[0410] In the computer configured as described above, for example, the CPU 301 loads and executes a program stored in the storage unit 308 via the input / output interface 305 and the bus 304 into the RAM 303, thereby performing the above-described series of processes. The RAM 303 also appropriately stores data and the like necessary for the CPU 301 to execute various processes.

[0411] The program executed by the computer (CPU 301) can be recorded and provided on a removable recording medium 311 as a package medium or the like. Further, the program can be provided via a wired or wireless transmission medium such as a local area network, the Internet, or digital satellite broadcasting.

[0412] In the computer, the program can be installed in the storage unit 308 via the input / output interface 305 by attaching the removable recording medium 311 to the drive 310. Further, the program can be received by the communication unit 309 via a wired or wireless transmission medium and installed in the storage unit 308. Additionally, the program can be installed in advance in the ROM 302 or the storage unit 308.

[0413] In this specification, a system refers to a collection of a plurality of components (devices, modules (parts), etc.), regardless of whether all the components are in the same housing. Therefore, a plurality of devices housed in separate enclosures and connected via a network, and a single device with a plurality of modules housed in one enclosure are both systems.

[0414] Also, in this specification, the steps described in the flowchart may be executed in parallel or at a necessary timing such as when a call is made, even if they are not necessarily processed in chronological order, as long as they are executed in chronological order according to the described order.

[0415] The embodiments of the present technology are not limited to the above-described embodiments, and various modifications are possible without departing from the gist of the present technology.

[0416] The above-described embodiments have been described by taking as an example a prediction system that predicts the demand for taxi rides as a business vehicle. However, the present technology can also be applied to other business vehicles that carry passengers (people), specifically, buses, trains, airplanes, ships, helicopters, etc., or to systems that predict the demand for business vehicles that carry goods (cargo), such as trucks and dump trucks. Also, the business vehicle may be an unmanned transport vehicle such as a drone.

[0417] For example, a form in which all or part of the above-described embodiments are appropriately combined can be adopted.

[0418] For example, the present technology can take a cloud computing configuration in which one function is shared and jointly processed by a plurality of devices via a network.

[0419] Also, each step described in the above flowchart can be executed by one device or can be shared and executed by a plurality of devices.

[0420] Furthermore, when a single step includes a plurality of processes, the plurality of processes included in that single step can be executed by a single device or can be shared and executed by a plurality of devices.

[0421] Note that the effects described in this specification are merely examples and are not limiting. There may be effects other than those described in this specification.

[0422] Note that the present technology can also adopt the following configurations. (1) An information processing apparatus including a display control unit that divides the business area of a business vehicle into a plurality of areas, predicts the boarding demand for each area, and displays, as a prediction result, the moving direction and moving distance of passengers in a target area, which is one of the plurality of areas, on a display unit. Information processing apparatus. (2) The display control unit displays the moving direction and the moving distance on the display unit as arrows pointing outward from the target area. The information processing apparatus according to (1) above. (3) The direction of the arrow represents the moving direction, the length of the arrow represents the moving distance, and the thickness of the arrow represents the boarding ratio in the moving direction in all directions. The information processing apparatus according to (2) above. (4) The display control unit displays, as a prediction result, the boarding positions with high boarding frequency and the number of boardings at those boarding positions within the target area on the display unit. The information processing apparatus according to any one of (1) to (3) above. (5) The display control unit displays, as a prediction result, the boarding positions with high boarding frequency and the number of boardings at those boarding positions within the target area, and the number of boardings in the target area, on the display unit. The information processing apparatus according to any one of (1) to (4) above. (6) The display control unit causes the display unit to display, as a prediction result, a predetermined boarding position within the target area, the number of passengers boarding at the boarding position, and the time required to board the waiting passengers at the boarding position. The information processing apparatus according to any one of (1) to (5) above. (7) The display control unit causes the display unit to display, as a prediction result, the ratio of boardings in the target area where the boarding distance is equal to or greater than a predetermined distance. The information processing apparatus according to any one of (1) to (6) above. (8) The display control unit divides the boarding distance in the target area into a plurality of sections, and causes the display unit to display, as a prediction result, the ratio of boardings for each divided section. The information processing apparatus according to any one of (1) to (7) above. (9) The display control unit also causes the display unit to display, as a prediction result, the ratio of boardings for each section of the entire plurality of areas. The information processing apparatus according to (8) above. (10) The display control unit causes the display unit to display, as a prediction result, the time and fare required for the movement to the destination, and the time and fare required for the movement for each divided unit obtained by dividing the movement route to the destination into predetermined units. The information processing apparatus according to any one of (1) to (9) above. (11) The display control unit causes the display unit to display, as a prediction result, the average boarding distance of the passengers in the target area and its confidence interval. The information processing apparatus according to any one of (1) to (10) above. (12) The display control unit causes the display unit to display, as a prediction result, the average boarding fare of the passengers in the target area and its confidence interval. The information processing apparatus according to any one of (1) to (10) above. (13) The display control unit causes the display unit to display, as a prediction result, the average boarding time of the passengers in the target area and its confidence interval. The information processing apparatus according to any one of (1) to (10) above. (14) Among the plurality of divided areas, for the plurality of adjacent areas where the number of boardings in the adjacent areas is equal to or less than a predetermined threshold, the display control unit combines them as one area and causes the display unit to display the prediction result. The information processing apparatus according to any one of (1) to (10) above. (15) The information processing apparatus further includes a notification unit that notifies, by sound, a location with high boarding demand in the traveling direction. The information processing apparatus according to any one of (1) to (14) above. (16) The notification unit notifies, in units of the areas, a location with high boarding demand. The information processing apparatus according to (15) above. (17) The notification unit changes the type of sound according to the scale of the prediction result displayed on the display unit and notifies a location with high boarding demand. The information processing apparatus according to (15) or (16) above. (18) The notification unit changes the type of sound according to the distance to a location with high boarding demand and notifies the location with high boarding demand. The information processing apparatus according to any one of (15) to (17) above. (19) The notification unit changes the type of sound according to the magnitude of the boarding demand and notifies a location with high boarding demand. The information processing apparatus according to any one of (15) to (18) above. (20) The notification unit notifies, by sound, every time it passes through a location with high boarding demand. The information processing apparatus according to any one of (15) to (19) above. (21) The notification unit is the information processing device according to any one of (15) to (20) above, which turns the notification on and off in conjunction with the status of "occupied vehicle" or "empty vehicle". (22) The sound is an effect sound or a voice message. The information processing device according to any one of (15) to (21) above. (23) The notification unit further notifies at least one of train operation information, event information, and weather information by voice message. The information processing device according to any one of (15) to (22) above. (24) The display control unit causes the display unit to display a recommended route based on a prediction result of predicting the boarding demand. The information processing device according to any one of (1) to (23) above. (25) The display control unit searches for a route to the set destination and causes the display unit to display it as the recommended route. The information processing device according to (24) above. (26) The destination is an area or place that the driver is good at. The information processing device according to (25) above. (27) The area that the driver is good at is an area where the time the driver has moved is a predetermined value or more. The information processing device according to (26) above. (28) The area that the driver is good at is an area where the time the driver has carried passengers is a predetermined value or more. The information processing device according to (26) or (27) above. (29) The display unit of the area that the driver is good at is a city, town, or village, The display unit of the place that the driver is good at is a boarding position with a large number of passengers. The information processing device according to any one of (26) to (28) above. (30) The display control unit searches for a route that passes through a location where a boarding demand is predicted near the current location, and causes the display unit to display the route as the recommended route. The information processing apparatus according to (24) above. (31) The display control unit causes the display unit to display, as the recommended route, a route with a high total score obtained by summing the scores of the routes to be passed through. The information processing apparatus according to any one of (24) to (30) above. (32) The total score is calculated by summing the scores for each area based on the level of boarding demand. The information processing apparatus according to (31) above. (33) The total score is calculated by summing the scores for each location where a boarding demand is predicted. The information processing apparatus according to (31) above. (34) The score for a location where it is necessary to cross the oncoming lane is set lower than the score for a location where it is not necessary to cross the oncoming lane. The information processing apparatus according to any one of (31) to (33) above. (35) The closer the predicted result of the passenger's moving direction is to the direction of the destination, the higher the score. The information processing apparatus according to any one of (31) to (34) above. (36) The prediction time for predicting the boarding demand in a predetermined area when searching for the recommended route is changed according to the distance from the current location. The information processing apparatus according to any one of (24) to (35) above. (37) The display control unit causes the display unit to display, as the recommended route, a route for which the predicted number of boarding demands is equal to or greater than the number of empty commercial vehicles. The information processing apparatus according to any one of (24) to (36) above. (38) The display control unit further causes the display unit to display a no-boarding area. The information processing apparatus according to any one of (1) to (37) above. (39) The display control unit further causes the display unit to display the location where "awaiting boarding" is performed. The information processing apparatus according to any one of (1) to (38) above. (40) The display control unit further causes the display unit to display the name of the company that can use the location where the "awaiting boarding" is performed. The information processing apparatus according to (39) above. (41) The display control unit further causes the display unit to display the locations where other "awaiting boarding" is performed that can be used by the business vehicle. The information processing apparatus according to (39) or (40) above. (42) The display control unit further causes the display unit to display the train times of the stations on the map displayed on the display unit. The information processing apparatus according to any one of (1) to (41) above. (43) The display control unit causes the train times of a plurality of the stations to be listed in the order of arrival times or in the order of the distances from the vehicle's position to the stations. The information processing apparatus according to (42) above. (44) The display control unit displays only the boarding point where the passenger boarded, with the designated location as the destination of the passenger. The information processing apparatus according to any one of (1) to (43) above. (45) The display control unit displays a plurality of the destinations and displays only the boarding points of the selected destination. The information processing apparatus according to (44) above. (46) The display control unit further causes the display unit to display a prediction of the demand for boarding by distinguishing the differences in the boarding methods of "vehicle allocation", "skipping", and "awaiting boarding". The information processing apparatus according to any one of (1) to (45) above. (47) The display control unit turns on and off the display of the predicted demand for boarding in units of boarding methods of "dispatch", "through", and "waiting". The information processing apparatus according to (46) above. (48) The display control unit causes the display unit to display, as a prediction result, the average fare of the passengers in the area of interest and its confidence interval. The information processing apparatus according to any one of (1) to (47) above. (49) The display control unit causes the display unit to display, as a prediction result, the average fare of the passengers at a predetermined boarding position and its confidence interval. The information processing apparatus according to any one of (1) to (47) above. (50) The display control unit further causes the display unit to display real-time empty vehicle information. The information processing apparatus according to any one of (1) to (49) above. (51) The display control unit causes the display unit to display the number of empty vehicles for each area as the empty vehicle information. The information processing apparatus according to (50) above. (52) The display control unit causes the display unit to display icons of the empty commercial vehicles as the empty vehicle information. The information processing apparatus according to any one of (50) or (51) above. (53) After the end of the business for one day, the display control unit further causes the display unit to display business evaluation information for evaluating the business of that day. The information processing apparatus according to any one of (1) to (52) above. (54) The display control unit causes the display unit to display the actual sales and the virtual sales as part of the business evaluation information. The information processing apparatus according to (53) above. (55) The display control unit causes the display unit to display an operation route including the status of the actual vehicle and the empty vehicle as part of the business evaluation information. The information processing apparatus according to the above (53) or (54). (56) The display control unit further causes the additional information to be displayed on the map of the display unit according to the distance or direction of the additional information. The information processing apparatus according to any one of the above (1) to (55). (57) The display control unit causes the information in the traveling direction to be displayed on the display unit such that the amount of information to be displayed is larger than that of the information in the reverse direction of the traveling direction. The information processing apparatus according to any one of the above (1) to (56). (58) An information processing apparatus divides the business area of a business vehicle into a plurality of areas, predicts the boarding demand for each area, and causes the moving direction and moving distance of passengers in a target area, which is the target area among the plurality of areas, to be displayed on a display unit as a prediction result. An information processing method. (59) Causes a computer to divide the business area of a business vehicle into a plurality of areas, predict the boarding demand for each area, and cause the moving direction and moving distance of passengers in a target area, which is the target area among the plurality of areas, to be displayed on a display unit as a prediction result. A program for causing the execution of the process.

Explanation of Reference Numerals

[0423] 1 Prediction system, 11 Taxi, 12 Server, 22 Vehicle management device, 23 Terminal device, 63 Demand prediction mesh, 121 Control unit, 131 Data generation unit, 132 Learning unit, 133 Prediction unit, 141 Control unit, 142 Operation unit, 143 Display unit, 145 Speaker, 146 Microphone, 211 Focus area frame, 212 Arrow, 221 Pinpoint boarding position mark, 222 Boarding number display, 223 Pickup start button, 224 Pickup display, 241 Long display, 251 Boarding distance display, 261 Individual display, 262 Destination display, 301 CPU, 302 ROM, 303 RAM, 306 Input unit, 307 Output unit, 308 Storage unit, 309 Communication unit, 310 Drive, 521 Boarding prohibited area display, 531 Additional information display unit, 551 Pickup location display, 553 Detailed information, 581 List display unit, 582 Sort button, 601 Reverse pickup point display unit, 641 Pickup display button, 642 Scroll display button, 643 Dispatching display button, 711 Fare display, 741 Area information display unit, 742 Available vehicle information, 751 Icon, 812 Evaluation score display, 813 Radar chart, 814 Revenue graph, 815 Route history display button, 821, 822 Solid line, 831, 832 Additional information

Claims

1. A display control unit that controls a display unit to display a map including a search result of a recommended route among the plurality of routes based on a total score obtained by summing scores on routes through which a plurality of routes pass respectively based on predicted boarding demand data of a business vehicle, and operation data of each of the business vehicles including empty vehicle information. An information processing apparatus.

2. The display control unit causes the display unit to display, as the recommended route, a route among the plurality of routes having a high total score. The information processing apparatus according to claim 1.

3. The total score is calculated by summing the scores for each boarding point, which is a place where boarding demand is predicted. The information processing apparatus according to claim 1.

4. The score for the boarding point is set based on at least one of the date and time of boarding, the number of boarding times, the total boarding time, the average boarding time, and the ratio of boardings with a long boarding distance at the boarding point. The information processing apparatus according to claim 3.

5. The score is set based on the width of the road in the passing route. The information processing apparatus according to claim 1.

6. The total score is calculated by summing the scores for each area obtained by dividing the business area of the business vehicle based on the level of boarding demand. The information processing apparatus according to claim 1.

7. The score for a place where it is necessary to cross the oncoming lane is set lower than the score for a place where it is not necessary to cross the oncoming lane. The information processing apparatus according to claim 1.

8. The closer the predicted result of the passenger's moving direction is to the direction of the destination, the greater the score. The information processing apparatus according to claim 1.

9. The display control unit searches for a route that passes through a place where a riding demand is predicted from among the routes leading to the set destination, and causes the display unit to display the route as the recommended route. The information processing apparatus according to claim 1.

10. The destination is an area or a place that the driver is good at. The information processing apparatus according to claim 9.

11. The area that the driver is good at is an area where the time the driver has moved is equal to or more than a predetermined value. The information processing apparatus according to claim 10.

12. The area that the driver is good at is an area where the time the driver has carried passengers is equal to or more than a predetermined value. The information processing apparatus according to claim 10.

13. The display unit of the area that the driver is good at is a city, town, or village, and The display unit of the place that the driver is good at is a boarding position with a large number of boardings. The information processing apparatus according to claim 10.

14. The display control unit searches for a route that passes through a place where a riding demand is predicted in the vicinity of the current location, and causes the display unit to display the route as the recommended route. The information processing apparatus according to claim 1.

15. The prediction time for predicting the riding demand in a predetermined area when searching for the recommended route is changed according to the distance from the current location. The information processing apparatus according to claim 1.

16. A route in which the predicted number of riding demands is equal to or more than the number of the empty business vehicles is caused to be displayed on the display unit as the recommended route. The information processing apparatus according to claim 1.

17. The display control unit causes the display unit to display, as the recommended route, a route that the empty business vehicle has not passed through for a certain period of time. The information processing apparatus according to claim 1.

18. The display control unit searches for the recommended route based on either a mode of searching for the recommended route based on a destination or a mode of searching for the recommended route based on the current location of the business vehicle without a specific destination, and causes the display unit to display it. The information processing apparatus according to claim 1.

19. An information processing apparatus, controls a display unit to display a map including a search result of a recommended route among the plurality of routes based on a total score obtained by summing scores on routes through which the plurality of routes respectively pass based on passenger demand prediction data of a business vehicle and operation data of each of the business vehicles including empty vehicle information. An information processing method.

20. causing a computer to, control a display unit to display a map including a search result of a recommended route among the plurality of routes based on a total score obtained by summing scores on routes through which the plurality of routes respectively pass based on passenger demand prediction data of a business vehicle and operation data of each of the business vehicles including empty vehicle information. A program for causing the processing to be executed.

Images