Vehicle dispatch management device and vehicle dispatch management method

Abstract

In a ride-hailing service that pre-deploys vehicles according to the predicted vehicle demand within a target area, vehicles are deployed while taking into account fluctuations in fairness between sub-areas. [Solution] The present invention is a dispatch management device for managing dispatch services within a target area. The device calculates a forecast value of vehicle demand based on dispatch requests in a predetermined time period in the future, based on actual values ​​of vehicle demand based on past dispatch requests, for each sub-area within the target area. It also calculates a fairness index regarding fairness among users enjoying the dispatch service based on actual dispatch times based on past dispatch requests for the predetermined time period in each sub-area. For a specific sub-area where the device determines that the fairness index should be applied based on the forecast value of vehicle demand and the fairness index, it determines which vehicles should be deployed based on the forecast value of vehicle demand and the fairness index, and issues instructions to the vehicles for deployment.

Description

Technical Field

[0001] The present invention relates to a vehicle allocation management device and a vehicle allocation management method.

Background Art

[0002] There is known a vehicle allocation service that allocates (assigns) an appropriate vehicle to a user in response to the user's vehicle allocation request. In the vehicle allocation service, fleet management that manages vehicle allocation so as to improve the convenience of users while increasing the operating rate of vehicles in the service target area is important. For example, Patent Document 1 below discloses a technique for efficiently performing vehicle allocation by eliminating waste in reallocation when the ride demand data and the actual ride demand deviate.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the technique disclosed in Patent Document 1 above, when there is a large regional bias in the regional distribution of vehicle demand during a predetermined time period in the service target area, the number of vehicles allocated to areas with low vehicle allocation demand tends to be small. For users in such areas, it is difficult to have a vehicle allocated with a short waiting time even if a vehicle allocation request is made, so the convenience is impaired and inconvenient from the on-demand perspective. In addition, when the vehicle allocation service plays a role as a regional public transportation agency, it is necessary to ensure a certain degree of fairness between regions from the perspective of convenience.

[0005] In addition, in ensuring fairness between regions, depending on a certain time period in a certain region, it may be possible to provide a sufficient vehicle allocation service without considering fairness.

[0006] Therefore, the present invention aims to propose a dispatch management device and dispatch management method that, in a dispatch service in which vehicles are pre-deployed to each region (sub-area) within a service area in accordance with the predicted dispatch demand for each region, appropriately judges circumstances that should be considered from the perspective of fairness and circumstances that do not, and determines the number of vehicles to be dispatched to the region. [Means for solving the problem]

[0007] The present invention, which solves the above problems, is comprised of the following inventive features or technical characteristics.

[0008] According to a certain perspective, the present invention is a dispatch management device that manages vehicle dispatch services in a target area in response to user dispatch requests. The dispatch management device includes: a vehicle demand forecasting unit that calculates a predicted value of vehicle demand based on future dispatch requests in a predetermined time period, based on actual values ​​of vehicle demand based on past dispatch requests, for each of a plurality of sub-areas in the target area; a fairness index calculation unit that calculates a fairness index relating to fairness among users enjoying the dispatch service, based on actual values ​​of time related to dispatch based on past dispatch requests, for each of the predetermined time periods in each of the sub-areas; a vehicle deployment determination unit that determines which vehicles to be deployed for the predetermined time period in each of the sub-areas based on the calculated predicted value of vehicle demand; and a vehicle instruction unit that gives instructions to the vehicles for deployment to the sub-areas. The fairness index calculation unit determines whether or not to apply the fairness index to each of the sub-areas based on fairness fluctuations caused by predetermined external factors. Furthermore, the vehicle deployment determination unit determines which vehicles to deploy to a specific sub-area among the plurality of sub-areas to which the fairness index is determined to apply, based on the calculated fairness index.

[0009] Furthermore, the present invention also constitutes a dispatch management method using a dispatch management device that manages vehicle dispatch in response to dispatch requests, a computer program for executing the method, and a recording medium that non-temporarily stores the same. [Effects of the Invention]

[0010] According to the present invention, in a ride-hailing service that pre-deploys vehicles to each region (sub-area) within a service area in accordance with the predicted ride-hailing demand for each region, it becomes possible to ensure fairness between regions without compromising user convenience.

[0011] Other technical features, objectives, and effects or advantages of the present invention will be illustrated by the following embodiments described with reference to the accompanying drawings. The effects described herein are illustrative and not limiting, and other effects may also occur. [Brief explanation of the drawing]

[0012] [Figure 1] Figure 1 is a diagram illustrating an example of a schematic configuration of a dispatch management system according to one embodiment of the present invention. [Figure 2] Figure 2 is a block diagram showing an example of a functional configuration model of a dispatch management device according to one embodiment of the present invention. [Figure 3] Figure 3 is a block diagram showing an example of the specific configuration of the dispatch planning unit in a dispatch management device according to one embodiment of the present invention. [Figure 4] Figure 4 is a diagram illustrating the predicted values ​​of vehicle demand for each sub-area during each predetermined time period in a dispatch management device according to one embodiment of the present invention. [Figure 5] Figure 5 is a diagram illustrating the fairness indicators for each predetermined time period in each sub-area of ​​a dispatch management device according to one embodiment of the present invention. [Figure 6] Figure 6 is a diagram illustrating an example of fairness indicators for each sub-area in a dispatch management device according to one embodiment of the present invention. [Figure 7] Figure 7 is a diagram illustrating an example of a fairness index for a block area in a dispatch management device according to one embodiment of the present invention. [Figure 8]Figure 8 is a diagram illustrating an example of a fairness index for a block area in a dispatch management device according to one embodiment of the present invention. [Figure 9A] Figure 9A is a flowchart showing an example of vehicle deployment processing in a vehicle dispatch management device according to one embodiment of the present invention. [Figure 9B] Figure 9B is a flowchart showing an example of vehicle deployment processing in a vehicle dispatch management device according to one embodiment of the present invention. [Figure 10] Figure 10 is a flowchart showing an example of the fairness index calculation process in a dispatch management device according to one embodiment of the present invention. [Modes for carrying out the invention]

[0013] Figure 1 is a diagram illustrating an example of a schematic configuration of a dispatch management system according to one embodiment of the present invention. As shown in the figure, the dispatch management system 1 includes a dispatch management device 10, an information and communication terminal device 20, and vehicles V, which are connected to each other so as to be able to communicate with each other via a communication network N. The dispatch management system 1 may also be connected via the communication network N to a road traffic information management system 30 that provides various traffic information in real time. The dispatch management system 1 provides, for example, a dispatch service using multiple vehicles V in response to a user's dispatch request for a certain area. In this disclosure, the vehicles V may be manned vehicles driven by a driver, or unmanned (autonomous) vehicles without a driver.

[0014] The vehicle allocation management device 10 is a computing device that comprehensively manages the vehicle allocation service by a plurality of vehicles V in the target area of the vehicle allocation service. The target area is composed of a plurality of sub-areas. The sub-areas may be formed in a grid pattern according to latitude and longitude, or may be formed according to geographical administrative divisions. The vehicle allocation management device 10 realizes the vehicle allocation service, for example, by implementing a vehicle allocation management server program and executing the vehicle allocation management server program under the control of a processor. In particular, the vehicle allocation management server program of the present disclosure causes the vehicle allocation management device 10 to calculate, for each sub-area, a predicted value of vehicle demand in a predetermined future time period based on the actual value of vehicle demand based on past vehicle allocation requests, and to calculate a fairness index regarding fairness among users who enjoy the vehicle allocation service based on the actual value of time related to vehicle allocation based on past vehicle allocation requests. Based on the calculated predicted value of vehicle demand, it determines the vehicles to be deployed in each sub-area for a predetermined time period, and further determines whether the fairness index should be applied based on the variation of fairness. When it is determined that the fairness index should be applied, it performs a process of determining the vehicles to be deployed for a specific sub-area among the plurality of sub-areas based on the fairness index, and includes a sub-program or module for this purpose.

[0015] The vehicle allocation management device 10 includes various databases 12 that store and manage the data necessary to realize such a vehicle allocation service. The database 12 may include, for example, a user information database 12a, a road map information database 12b, a vehicle information database 12c, a vehicle allocation plan database 12d, and a vehicle allocation performance information database 12e (see FIG. 2). The database 12 may be configured as a part of the vehicle allocation management device 10, or all or part of it may be configured separately from the vehicle allocation management device 10.

[0016] In general terms, the dispatch management device 10 receives dispatch requests from users who wish to be dispatched for travel to their destination. Based on the received dispatch request, it refers to the road map information database 12b to determine the travel route to the destination, and also refers to the vehicle information database 12c to select a vehicle V to assign (dispatch) to the user from among multiple vehicles V, creates a dispatch plan, and proposes the created dispatch plan to the user. The dispatch management device 10 confirms the dispatch plan based on the user's acceptance action, and issues dispatch instructions to the vehicle V according to the confirmed dispatch plan. If the user rejects the dispatch plan, the created dispatch plan is not applied to the user, and the user will, for example, make a new dispatch request with modified conditions. The user's rejection action may also be used to calculate a fairness index.

[0017] In this disclosure, the dispatch management device 10 calculates a forecast value of vehicle demand for a predetermined time period in the future for each sub-area based on actual values ​​of vehicle demand based on past dispatch requests, and calculates a fairness index regarding fairness among users enjoying the dispatch service based on actual values ​​of dispatch time based on past dispatch requests. Furthermore, the dispatch management device 10 determines which vehicles should be deployed to each sub-area for a predetermined time period based on the calculated forecast value of vehicle demand. In addition, the dispatch management device 10 further determines whether or not to apply the fairness index based on the circumstances of fluctuations in fairness, and if it determines that it should be applied, it determines which vehicles should be deployed to a specific sub-area among the multiple sub-areas based on the calculated fairness index. This means that in a ride-hailing service where vehicles V are pre-deployed to each sub-area within the service area according to the predicted vehicle demand, vehicles V will be deployed even in sub-areas where vehicle demand is low and vehicles V would not normally be deployed, ensuring user convenience from the standpoint of public interest and fairness. Furthermore, in cases where sufficient ride-hailing service can be provided in a certain sub-area at a certain time without considering fairness, vehicles V can be deployed according to the predicted vehicle demand, thereby improving ride-hailing efficiency.

[0018] In the present disclosure, "vehicle dispatch" refers to a form of transporting a user to a destination by selecting a predetermined vehicle V in response to a user's vehicle dispatch request and following a vehicle dispatch plan that determines a driving route from the user's departure location (boarding location) to the destination (alighting location). "Vehicle deployment" refers to a form of moving a vehicle to a predetermined waiting location within a sub-area for such vehicle dispatch.

[0019] The vehicle V is a vehicle registered in the vehicle information database 12c and provided for user use in the vehicle dispatch service. The vehicle V may be a manned vehicle driven by a driver, or alternatively, an unmanned (autonomous driving) vehicle without a driver, regardless of its level of automated driving.

[0020] The vehicle V includes a control device CTL that controls the vehicle V itself or its on-board devices. The control device CTL includes a processor, a memory, etc., and realizes various functions by executing a vehicle control program. Further, the control device CTL can acquire the geographical position information (vehicle position information) of the vehicle V in real time by a positioning system such as GPS, use the acquired vehicle position information for the navigation function, and transmit it to the vehicle dispatch management device 10. Also, the vehicle V acquires driving management information necessary for driving from various sensors under the control of the control device CTL and transmits this to the vehicle dispatch management device 10. The vehicle dispatch management device 10 can remotely monitor and control the driving of the vehicle V according to the driving management information transmitted from the vehicle V.

[0021] The information and communication terminal device 20 is a computing device for users of a ride-hailing service to operate a user interface and make ride-hailing requests. The information and communication terminal device 20 is, for example, a smartphone, a tablet computer, a notebook personal computer, a desktop computer, etc., but is not limited to these. The information and communication terminal device 20 may implement, for example, a ride-hailing management client program (a so-called ride-hailing app). Typically, the information and communication terminal device 20 has a positioning function that acquires its own geographical location information in real time using a positioning system such as GPS. Under the control of the processor, the information and communication terminal device 20 enables the user to use the ride-hailing service of the ride-hailing management device 10 by executing the ride-hailing app. While the ride-hailing app is running, the information and communication terminal device 20 transmits the location information acquired from the positioning system to the ride-hailing management device 10 as user location information.

[0022] Figure 2 is a block diagram showing an example of a functional configuration model of a dispatch management device according to one embodiment of the present invention. As shown in the figure, the dispatch management device 10 may be configured as a functional configuration model that includes various functional components such as a front-end processing unit 110, a dispatch planning unit 130, and a vehicle communication unit 120. The dispatch management device 10 is configured to be able to access various databases 12. Such a functional model is realized by the dispatch management device 10 executing a dispatch management server program under the control of a processor, thereby cooperating with various hardware resources. The functional configuration model shown herein is an example, and all or part of the functions of one functional component may be realized by other functional components.

[0023] The user information database 12a stores information about users who use the ride-hailing service (hereinafter referred to as "user information"). User information includes, for example, the user's user ID and password, email address, personal attributes (gender, age, and address, etc.), and usage history. User information is registered in the user information database 12a under the control of the ride-hailing management device 10 when the user enters predetermined information, for example, when the ride-hailing app is first launched.

[0024] The road map information database 12b stores road map information for the service area of ​​the ride-hailing service. The road map information includes, for example, map data related to addresses, place names, road networks (links and nodes), pick-up and drop-off locations (bus stops), waiting areas, and various facilities. Waiting areas are provided, for example, one or more per sub-area. The road map information also includes information necessary for searching for a driving route, such as road information and feature information. The road map information is used, for example, to determine the driving route of vehicle V indicated in the ride-hailing plan.

[0025] The vehicle information database 12c stores information about the vehicle V provided for use by the user (vehicle information). The vehicle information includes information such as vehicle ID, vehicle attributes (vehicle registration number, vehicle type, and maximum passenger capacity / maximum load capacity, level of automation, etc.), vehicle characteristics, other specifications (overall length, overall width, vehicle height, and weight, etc.), current dispatch plan, current location, and current status (service status, battery remaining capacity (driving range), and number of passengers, etc.). The latest vehicle information is obtained from the vehicle V via the vehicle communication unit 120 and stored in the vehicle information database 12c. The vehicle information is used, for example, to select the optimal vehicle V to dispatch to the user.

[0026] The dispatch plan database 12d stores the dispatch plan for each vehicle V, which is created by the dispatch plan unit 130 based on the dispatch request. The dispatch plan includes the route of vehicle V, which is defined from the vehicle V's current location (e.g., a starting waiting location) to a final waiting location (not necessarily the same as the initial waiting location), passing through several points (stops) along the way where users board and alight. If vehicle V is waiting for a dispatch request, a new dispatch plan is created based on the received dispatch request. If vehicle V is currently operating with a user on board, the dispatch plan may be updated based on the dispatch request.

[0027] The dispatch performance information database 12e stores information regarding past dispatch performance in response to dispatch requests (hereinafter referred to as "dispatch performance information"). For example, the dispatch performance information includes various specifications for each vehicle V, such as the operating days and times (start and end times of operation, etc.), the route traveled, the arrival and departure times at each point (e.g., bus stops or intermediate points), the distance traveled, and fuel consumption (electricity consumption). The dispatch performance information also includes actual values ​​of vehicle demand based on dispatch requests during a predetermined time period in the past. The actual values ​​of vehicle demand in the dispatch performance information are used to calculate predicted values ​​of vehicle demand during a predetermined time period in the future.

[0028] Furthermore, the dispatch performance information database 12e can store various times related to dispatching based on dispatch requests. Dispatch times include, for example, actual waiting time, scheduled waiting time, allowable waiting time, and allowable delay time. Waiting time is the time from when the user's dispatch request is received until the user boards vehicle V at the pick-up location. Scheduled waiting time is the time calculated in creating the dispatch plan in response to the dispatch request. Allowable waiting time is the time the user may be made to wait from when the user's dispatch request is received until vehicle V is dispatched to the pick-up location, and can be arbitrarily set by the person providing and operating the dispatch service, for example, 5 minutes or 10 minutes. Allowable delay time is the delay time that is acceptable to the user from the scheduled arrival time of vehicle V at the destination. Dispatch times are used to calculate fairness indicators.

[0029] Furthermore, the dispatch performance information database 12e may store information regarding dispatches that were not fulfilled in response to a dispatch request. Specifically, if a user rejects a dispatch plan created in response to a dispatch request, the dispatch request (and / or dispatch plan) is canceled, and the dispatch performance information database 12e may store the dispatch request as an unfulfilled dispatch. Also, if no vehicle V is found whose scheduled waiting time meets the allowable waiting time for a dispatch request, the dispatch request (and / or dispatch plan) is canceled, and the dispatch performance information database 12e may store the dispatch request as an unfulfilled dispatch.

[0030] The front-end processing unit 110 performs various processes by communicating information with the user's information and communication terminal device 20. For example, the front-end processing unit 110 performs login authentication processing in response to login requests with the ride-hailing application running on the user's information and communication terminal device 20. As another example, under the control of the ride-hailing planning unit 130, the front-end processing unit 110 proposes a vehicle dispatch plan to the passenger and accepts or rejects it in response to the ride-hailing application.

[0031] The vehicle communication unit 120 exchanges various types of information with the vehicle V via the communication network N. For example, the vehicle communication unit 120 receives driving management information from the vehicle V. This driving management information includes, for example, location information from the vehicle V's GPS receiver, environmental information from imaging cameras and spatial measurement sensors, and road traffic information from the road traffic information management system 30. The vehicle communication unit 120 also transmits dispatch / deployment instructions from the dispatch planning unit 130 to the vehicle V.

[0032] The dispatch planning unit 130 determines the travel route from the departure point to the destination based on the user's dispatch request, selects the most suitable vehicle V from among the vehicles V in the target area, and creates a dispatch plan. Dispatch plans may be created for several different travel routes and candidate vehicles V. For example, based on the user's pick-up and drop-off locations indicated in the dispatch request, the dispatch planning unit 130 determines the travel route by referring to the road map information database 12b, selects an appropriate vehicle V by referring to the vehicle information database 12c, and creates a provisional dispatch plan for the selected vehicle V. For example, the dispatch planning unit 130 selects a vehicle V that is waiting or traveling near the pick-up location. Subsequently, the dispatch planning unit 130, in cooperation with the front-end processing unit 110, proposes the provisionally created dispatch plan to the user, and confirms the dispatch plan after receiving the user's approval. Then, the dispatch planning unit 130 stores the confirmed dispatch plan in the dispatch plan database 12d and issues a dispatch instruction to the corresponding vehicle V via the vehicle communication unit 120.

[0033] Furthermore, the dispatch planning unit 130 instructs vehicles V in the target area to wait at designated sub-area waiting locations by the start time of a predetermined time period in preparation for user dispatch requests. Specifically, for each of the multiple sub-areas in the target area, the dispatch planning unit 130 refers to the dispatch performance information database 12e and calculates a predicted value of vehicle demand based on future dispatch requests in a predetermined time period, based on the actual value of vehicle demand based on past dispatch requests. The dispatch planning unit 130 also calculates a fairness index regarding fairness among users enjoying the dispatch service for each sub-area, based on the actual value of dispatch time based on past dispatch requests. The dispatch management device 10 then determines which vehicles should be deployed to each sub-area at the start time of a predetermined time period in the future, based on the calculated predicted value of vehicle demand. Furthermore, the dispatch management device 10 determines whether or not to apply the calculated fairness index based on the fluctuations in fairness. If it determines that it should not be applied, it sets the fairness index to its minimum value. If it determines that it should be applied, it determines which vehicles should be deployed to a specific sub-area among several sub-areas based on the fairness index. The fluctuations in fairness are based, for example, on information about external factors during a predetermined time period in the past, stored in an external factors database (not shown). Information about external factors includes, for example, weather conditions and public transport operation information (e.g., service suspension information). The dispatch management device 10 then issues instructions to the vehicles for deployment to each sub-area by the start time of a predetermined time in the future.

[0034] Figure 3 is a block diagram showing an example of the specific configuration of the dispatch planning unit in a dispatch management device according to one embodiment of the present invention. The figure illustrates components that are particularly relevant to the technology of this disclosure. Specifically, in the dispatch planning unit 130 of this example, a vehicle demand forecasting unit 131, a fairness index calculation unit 132, a vehicle deployment determination unit 133, and a vehicle instruction unit 134 are shown, while components related to the creation of a dispatch plan based on a user's dispatch request (for example, a dispatch request reception unit, a vehicle information acquisition unit, a driving route determination unit, a vehicle selection unit, and a dispatch plan creation unit, etc.) are omitted.

[0035] The vehicle demand forecasting unit 131 calculates a predicted value of vehicle demand based on future dispatch requests for a predetermined time period, for each of the multiple sub-areas in the target area, based on the actual value of vehicle demand based on past dispatch requests. For example, as shown in Figure 4, the vehicle demand forecasting unit 131 refers to the dispatch performance information database 12e and calculates a predicted value of vehicle demand for a corresponding future time period for each sub-area, based on the actual value of vehicle demand based on past dispatch requests for a predetermined time period in the past. The predetermined past period is a fixed period such as one month, two months, or three months, and the predetermined time period is, for example, the same time period on the same day of the week (for example, 9:00 to 9:30 every Monday, but it is not limited to the same day of the week, and any time period that is in line with the user's behavior pattern, such as holidays and weekdays).

[0036] The fairness index calculation unit 132 calculates a fairness index for each sub-area, based on actual values ​​of the time spent on dispatching based on past dispatch requests, as shown in Figure 5, for example, regarding the fairness among users who enjoy the dispatch service. In the figure, the rectangular areas represent sub-areas, and the numerical values ​​within the rectangular areas represent the fairness index. The dispatch time is, for example, the expected waiting time for the user from the time the dispatch management device 10 receives the user's dispatch request until the vehicle V is dispatched to the pick-up location.

[0037] The fairness index is an index based on the role of the ride-hailing service as a public transportation service, and for example, it is set to a relatively large value when a high level of fairness is required. Figure 6 is a diagram illustrating an example of the fairness index in a ride-hailing management device according to one embodiment of the present invention. In the example shown in the figure, for each sub-area, the higher the value of the fairness index, the greater the difference in the fairness of users' opportunities to ride in the ride-hailing service. For example, in a sub-area with low population density and insufficient public transportation such as railways and buses, the number of vehicles V deployed in advance for the ride-hailing service will inevitably be small, from the perspective of vehicle demand based on ride-hailing requests alone. However, in light of the role of the ride-hailing service as a public transportation service, it is desirable that a certain number of vehicles V be deployed to ensure users have opportunities to ride, even while accepting a decrease in dispatch efficiency. Therefore, in this disclosure, a fairness index is introduced in such sub-areas as an evaluation index to ensure users have opportunities to ride, and the fairness index is used to determine the number of vehicles V to be deployed in each sub-area from the perspective of fairness, separate from the actual vehicle demand, in preparation for users' ride-hailing requests.

[0038] In this disclosure, the fairness index calculation unit 132 calculates the degree of variability (e.g., variability or dispersion) of scheduled waiting times for past dispatch requests, and calculates a fairness index using the scheduled waiting time that takes into account the magnitude of the degree of variability. The degree of variability of scheduled waiting times for dispatch is shown in relation to the number of dispatches (dispatch requests) per day (including 0) where the scheduled waiting time exceeds the allowable waiting time, and the number of days on which that number occurred. For example, it can be expressed as a distribution function with the number of occurrences on the horizontal axis and the corresponding number of days on the vertical axis. For example, for each sub-area, the fairness index calculation unit 132 counts the number of dispatch requests corresponding to the scheduled waiting time, based on whether the scheduled waiting time for a dispatch request exceeds the allowable waiting time for each predetermined time period in a predetermined past period, and calculates the degree of variability regarding the number of dispatch requests per day that exceed the allowable waiting time. This makes it possible to determine whether or not to apply the fairness index to each sub-area according to the degree of variability of the allowable waiting time for dispatch.

[0039] Furthermore, the fairness index calculation unit 132 determines whether the calculated variability is below a predetermined variability threshold. If the fairness index calculation unit 132 determines that the variability is below a predetermined variability threshold, it calculates a fairness index for each sub-area based on the actual values ​​of the scheduled waiting time for each predetermined time period in the past. More specifically, the fairness index calculation unit 132 may calculate the difference between the average scheduled waiting time and the allowable waiting time for each sub-area in a predetermined time period in the past, and determine this as the fairness index. Alternatively, the fairness index calculation unit 132 may calculate the fairness index by assigning a predetermined coefficient to the difference between the calculated average value and the allowable waiting time. This makes it possible to replenish vehicles V from a fairness perspective in sub-areas within the target area where there are many dispatches where the user's scheduled waiting time exceeds a predetermined allowable waiting time. If the dispatch management device 10 determines that there are no vehicles V available that exceed the allowable waiting time for the user's dispatch request, it notifies the user that there are no vehicles V available for dispatch (that the dispatch service is unavailable).

[0040] Furthermore, the fairness index calculation unit 132 extracts dispatch requests that exceed the allowable waiting time when the degree of variation exceeds a predetermined variation threshold, and calculates a fairness index based on the scheduled waiting time for the extracted dispatch requests. This makes it possible to supplement vehicles in sub-areas where the degree of variation in scheduled waiting time for dispatch based on dispatch requests is large, from the standpoint of fairness.

[0041] Furthermore, the fairness index calculation unit 132 identifies combinations of sub-areas and time periods in which the degree of variability exceeds a predetermined variability threshold. In addition, the fairness index calculation unit 132 creates a predetermined model that represents the circumstances leading to the dispatch of vehicles exceeding the allowable waiting time, based on predetermined external factors. The predetermined model may be, but is not limited to, an AI model that has been machine-learned using a predetermined machine learning algorithm, with predetermined external factors and past dispatch requests as explanatory variables (inputs) and a label indicating whether or not the scheduled waiting time has exceeded a predetermined threshold (e.g., allowable waiting time) as the dependent variable (output). Furthermore, the fairness index calculation unit 132 determines whether or not to apply the fairness index to the combination of sub-areas and time periods, according to the output of the predetermined model that has been created. This makes it possible to appropriately apply the fairness index to sub-areas with a large degree of variability in the allowable waiting time for vehicle dispatch, in accordance with the circumstances of fairness fluctuations.

[0042] Furthermore, the fairness index calculation unit 132 sets the fairness index to the minimum value for sub-areas where it has determined, based on a predetermined model, that the fairness index should not be applied. In other words, if, for a certain sub-area during a time period (which may include not only the date but also the day of the week), the significance of public service is considered to decrease due to fluctuations in fairness, the fairness index is set to the minimum value, and the vehicle V is not compensated for, allowing that vehicle V to be allocated to other sub-areas. As a result, in that sub-area, vehicle V is not compensated for, taking into account fluctuations in fairness, thereby improving dispatch efficiency.

[0043] Returning to Figure 3, the Vehicle Deployment Determination Unit 133 determines which vehicles to deploy for a predetermined time period in each sub-area based on the predicted vehicle demand calculated by the Vehicle Demand Forecasting Unit 131. That is, if the predicted vehicle demand is high for a predetermined time period in a certain sub-area, the Vehicle Deployment Determination Unit 133 determines which vehicles V to deploy for that predetermined time period in that sub-area so as to supply a number of vehicles V that match the predicted vehicle demand. The vehicles V whose deployment has been decided are deployed to the sub-area by the start time of the predetermined time period, under the instructions of the Vehicle Instruction Unit 134.

[0044] Furthermore, the vehicle deployment determination unit 133 can set as a specific sub-area any sub-area in which the fairness index exceeds a predetermined fairness threshold among multiple sub-areas. In other words, a specific sub-area is an area where vehicle V should be deployed from the perspective of fairness, regardless of vehicle demand. Therefore, a sub-area in which the fairness index is set to the minimum value is not set as a specific sub-area. This makes it possible to extract and identify sub-areas where user convenience is low, such as when vehicle V is not immediately dispatched in response to a dispatch request, so that more users within the target area can enjoy a certain level of dispatch service from the perspective of fairness.

[0045] Furthermore, the vehicle deployment determination unit 133 calculates the number of replacement vehicles V according to the fairness index for each specific sub-area determined according to a predetermined fairness threshold. For example, the vehicle deployment determination unit 133 may refer to a lookup table (not shown) that shows the relationship between the fairness index and the number of replacement vehicles, and determine the number of replacement vehicles for a specific sub-area according to the fairness index. For example, even if vehicle demand is low and no vehicles V are deployed to a specific sub-area during a predetermined time period, a number of vehicles V (replacement vehicles) according to the fairness index may be deployed. This ensures that, from the standpoint of fairness, the number of vehicles V is secured in a specific sub-area even if the predicted value of vehicle demand for vehicles V is low.

[0046] Furthermore, the vehicle deployment determination unit 133 may set the number of vehicles to be supplemented for a specific sub-area to a minimum value if, in the target area, the predicted value of vehicle demand exceeds a predetermined demand threshold and there are no sub-areas where the fairness index falls below a predetermined fairness threshold. In other words, if high vehicle demand is predicted in a certain sub-area during a predetermined time period in the future, and at the same time, the fairness index for that sub-area exceeds a predetermined fairness threshold, it is expected that there will be a shortage of vehicles V that can be supplied to that sub-area. If all sub-areas with high vehicle demand are in this state, it can be determined that there is a shortage of vehicles in the entire target area. Therefore, in such cases, the vehicle deployment determination unit 133 sets the number of vehicles to be supplemented to, for example, 0, based on the balance between vehicle demand and fairness. This makes it possible to secure as many vehicles V as possible that meet the vehicle demand in the target area by not supplementing vehicles V to a specific sub-area when it is expected that there will be a shortage of vehicles V in the target area due to strong vehicle demand during a predetermined time period.

[0047] Furthermore, the vehicle deployment determination unit 133 calculates the required number of vehicles V to be deployed for each specific sub-area based on the number of vehicles in demand and the number of vehicles to be replenished, based on the predicted vehicle demand. For example, in a certain sub-area, the fairness index value is relatively low, so it is not set as a specific sub-area, and the number of vehicles in demand based on the predicted vehicle demand is calculated as the required number of vehicles V to be deployed. On the other hand, in another sub-area, the fairness index value is relatively high, so it is set as a specific sub-area, and the required number is calculated by adding the number of vehicles in demand based on the predicted vehicle demand to the number of vehicles to be replenished according to the fairness index. This makes it possible to calculate the required number of vehicles V in a specific sub-area, taking into account the perspective of fairness, at the start time of a predetermined time period in the future. In addition, the vehicle deployment determination unit 133 obtains the number of vehicles V available for standby at the start time of a predetermined time period in the future for each specific sub-area. For example, the vehicle deployment determination unit 133 obtains the number of vehicles V available for standby by referring to the vehicle dispatch plan database 12d. Next, the vehicle deployment determination unit 133 calculates the number of vehicles V that need to be deployed based on the acquired number of vehicles on standby and the required number of vehicles V to be deployed. This makes it possible to calculate the number of vehicles V that should be deployed from other sub-areas to a specific sub-area at the start time of a predetermined time period in the future.

[0048] Furthermore, the vehicle deployment determination unit 133 calculates the ratio of the number of vehicles V that need to be deployed to the number of vehicles V that can handle the dispatch service at a predetermined start time in the future for each specific sub-area. This allows for proportional allocation according to the ratio of the number of vehicles V that need to be deployed to handle the dispatch service. In addition, the vehicle deployment determination unit 133 deploys vehicles V that can handle the dispatch service to each specific sub-area according to the calculated ratio. For example, the vehicle deployment determination unit 133 determines the vehicles V to be allocated to each sub-area in such a way that the travel distance between sub-areas is minimized during deployment. This ensures fairness by allowing vehicles V that can handle the dispatch service at a predetermined start time to be allocated in order from the sub-areas with the largest shortages.

[0049] Furthermore, the vehicle deployment determination unit 133 may calculate the number of replacement vehicles V according to the fairness index for each specific block area composed of multiple sub-areas. A specific block area may always consist of the same multiple sub-areas, or it may consist of multiple sub-areas dynamically selected according to the fairness index. For example, it is conceivable that several adjacent sub-areas each have relatively low fairness index values, and therefore, from a fairness standpoint, vehicles V will not be deployed as replacements. Even in such a case, if the fairness index is relatively high when viewed from the perspective of these sub-areas as a whole, then several vehicles V may be deployed to such adjacent sub-areas. For example, as shown in Figure 7(a), if the fairness index values ​​of four sub-areas are relatively low, the vehicle deployment determination unit 133 will calculate the number of replacement vehicles V to be deployed to one specific block area consisting of the four areas, as shown in Figure 7(b), if it determines that the sum of the fairness index values ​​of the four sub-areas is relatively high. This allows for the securing of sufficient replacement vehicles for an entire adjacent sub-area (a specific block area), even if a single sub-area falls short of the number of replacement vehicles required for one vehicle (V) according to fairness indicators, thereby maintaining a consistent level of ride-hailing service.

[0050] Furthermore, the vehicle deployment determination unit 133 determines which vehicle to deploy to which subblock constituting a specific block area and its deployment location based on the fairness index. For example, if one vehicle V is deployed to four sub-areas as one specific block area, the vehicle V is deployed to the sub-area with the highest fairness index among the four sub-areas (see Figure 7). As another example, as shown in Figure 8(a), if two vehicles V are deployed to four sub-areas as one specific block area, the sub-areas and their waiting locations are determined based on the weighting of the fairness index. In Figure 8(b), one vehicle is selected for the sub-area with the highest fairness index, and the waiting location is in a sub-area adjacent to that sub-area and closest to that sub-area. This enables efficient vehicle deployment by considering the balance of fairness indices for each sub-block in a specific block area and deploying the vehicle V to an appropriate sub-block and in an appropriate location within that sub-block.

[0051] The vehicle instruction unit 134 issues instructions to the vehicles V for deployment to each sub-area. That is, once the vehicles V to be deployed to each sub-area are determined, the vehicle instruction unit 134 instructs each of the vehicles V to move to the designated waiting area in the sub-area. The vehicle instruction unit 134 may also instruct the vehicles to travel from the waiting area to the boarding area and then to the disembarking area according to the dispatch plan.

[0052] Figures 9A and 9B are flowcharts showing an example of vehicle deployment processing in a dispatch management device according to one embodiment of the present invention. This processing is realized by the dispatch management device 10 executing a dispatch management server program under the control of a processor, thereby cooperating with various hardware resources.

[0053] As shown in Figure A, the dispatch management device 10 monitors whether it is time to update the vehicle deployment for a predetermined time period (S901). The timing for updating the vehicle deployment may be, for example, three hours before the start time of a predetermined time period in the future, or before the start of the day's dispatch service. If the dispatch management device 10 determines that it is time to update the vehicle deployment (Yes in S901), it refers to the dispatch performance information database 12e and calculates a predicted value of vehicle demand based on dispatch requests for predetermined time periods in each sub-area (S902).

[0054] Next, the dispatch management device 10 refers to the dispatch performance information database 12e and calculates a fairness index for each sub-area based on the actual time values ​​related to dispatches based on past dispatch requests, relating to the fairness among users who enjoy the dispatch service (S903). The calculation process of the fairness index will be explained later using Figure 10.

[0055] Next, the dispatch management device 10 determines whether there is a sub-area in the target area where the predicted value of vehicle demand exceeds a predetermined demand threshold and the fairness index is below a predetermined fairness threshold (S904). If the dispatch management device 10 determines that there is a sub-area where the predicted value of vehicle demand exceeds a predetermined demand threshold and the value of the fairness index is below a predetermined fairness threshold (Yes in S904), the dispatch management device 10 then determines whether there is a sub-area where the fairness index exceeds a predetermined fairness threshold (S905). If the dispatch management device 10 determines that there is a sub-area where the fairness index exceeds the predetermined fairness threshold (Yes in S905), the dispatch management device 10 calculates the number of replacement vehicles V based on the fairness index for the sub-area where the fairness index exceeds the predetermined fairness threshold (specific sub-area) (S906).

[0056] On the other hand, in processing step S904, if the dispatch management device 10 determines that there are no sub-areas where the predicted value of vehicle demand exceeds a predetermined demand threshold and the fairness index falls below a predetermined fairness threshold (No in S904), the dispatch management device 10 sets the number of vehicles V to be compensated based on the fairness index to the minimum value (e.g., 0 vehicles) for all sub-areas in the target area (S907). In other words, if there are no sub-areas where the predicted value of vehicle demand exceeds a predetermined demand threshold and the fairness index falls below a predetermined fairness threshold, it is assumed that there is a tendency for the number of vehicles V to be used for the dispatch service in the target area to be insufficient, and the compensation of vehicles V based on the fairness index is suppressed for all sub-areas.

[0057] Furthermore, if the dispatch management device 10 determines in processing step S905 that there are no sub-areas where the fairness index exceeds a predetermined fairness threshold (No. in S905), the dispatch management device 10 sets the number of vehicles V to be compensated based on the fairness index to the minimum value (for example, 0 vehicles) for all sub-areas in the target area (S907). In other words, in this case, there are no sub-areas in the target area that require compensation based on fairness, so no compensation of vehicles V is performed.

[0058] Next, the dispatch management device 10 performs the processing steps S908 to S910 for each sub-area, as shown in Figure 9B. That is, for each sub-area, the dispatch management device 10 first calculates the required number of vehicles V to be deployed based on the number of vehicles V in demand and the number of vehicles to be replenished (S908). Note that in a sub-area where the number of vehicles to be replenished is, for example, 0, the number of vehicles in demand becomes the required number. Subsequently, for each sub-area, the dispatch management device 10 obtains the number of vehicles V that are available to wait at the start time of a predetermined time period in the future (S909), and further, for each sub-area, the dispatch management device 10 calculates the number of vehicles V that are needed to be deployed based on the obtained number of vehicles in wait and the required number of vehicles V to be deployed (S910).

[0059] After calculating the number of vehicles V needed to be deployed to each sub-area, the dispatch management device 10 then calculates the ratio of the number of vehicles V needed to be deployed to the number of vehicles V available to handle the dispatch service at a predetermined start time in the future (S911). This allows for proportional allocation according to the ratio of the number of vehicles V needed to handle the dispatch service.

[0060] As described above, once the proportion of vehicles needed in each sub-area is calculated, the dispatch management device 10 allocates vehicles V to each sub-area according to the calculated proportion of vehicles needed (S912). When deploying, the dispatch management device 10 determines which vehicles V to allocate to each sub-area, for example, so that the travel distance between sub-areas is minimized. Subsequently, the dispatch management device 10 issues deployment instructions to the vehicles V allocated to each sub-area so that they should wait at the designated waiting area by the start time of a predetermined time slot (S913).

[0061] Figure 10 is a flowchart showing an example of the fairness index calculation process in a dispatch management device according to one embodiment of the present invention. Specifically, the figure is a flowchart detailing the processing step S903 shown in Figure 9.

[0062] As shown in the figure, the dispatch management device 10 refers to the dispatch performance information database 12e and extracts dispatch groups in which the scheduled waiting time for a dispatch request exceeds the allowable waiting time, based on past dispatch performance information for a predetermined time period in each sub-area (S1001). Next, the dispatch management device 10 calculates the daily variation in the number of occurrences for the extracted dispatch groups (S1002). The variation in the number of occurrences for a dispatch group is shown as the relationship between the daily number of occurrences (including 0) for the dispatch group in a predetermined time period and the number of days on which that number occurred.

[0063] Next, the dispatch management device 10 determines whether the calculated variability is below a predetermined variability threshold (S1003). If the dispatch management device 10 determines that the calculated variability is below a predetermined variability threshold (Yes in S1003), the dispatch management device 10 calculates the difference between the average scheduled waiting time and the allowable waiting time for the extracted dispatch group (S1004). Subsequently, the dispatch management device 10 calculates a fairness index using a predetermined coefficient based on the calculated difference between the average scheduled waiting time and the allowable waiting time (S1005). For example, the dispatch management device 10 can calculate the fairness index by normalizing the calculated difference between the average scheduled waiting time and the allowable waiting time by multiplying it by a predetermined coefficient. Alternatively, the predetermined coefficient may be a weighting coefficient.

[0064] On the other hand, if the dispatch management device 10 determines that the calculated variability is not below a predetermined variability threshold (No. in S1003), the dispatch management device 10 creates a predetermined model for the dispatch group based on predetermined external factors to represent the circumstances under which the scheduled waiting time exceeded a predetermined threshold (S1006). The predetermined model outputs a label indicating whether or not the scheduled waiting time exceeded a predetermined threshold for inputs such as predetermined external factors or past dispatch requests. Note that the predetermined model does not have to be created sequentially; a predetermined model created in the past may be used, in which case processing step S1006 is omitted.

[0065] Next, the dispatch management device 10 uses the created predetermined model to determine whether or not to apply the fairness index to sub-areas that exceed a predetermined variability threshold during a predetermined time period (S1007).

[0066] If the dispatch management device 10 determines that it will apply the fairness index to a sub-area that exceeds a predetermined variability threshold during a predetermined time period (Yes in S1007), the dispatch management device 10 calculates the average value of the difference between the scheduled waiting time and the predetermined threshold for the dispatch group (S1008). Subsequently, the dispatch management device 10 calculates the fairness index based on the calculated average value (S1009). In other words, if it is determined that a combination of such sub-area and predetermined time period is likely to result in extremely long scheduled waiting times and thus a high probability of no dispatch, the fairness index will be applied.

[0067] On the other hand, if the dispatch management device 10 determines that it will not apply the fairness index to a sub-area that exceeds a predetermined variability threshold during a predetermined time period (No. in S1007), the dispatch management device 10 sets the fairness index for the sub-area that exceeds the predetermined variability threshold during a predetermined time period (combination of sub-area and predetermined time period) to the minimum value (S1010). In other words, if it is determined that the situation is such that it is unlikely that the scheduled waiting time will be long and no vehicle will be dispatched, the fairness index will not be applied to the sub-area during that predetermined time period by setting the fairness threshold to the minimum value.

[0068] In this way, the dispatch management device 10 can refer to the dispatch performance information database 12e and calculate different fairness indicators according to the daily variation in the number of cases where the scheduled waiting time exceeds a threshold.

[0069] As described above, according to this embodiment, in a ride-hailing service in which vehicles V are pre-deployed to each sub-area within the service area according to the predicted vehicle demand, the dispatch management device 10 can deploy vehicles V even in sub-areas where vehicle demand is low and vehicles V are not deployed, from the standpoint of public interest and fairness, and also ensure user convenience. In particular, according to this embodiment, if the dispatch management device 10 can provide sufficient ride-hailing service in a certain sub-area at a certain time of day without considering fairness, it can deploy vehicles V according to the predicted vehicle demand, thereby improving dispatch efficiency.

[0070] The embodiments described above are illustrative examples for illustrating the present invention and are not intended to limit the invention to these embodiments only. The present invention can be implemented in various forms without departing from its spirit.

[0071] For example, in the methods disclosed herein, steps, operations, or functions may be performed in parallel or in different orders, as long as this does not result in a contradiction in the outcome. The steps, operations, and functions described are provided merely as examples, and some of the steps, operations, and functions may be omitted, combined with each other to form a single unit, or other steps, operations, or functions may be added, without departing from the spirit of the invention.

[0072] Furthermore, although various embodiments are disclosed herein, specific features (technical matters) in one embodiment can be added to or replaced in other embodiments, with appropriate modifications, and such forms are also included in the gist of the present invention. [Explanation of Symbols]

[0073] 1… Dispatch management system 10… Dispatch management system 110…Front-end processing unit 120... Vehicle Communications Department 130... Dispatch Planning Department 131... Vehicle Demand Forecasting Department 132…Fairness index calculation unit 133... Vehicle Deployment Decision Department 134... Vehicle Indicator 12…Database 12a...User Information Database 12b...Road Map Information Database 12c... Vehicle Information Database 12d... Vehicle dispatch planning database 12e... Dispatch performance information database 20… Information and communication terminal equipment 30…Road Traffic Information Management System N...communication network V...vehicle CTL...Control Unit

Claims

1. A dispatch management device that manages vehicle dispatch services in a target area in response to a user's dispatch request, A vehicle demand forecasting unit calculates a predicted value for vehicle demand based on future dispatch requests during a predetermined time period, based on actual values ​​of vehicle demand based on past dispatch requests, for each of the multiple sub-areas within the target area. A fairness index calculation unit calculates a fairness index relating to fairness among users enjoying the ride-hailing service based on actual time values ​​for ride-hailing based on past ride-hailing requests for the predetermined time period in each of the sub-areas, A vehicle deployment determination unit determines which vehicles to be deployed for each sub-area during the predetermined time period based on the calculated forecast values ​​of vehicle demand, The vehicle includes a vehicle instruction unit that gives instructions to the vehicle for deployment to the sub-area, The fairness index calculation unit determines whether or not to apply the fairness index to each sub-area based on the circumstances of fairness fluctuations caused by predetermined external factors. The vehicle deployment determination unit determines which vehicles to deploy to a specific sub-area among the plurality of sub-areas to which the fairness index is determined to apply, based on the calculated fairness index. Vehicle dispatch management system.

2. The fairness index calculation unit calculates the degree of variation in the scheduled waiting time for the dispatch request for each sub-area, based on whether the scheduled waiting time for the dispatch request exceeds the allowable waiting time for each predetermined time period within a predetermined period in the past. The vehicle dispatch management device according to claim 1.

3. The fairness index calculation unit calculates the fairness index for each sub-area, based on the scheduled waiting time for the dispatch request, for each predetermined time period within the predetermined past period, when the degree of variation is less than or equal to a predetermined variation threshold. The vehicle dispatch management device according to claim 2.

4. The fairness index calculation unit is, If the degree of variation exceeds a predetermined variation threshold, the dispatch requests that exceed the allowable waiting time are extracted, and the fairness index is calculated based on the scheduled waiting time for the extracted dispatch requests. The vehicle dispatch management device according to claim 2.

5. The fairness index calculation unit is, Identify the combination of the sub-area and the predetermined time period in which the degree of variation exceeds the predetermined variation threshold. Based on the aforementioned predetermined external factors, a predetermined model is created to represent the circumstances that led to the aforementioned vehicle dispatch. With respect to the combination of the sub-area and the predetermined time period, it is determined whether or not the fairness indicator should be applied based on the predetermined model. The vehicle dispatch management device according to claim 4.

6. The fairness index calculation unit sets the fairness index to the minimum value for combinations of sub-areas and predetermined time periods to which the fairness index should not be applied. The vehicle dispatch management device according to claim 5.

7. The vehicle deployment determination unit determines, among the plurality of sub-areas, a sub-area in which the fairness index exceeds a predetermined fairness threshold as the specific sub-area. The vehicle dispatch management device according to claim 5.

8. The vehicle deployment determination unit determines the number of replacement vehicles to be deployed to a specific sub-area to be the minimum value if, in the target area, the predicted value of the vehicle demand exceeds a predetermined demand threshold and there are no sub-areas where the fairness index is below a predetermined fairness threshold. The vehicle dispatch management device according to claim 6.

9. The vehicle deployment determination unit calculates the number of replacement vehicles for each specific sub-area according to the fairness index. The vehicle dispatch management device according to claim 6.

10. The vehicle deployment determination unit calculates the required number of vehicles to be deployed for each specific sub-area based on the number of vehicles in demand based on the forecast value of vehicle demand and the number of vehicles to be replenished. The dispatch management device according to claim 9.

11. The vehicle deployment determination unit calculates the number of vehicles that need to be deployed for each specific sub-area, based on the number of vehicles available to be on standby at the start time of the predetermined time period and the required number of vehicles to be deployed. The vehicle dispatch management device according to claim 10.

12. For each of the specified sub-areas, calculate the ratio of the number of vehicles to be deployed to the number of vehicles available to handle the dispatch service at the predetermined start time. In accordance with the calculated proportion, vehicles capable of handling the dispatch service are allocated to each of the specified sub-areas. The vehicle dispatch management device according to claim 11.

13. A dispatch management method using a dispatch management device that manages vehicle dispatch services in a target area in response to a user's dispatch request, For each of the multiple sub-areas within the aforementioned target area, a forecast value of vehicle demand based on future dispatch requests during a predetermined time period is calculated based on the actual vehicle demand values ​​based on past dispatch requests. For each of the aforementioned sub-areas, within the predetermined time period, a fairness index relating to fairness among users enjoying the ride-hailing service is calculated based on the actual time values ​​for ride-hailing based on past ride-hailing requests. Based on the calculated forecast values ​​for vehicle demand, the vehicles to be deployed for the predetermined time period in each of the sub-areas are determined. This includes giving instructions to the vehicle for deployment to the aforementioned sub-area, Calculating the fairness index includes determining whether or not to apply the fairness index to each sub-area based on the circumstances of fairness fluctuations caused by predetermined external factors, Determining the vehicles to be deployed includes determining the vehicles to be deployed based on the calculated fairness index for a specific sub-area among the plurality of sub-areas to which the fairness index is determined to apply. Vehicle dispatch management method.

Images