Traffic prediction system, traffic prediction method, and computer readable medium

Abstract

The present application provides a traffic prediction system, a traffic prediction method and a computer readable medium, which efficiently predict the change in required time for a route in a case where the traffic volume of the route changes along with a change in action. A traffic volume data acquisition unit acquires traffic volume data. A required time acquisition unit acquires an actual value of required time. A traffic volume calculation unit calculates the inflow traffic volume of each route using the traffic volume data and the actual value of required time. A traffic state prediction unit calculates, for each route, a predicted value of required time for the calculated inflow traffic volume and a predicted value of required time in a case where the inflow traffic volume changes, using a traffic model generated in advance. A change amount calculation unit calculates, for each route, a change amount of the predicted value of required time in a case where the inflow traffic volume changes.

Description

Technical Field

[0001] This disclosure relates to traffic prediction systems, traffic prediction methods, and computer-readable media, and more particularly to traffic prediction systems, traffic prediction methods, and computer-readable media for monitoring vehicle traffic. Background Technology

[0002] There are technologies for predicting traffic in transportation networks (roads) where vehicles pass through. For example, Japanese Patent No. 5523886 discloses a prediction calculation device that defines a road with multiple entrances and multiple exits as being formed by multiple road segments of a predetermined length, and predicts the movement of multiple vehicles moving from entrances to exits for each road segment. In the prediction calculation device disclosed in Japanese Patent No. 5523886, the entrance traffic volume prediction unit calculates the number of vehicles flowing in from each entrance as the number of inflow vehicles based on traffic volume information indicating the traffic volume of vehicles observed on the road. The emergency response unit estimates the possible exits that vehicles may choose based on obstruction information indicating events that may obstruct the movement of vehicles, and generates origin-end point information that is corrected by multiplying standard origin-end point information indicating the traffic volume of a standard movement state from entrance to exit by a conversion rate generated based on the estimated exit. In addition, the baseline condition calculation unit calculates the vehicle speed for each road segment based on the corrected origin-end point information. The forecasting unit calculates the travel time between destinations until the vehicles reach their destinations before a predetermined time, based on the calculated number of inbound vehicles and the calculated speed of movement, and forecasts traffic conditions. Summary of the Invention

[0003] Traffic volume on a road (path) can change due to changes in the mode of transportation a vehicle uses from its entrance to its exit. In such cases, it is desirable to efficiently predict changes in the travel time required to traverse the road (path). Japanese Patent No. 5523886 utilizes data (infrastructure data) collected by equipment (traffic counters, etc.) managed by road administrators. However, obtaining such data (infrastructure data) at desired locations is not easy. Therefore, the technology disclosed in Japanese Patent No. 5523886 may not be able to efficiently predict changes in travel time when traffic volume on a path changes due to changes in movement.

[0004] This disclosure provides a traffic prediction system, method, and procedure that can efficiently predict changes in the time required to travel a route in the event of changes in traffic volume accompanying changes in movement.

[0005] The traffic prediction system disclosed herein includes: a traffic volume data acquisition unit that acquires traffic volume data representing the relationship between time and traffic volume at a specified location; a required time acquisition unit that, for each of at least one path leading to the specified location, acquires an actual value of the required time for a vehicle to travel from a reference point along the path to the specified location; a traffic volume calculation unit that uses the traffic volume data and the actual value of the required time to calculate the inflow traffic volume flowing from the reference point to the path; a traffic state prediction unit that, with respect to the path, uses a pre-generated traffic model to calculate a predicted value of the required time for the calculated inflow traffic volume and a predicted value of the required time under conditions that cause changes in the inflow traffic volume; and a change calculation unit that, with respect to the path, calculates the change in the predicted value of the required time under conditions that cause changes in the inflow traffic volume.

[0006] Furthermore, the traffic prediction method disclosed herein obtains traffic volume data representing the relationship between time and traffic volume at a specified location; for each path of at least one path to the specified location, it obtains the actual value of the time required for a vehicle to travel from a reference point on the path to the specified location; it uses the traffic volume data and the actual value of the required time to calculate the inflow traffic volume flowing from the reference point to the path; with respect to the path, it uses a pre-generated traffic model to calculate the predicted value of the required time for the calculated inflow traffic volume and the predicted value of the required time under the condition that the inflow traffic volume changes; with respect to the path, it calculates the amount of change in the predicted value of the required time under the condition that the inflow traffic volume changes.

[0007] Furthermore, the program involved in this disclosure causes a computer to perform the following steps: obtaining traffic volume data representing the relationship between time and traffic volume at a specified location; for each of at least one path leading to the specified location, obtaining an actual value of the time required for a vehicle to travel from a reference point on the path to the specified location; using the traffic volume data and the actual value of the required time to calculate the inflow traffic volume flowing from the reference point to the path; using a pre-generated traffic model for the path to calculate a predicted value of the required time for the calculated inflow traffic volume and a predicted value of the required time in the event of a change in the inflow traffic volume; and for the path to calculate the amount of change in the predicted value of the required time in the event of a change in the inflow traffic volume.

[0008] In this disclosure, by means of such a configuration, it is possible to predict changes in the time required to travel a route when there are changes in inflow traffic volume, even without obtaining infrastructure data. Therefore, this disclosure can efficiently predict changes in the time required to travel a route when there are changes in traffic volume due to changes in movement.

[0009] Alternatively, preferably, the required time acquisition unit uses actual driving data obtained from multiple specific vehicles traveling on the path to obtain the actual value of the required time.

[0010] Actual driving data is obtained from specific vehicles that the system administrator can manage. Therefore, the system administrator can easily obtain actual driving data. Consequently, this disclosure allows for the easy acquisition of the actual value of the required time.

[0011] Alternatively, preferably, the required time acquisition unit acquires the actual value of the required time for each of the multiple paths to the designated location, the traffic volume calculation unit uses the traffic volume data and the actual value of the required time to calculate the inflow traffic volume for each of the multiple paths, the traffic state prediction unit calculates the predicted value of the required time for the calculated inflow traffic volume and the predicted value of the required time under the condition of changing the inflow traffic volume for each of the multiple paths, and calculates the change in the predicted value of the required time under the condition of changing the inflow traffic volume for each of the multiple paths.

[0012] This structure allows us to predict which path will have the greatest variation in time required. Therefore, we can determine the path where the change in action will have a higher impact.

[0013] Alternatively, preferably, the required time acquisition unit uses actual driving data obtained from multiple specific vehicles traveling on multiple paths to obtain the actual value of the required time, and the traffic volume calculation unit uses the modal share of the traffic volume for each path in the traffic volume data to calculate the inflow traffic volume for each of the multiple paths, wherein the modal share of each path is calculated based on the number of specific vehicles traveling on the multiple paths.

[0014] With this structure, it is possible to calculate the inflow traffic volume for each path even without obtaining infrastructure data.

[0015] Alternatively, preferably, the traffic volume calculation unit uses the modal share to calculate the cumulative outflow traffic volume, which is the cumulative outflow traffic volume from the traffic volume data that flows out from the multiple paths respectively. The traffic volume calculation unit uses the cumulative outflow traffic volume and the actual value of the required time to calculate the cumulative inflow traffic volume, i.e., the cumulative inflow traffic volume. The traffic volume calculation unit calculates the inflow traffic volume at each time based on the cumulative inflow traffic volume.

[0016] With this structure, it is possible to calculate the inflow traffic volume for each path even without obtaining infrastructure data.

[0017] Alternatively, preferably, the traffic model is generated by adjusting the parameters of the traffic model in a manner that reproduces the actual value of the required time when the inflow traffic volume is input into the traffic model.

[0018] This structure allows for the calculation of highly accurate time forecasts based on inflow traffic volume.

[0019] In addition, preferably, the parameters of the traffic model include the bottleneck capacity of each path at each time step.

[0020] With this structure, it is possible to calculate a predicted value of the required time that accurately reproduces the actual value of the required time. Therefore, this disclosure can predict the required time and outflow traffic volume with high accuracy.

[0021] According to this disclosure, a traffic prediction system, traffic prediction method, and procedure are provided that can efficiently predict changes in the time required to travel a route in the event of changes in traffic volume accompanying changes in movement.

[0022] The above and other objects, features and advantages of this disclosure will be more fully understood from the detailed description and accompanying drawings given below, which are given by way of illustration only and should not be regarded as limiting the disclosure. Attached Figure Description

[0023] Figure 1 This is a diagram illustrating the traffic prediction system involved in Implementation 1.

[0024] Figure 2 This is a diagram showing the structure of the traffic prediction device according to Embodiment 1.

[0025] Figure 3 This is a flowchart illustrating a traffic prediction method performed by the traffic prediction system according to Embodiment 1.

[0026] Figure 4This is a flowchart illustrating a traffic prediction method performed by the traffic prediction system according to Embodiment 1.

[0027] Figure 5 This is a flowchart illustrating a traffic prediction method performed by the traffic prediction system according to Embodiment 1.

[0028] Figure 6 This is a diagram illustrating the path involved in Implementation Method 1.

[0029] Figure 7 This is a graph illustrating traffic volume data related to Implementation Method 1.

[0030] Figure 8 This is a diagram illustrating the method for obtaining the required time according to Embodiment 1.

[0031] Figure 9 This is a diagram used to illustrate the processing of the traffic volume calculation unit according to Embodiment 1.

[0032] Figure 10 This is a diagram illustrating bottlenecks in a path using the traffic model involved in Implementation 1.

[0033] Figure 11 This is a diagram illustrating the bottleneck capacity adjusted during the generation of the traffic model involved in Implementation 1.

[0034] Figure 12 This is a graph illustrating the predicted time calculated by the traffic model involved in Implementation 1.

[0035] Figure 13 This is a graph comparing the predicted time required before and during the implementation of the measures involved in Implementation Method 1.

[0036] Figure 14 This is a graph comparing the predicted time required before and during the implementation of the measures involved in Implementation Method 1.

[0037] Figure 15 This is a graph comparing the predicted time required before and during the implementation of the measures involved in Implementation Method 1.

[0038] Figure 16 This is a graph comparing the predicted time required before and during the implementation of the measures involved in Implementation Method 1.

[0039] Figure 17 This is a diagram showing the structure of the traffic prediction device according to Embodiment 2.

[0040] Figure 18 This is a flowchart illustrating a traffic prediction method performed by the traffic prediction system according to Embodiment 2.

[0041] Figure 19 This is a flowchart illustrating a traffic prediction method performed by the traffic prediction system according to Embodiment 2.

[0042] Figure 20 This is a graph illustrating the mixing rate and overall traffic volume involved in Implementation Method 2. Detailed Implementation

[0043] (Implementation Method 1)

[0044] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. It should be noted that substantially the same constituent elements are labeled with the same reference numerals. It should also be noted that in the following figures, the data in the graphs, etc., are illustrated for use in the description of the embodiments and may not correspond strictly to each other.

[0045] Figure 1 This is a diagram illustrating the traffic prediction system 1 according to Embodiment 1. The traffic prediction system 1 according to Embodiment 1 includes a traffic prediction device 100, a plurality of detection vehicles 20, and at least one traffic volume data detection device 50. The traffic prediction device 100 is communicatively connected to the detection vehicles 20 and the traffic volume data detection device 50 via a wired or wireless network 2.

[0046] The detection vehicle 20 is configured to transmit detection data to the traffic prediction device 100. The detection data is the actual operating data (driving data) of the detection vehicle 20. When the detection vehicle 20 is in motion, the information processing unit 22 mounted on the detection vehicle 20 acquires the detection data using sensors mounted on the detection vehicle 20. The information processing unit 22 then transmits the detection data to the traffic prediction device 100 via the network 2. It should be noted that the detection data can also be transmitted to a device different from the traffic prediction device 100. In this case, the traffic prediction device 100 receives (acquires) the detection data from that different device.

[0047] The detection data includes, for example, the identification information, time, location information, section information (road segment information), speed information, and congestion information of the corresponding detection vehicle 20. Section information (road segment information) is information related to the interval between a certain location. Congestion information can be detected, for example, by a car navigation system. Through the detection data, it is possible to determine the time when the detection vehicle 20 passed a certain location and its speed at that time. Alternatively, through the detection data, it is possible to determine the time (travel time) required for the detection vehicle 20 to traverse a certain section within a certain time period. Details regarding the detection data will be discussed later.

[0048] The traffic volume data detection device 50 is configured to detect traffic volume at a specific location (a designated location). Here, the traffic volume data detection device 50 is managed by the administrator (system administrator) of the traffic prediction system 1. Furthermore, the "designated location" is, for example, the entrance to a facility that can be managed by the system administrator. The "designated location" could be, for example, the entrance to a parking lot of a facility that can be managed by the system administrator (facility administrator). In this case, the traffic volume data detection device 50 is located near the entrance gate of the parking lot. The traffic volume data detection device 50 establishes a correspondence between the number of vehicles entering the parking lot and the time of entry and counts them. Thus, the traffic volume data detection device 50 detects the traffic volume at the entrance gate of the parking lot. Details regarding traffic volume data will be described later.

[0049] The traffic prediction device 100 is, for example, a computer such as a server. The traffic prediction device 100 can also be implemented using cloud computing. As described below, the traffic prediction device 100 uses traffic volume data and detection data to calculate the traffic volume for each of multiple paths ending at a predetermined location. Here, each of the multiple paths starts from a pre-determined reference location. Furthermore, the traffic prediction device 100 uses a pre-generated traffic model to predict the change in the required time (travel time) for a vehicle to travel from the reference location to the predetermined location along the path, causing a change in traffic volume for each of the multiple paths. It should be noted that, hereinafter, the required time (travel time) for a vehicle to travel from the reference location of the path to the predetermined location is sometimes simply referred to as the "required time of the path." Details regarding the traffic prediction device 100 will be described later. Additionally, the aforementioned "traffic model" is used to predict the required time for each path. Details will be described later.

[0050] Figure 2 This is a diagram showing the structure of the traffic prediction device 100 according to Embodiment 1. Additionally, Figures 3-5 This is a flowchart illustrating the traffic prediction method executed by the traffic prediction system 1 according to Embodiment 1. The traffic prediction device 100 according to Embodiment 1 includes a CPU (Central Processing Unit) 102, a ROM (Read Only Memory) 104, a RAM (Random Access Memory) 106, and an interface unit 108 (IF) as its main hardware structure. The CPU 102, ROM 104, RAM 106, and interface unit 108 are interconnected via a data bus, etc. It should be noted that the information processing device 22 may also have a structure substantially the same as the hardware structure of the traffic prediction device 100.

[0051] CPU 102 functions as a computing device for control and arithmetic processing. ROM 104 stores control programs and arithmetic programs executed by CPU 102. RAM 106 temporarily stores processed data. RAM 106 may also contain a database. Therefore, the traffic prediction device 100 can also implement a database. Interface 108 exchanges signals with external devices via wired or wireless means. Furthermore, interface 108 accepts data input from the user and displays information to the user.

[0052] Furthermore, as components of the representation function, the traffic prediction device 100 according to Embodiment 1 includes a route information storage unit 110, a traffic volume data acquisition unit 112, a detection data acquisition unit 114, a route share calculation unit 116, and a required time acquisition unit 118. Additionally, the traffic prediction device 100 according to Embodiment 1 includes a traffic volume calculation unit 120, a parameter setting unit 128, and a traffic model generation unit 130. Furthermore, the traffic prediction device 100 according to Embodiment 1 includes a change amount setting unit 140, a traffic state prediction unit 150, and a change amount calculation unit 160. Additionally, the traffic volume calculation unit 120 according to Embodiment 1 includes an outflow traffic volume calculation unit 122 and an inflow traffic volume calculation unit 124.

[0053] These components can be implemented, for example, by the CPU 102 executing a program stored in the ROM 104. Alternatively, the required program can be recorded on any non-volatile recording medium and installed as needed to implement each component. It should be noted that each component is not limited to being implemented in software as described above; it can also be implemented in hardware such as certain circuit elements. Furthermore, one or more of the above-mentioned components can be implemented separately by physically independent hardware. This is also true in other embodiments described later.

[0054] The route information storage unit 110 is configured to store route information, which is related to multiple routes ending at a specified location. The route information includes route identification information (route identification information) managed by the traffic prediction system 1, location information of the route's reference point (starting point), location information of intermediate points (sections) along the route, and location information of the specified location (ending point). Additionally, the route information may also include identification information of the detection vehicle 20 traveling along the route.

[0055] Figure 6 This is a diagram illustrating the path involved in Implementation Method 1. In Figure 6Two paths are shown: path R1 and path R2. As indicated by arrow A1, path R1 is the path from base point S1 to designated location P. As indicated by arrow A2, path R2 is the path from base point S2 to designated location P. For example, designated location P is a designated facility. More specifically, designated location P is the company's parking lot. And, for example, paths R1 and R2 are the routes taken by many employees' commuter vehicles (employee vehicles) during their commute. It should be noted that employee vehicles can also be referred to as vehicles entering the parking lot. In addition, employee vehicles can also be referred to as object vehicles that are monitored in this embodiment.

[0056] Here, a portion of the many employee vehicles traveling on paths R1 and R2 are detection vehicles 20 (specific vehicles). Furthermore, the number of detection vehicles 20 traveling on paths R1 and R2 is determined based on the proportion of the many employee vehicles traveling on each path. In other words, the number of detection vehicles 20 is set such that the more employee vehicles (target vehicles) traveling on a path, the more detection vehicles 20 (specific vehicles) travel on that path.

[0057] The following uses Figures 3-5 The flowchart shown illustrates the function of the aforementioned constituent elements. Figure 3 This is a flowchart illustrating the process of generating a traffic model according to Embodiment 1. The traffic volume data acquisition unit 112 acquires traffic volume data from the traffic volume data detection device 50 (step S102). That is, the traffic volume data acquisition unit 112 is configured to acquire traffic volume data representing the relationship between time and traffic volume at a specified location. The traffic volume data acquisition unit 112 can store the acquired traffic volume data in a database composed of RAM 106.

[0058] Here, traffic volume data represents the relationship between a given time and the number of vehicles passing a designated location P per unit of time at that time. In this embodiment, "time" does not need to refer to a strict instant; it can also be a predetermined time period (e.g., a time period every 5 minutes). Furthermore, in... Figure 6 In the example, "traffic volume" in traffic volume data is, for example, the number of parking lot occupancy per unit time (the number of facility occupancy).

[0059] Figure 7 This is a graph illustrating traffic volume data related to Implementation Method 1. Figure 7 The points represented by the quadrilaterals indicate the number of parking lot entries per 5 minutes over time, as a measure of traffic volume. Additionally, Figure 7 The circles representing points indicate the cumulative traffic volume over time. Cumulative traffic volume corresponds to the cumulative traffic volume over time. Additionally, in... Figure 7In the diagram, the horizontal axis represents time, the vertical axis on the left represents traffic volume (traffic volume per unit time), and the vertical axis on the right represents cumulative traffic volume.

[0060] Here, traffic volume data represents the traffic volume at the destination of each route, i.e., the designated location P. Therefore, the traffic volume represented by the traffic volume data corresponds to the total number of vehicles (employee vehicles) traveling on each route. Here, if we assume... Figure 6 and Figure 7 Correspondingly, Figure 7 The traffic volume data corresponds to the total number of vehicles that travel on path R1 and arrive at the designated location P, and the total number of vehicles that travel on path R2 and arrive at the designated location P.

[0061] The detection data acquisition unit 114 acquires detection data from the detection vehicles 20 traveling on each path (step S104). That is, the detection data acquisition unit 114 is configured to acquire detection data from each of the multiple detection vehicles 20 (specific vehicles) traveling on the path. The detection data acquisition unit 114 can store the acquired detection data in a database composed of RAM 106. As described above, the detection data includes the identification information, time, location information, section information (road segment information), speed information, congestion information, etc., of the corresponding detection vehicle 20. Furthermore, the detection data may also include path identification information of the path traveled by the corresponding detection vehicle 20.

[0062] The path share calculation unit 116 uses the detection data to calculate the path share ratio (step S106). That is, the path share calculation unit 116 is configured to use the detection data to calculate the path share ratio. Here, the path share ratio represents the proportion of traffic volume in the traffic volume data obtained in S102 that is allocated to each path.

[0063] Specifically, the modal share calculation unit 116 calculates the modal share of traffic volume for each path based on the number of specific vehicles traveling on each of the multiple paths. More specifically, the modal share calculation unit 116 calculates the modal share for each path as the ratio of the number of detection vehicles 20 traveling on each path to the total number of detection vehicles 20 arriving at the designated location P. For example, suppose that the total number of detection vehicles 20 arriving at the designated location P is 10, the number of detection vehicles 20 traveling on path R1 is 7, and the number of detection vehicles 20 traveling on path R2 is 3. In this case, the modal share calculation unit 116 calculates the modal share for path R1 as 70% and the modal share for path R2 as 30%.

[0064] The required time acquisition unit 118 acquires the actual value of the required time (step S108). That is, the required time acquisition unit 118 is configured to acquire the actual value (actual travel time) of each moment (each time period) of the required time for a vehicle to travel from a reference point S along at least one path to the designated location P, for each path. Specifically, the required time acquisition unit 118 uses multiple detection data to acquire (calculate) the required time. More specifically, the required time acquisition unit 118 calculates the required time for each time period for each path using detection data acquired from the detection vehicle 20 traveling along that path.

[0065] Figure 8 This is a diagram illustrating the method for obtaining the required time according to Embodiment 1. Figure 8 The image above is a contour diagram showing the average velocity over different time periods within each section of the path. Additionally, Figure 8 The following graph is a curve corresponding to the contour map, showing the actual values ​​of the required time (travel time) for each time period. Here, it is assumed that... Figure 8 The contour plot of path R1 and the required time are shown. It should be noted that the processing of path R1 will be described in the following description, but the same processing will be performed on path R2.

[0066] Figure 8 The contour maps and curves shown depict time periods horizontally. Furthermore, the contour maps vertically represent the various sections (segments) of the path. Vertically, the lower level is the reference point S1, and the upper level is the designated point P. Therefore, in the contour maps, the direction of travel for path R1 is upward. And, in... Figure 8 In the example, along path R1, from the reference point S1 to the designated point P, there are 13 predetermined intervals #1 to #13. Here, the position of each interval is predetermined, and therefore the distance between each interval is also predetermined.

[0067] Furthermore, the shadow lines for each interval within each time period correspond to the average (harmonic mean) speed of the probe 20 entering that interval during that time period. For example, in "Interval #1" of the time period "6:00", it shows that the average speed of the probe 20 entering Interval #1 from 6:00 to 6:15 is above 30 km / h. Similarly, in "Interval #13" of the time period "7:30", it shows that the average speed of the probe 20 entering Interval #13 from 7:30 to 7:45 is between 15.0 km / h and 19.9 km / h.

[0068] The time acquisition unit 118 calculates the average speed of the probe vehicle 20 for each section within each time period along the path R1. Then, the time acquisition unit 118 calculates the average travel time (average required time) of the probe vehicle 20 for each section within each time period based on the distance of each section and the average speed. For example, if the average speed of "section #13" in the time period "7:30" is V1 [km / h] and the distance of section #13 is L1 [km], the average travel time of "section #13" in the time period "7:30" is L1 / V1 [h].

[0069] Then, the time acquisition unit 118 calculates the required time (travel time) for path R1 in each time period by summing the average travel times of intervals #1 to #13. Thus, as... Figure 8 The following figure shows a graph representing the actual time required for each time period. It should be noted that... Figure 8 In the diagram below, the solid line represents the actual time required for path R1 over time. For example, the required time for path R1 in time slot "6:00" is approximately 5.5 minutes. Additionally, the required time for path R1 in time slot "7:30" is approximately 11 minutes. Furthermore, the required time for path R1 in time slot "9:45" is approximately 7 minutes. Therefore, since the required time for path R1 in time slot "7:30" is longer than in other time slots, it can be inferred that congestion may have occurred on path R1 in time slot "7:30". Furthermore, in Figure 8 In the diagram below, the dashed line represents the free-flow travel time (time required when not congested) of path R1 over time. The calculation method for free-flow travel time will be described later.

[0070] The traffic volume calculation unit 120 uses traffic volume data and the actual value of the required time to calculate the inflow traffic volume for each path (step S110). Here, the inflow traffic volume (generated traffic volume) for each path refers to the traffic volume flowing from the reference point S into that path (inflow traffic volume). That is, the traffic volume calculation unit 120 uses traffic volume data and the actual value of the required time to calculate the inflow traffic volume flowing from the reference point into the path at each time moment (each time period).

[0071] Figure 4 This is a flowchart illustrating the processing of the traffic volume calculation unit 120 according to Embodiment 1. Additionally, Figure 9 This is a diagram illustrating the processing of the traffic volume calculation unit 120 according to Embodiment 1. Figure 9 This is a graph showing the traffic volume at each time point (each time period) along path R1. Figure 9In the diagram, the horizontal axis represents time, the vertical axis on the left represents traffic volume (traffic volume per unit time), and the vertical axis on the right represents cumulative traffic volume. Figure 9 The points represented by the quadrilateral indicate the outflow traffic volume (traffic volume per unit time) of path R1 over time. Additionally, Figure 9 The circles represent the inflow traffic volume (traffic volume per unit time) of path R1 over time. Additionally, Figure 9 The solid line represents the cumulative outflow traffic volume (cumulative outflow traffic volume) of path R1 over time. Additionally, Figure 9 The dashed line represents the cumulative inflow traffic volume (cumulative inflow traffic volume) of path R1 over time.

[0072] The outflow traffic volume of path R1 represents the traffic volume of vehicles passing through path R1 and exiting from designated location P (that is, entering the parking lot from designated location P). In other words, the outflow traffic volume of path R1 represents the traffic volume exiting from path R1. In other words, the outflow traffic volume of path R1 represents the traffic volume of vehicles passing through path R1 and entering the parking lot from designated location P. Furthermore, the cumulative outflow traffic volume represents the cumulative outflow traffic volume over time. Additionally, the cumulative inflow traffic volume represents the cumulative inflow traffic volume over time. Through processing by the traffic volume calculation unit 120, the following can be calculated: Figure 9 The illustrated graph.

[0073] The outflow traffic volume calculation unit 122 calculates the cumulative outflow traffic volume for each path (step S112). Specifically, the outflow traffic volume calculation unit 122 uses the path share ratio and traffic volume data calculated through the processing in S106 to calculate the cumulative outflow traffic volume for each path. More specifically, for each path, the outflow traffic volume calculation unit 122 calculates the outflow traffic volume for each path by multiplying the traffic volume in the traffic volume data by the path share ratio of that path. For example, if the path share ratio of path R1 is 70%, the outflow traffic volume is calculated by multiplying the traffic volume in the traffic volume data by the path share ratio of that path. Figure 7 The outflow traffic volume for path R1 is calculated by multiplying the traffic volume by 0.7. Therefore, the outflow traffic volume for path R1 is obtained as follows: Figure 9 The points of the quadrilateral represent the outflow traffic volume.

[0074] Then, the outflow traffic volume calculation unit 122 calculates the cumulative outflow traffic volume for each path by accumulating the outflow traffic volume for that path. Thus, the following is obtained: Figure 9 The solid line represents the cumulative outflow traffic volume. It should be noted that the outflow traffic volume calculation unit 122 calculates the cumulative traffic volume after obtaining traffic volume data for a specified location P. Figure 7 In the case of a traffic flow, the cumulative outflow traffic volume for each path can also be calculated by multiplying the cumulative traffic volume of the traffic volume data by the path share ratio of that path.

[0075] The inflow traffic volume calculation unit 124 calculates the cumulative inflow traffic volume for each path (step S114). Specifically, for each path, the inflow traffic volume calculation unit 124 calculates the cumulative inflow traffic volume for that path using the actual values ​​of the cumulative outflow traffic volume and the required time. More specifically, for each time period, the inflow traffic volume calculation unit 124 calculates the cumulative outflow traffic volume in the past direction on the time axis (…). Figure 9 The actual value of the time required for the vehicle to move to the left (the direction of movement) corresponding to that time period. Therefore, the inflow traffic volume calculation unit 124 calculates the cumulative inflow traffic volume.

[0076] For example, in Figure 8 In the example, the position of the cumulative inflow traffic volume relative to the cumulative outflow traffic volume shifts to the left approximately 5.5 minutes during the time period "6:00", approximately 6 minutes during the time period "6:15", and approximately 6.5 minutes during the time period "6:30". Similarly, the position of the cumulative inflow traffic volume relative to the cumulative outflow traffic volume shifts to the left approximately 7.5 minutes during the time period "6:45" to "7:00", and approximately 11 minutes during the time period "7:15" to "8:00". Likewise, the position of the cumulative inflow traffic volume relative to the cumulative outflow traffic volume shifts to the left approximately 8.5 minutes during the time period "8:15", and approximately 7 minutes during the time period "8:30" to "9:45".

[0077] The inflow traffic volume calculation unit 124 calculates the inflow traffic volume (traffic volume per unit time) for each path based on the cumulative inflow traffic volume (step S116). Specifically, the inflow traffic volume calculation unit 124 calculates the inflow traffic volume by re-plotting the cumulative inflow traffic volume as a value for each unit time (e.g., every 5 minutes). In other words, the inflow traffic volume calculation unit 124 calculates the inflow traffic volume for each moment (each time period) by decomposing the cumulative inflow traffic volume into values ​​for each unit time. More specifically, the inflow traffic volume calculation unit 124 calculates the inflow traffic volume for each time period by calculating the difference between the value of the cumulative inflow traffic volume for a certain time period and the value for the previous time period.

[0078] Traffic model generation unit 130 generates a traffic model for each route. Figure 3(Step S120). The traffic model is generated by adjusting the parameters of the traffic model in a way that reproduces the actual value of the required time obtained by the processing in S108 when the inflow traffic volume calculated by the processing in S110 is input into the traffic model. Here, the traffic model can be generated for each path. The traffic model can be configured to output the predicted value of the outflow traffic volume and the predicted value of the required time for the corresponding path if the inflow traffic volume of the corresponding path is input. Moreover, the traffic model can be a model that takes into account the waiting queue of vehicles at the bottleneck of the path.

[0079] The parameter setting unit 128 sets the parameters of the traffic model. Here, the parameters of the traffic model include, for example, free-flow travel time and bottleneck capacity. The parameter setting unit 128 sets the free-flow travel time and bottleneck capacity. Bottleneck capacity will be discussed later. Here, the free-flow travel time and bottleneck capacity are not constant along the time axis. That is, the free-flow travel time may be a different value for each moment (each time period). Similarly, the bottleneck capacity may be a different value for each moment (each time period). It should be noted that the reasons for the free-flow travel time and bottleneck capacity differing for each moment (each time period) include, for example, differences in the signal light intervals for each time period, changes in the number of vehicles turning left or right due to increases or decreases in pedestrian traffic, and changes in the proportion of large vehicles entering the path.

[0080] use Figure 8 This explains the method for setting free-flow travel time. The free-flow travel time for a given path is the time (travel time) required when that path is not congested. For example, congestion on the path can be determined if, for instance, there are consecutive intervals where the average speed is below a predetermined number of predetermined speeds. Figure 8 In the time period after "8:00" (the time period enclosed by the double-dotted line in the diagram below), there were no more than three consecutive sections with speeds below 30 km / h. Therefore, it can be determined that no congestion occurred during this time period. Therefore, the parameter setting unit 128... Figure 8 The free-flow travel time after "8:00" is set as the average of the required time (travel time) for that period (approximately 7 minutes).

[0081] Furthermore, within the time period from 7:00 to 8:30 (the time period enclosed by the dotted line in the diagram below), there were three or more consecutive intervals with speeds below 30 km / h. Therefore, it can be determined that congestion occurred during this time period. Therefore, the parameter setting unit 128... Figure 8The free-flow travel time from 7:00 to 8:30 is set to the free-flow travel time (approximately 7 minutes) during periods without congestion, i.e., after 8:00. Furthermore, the time period before 7:00 (the time period enclosed by the dashed line in the diagram below) gradually approaches a critical state from low traffic volume, and the free-flow travel time increases. Therefore, the parameter setting unit 128... Figure 8 The free-flow travel time before "7:00" is set as the average of the required time (travel time) for that time period.

[0082] Figure 10 This is a diagram illustrating bottlenecks in a path using the traffic model described in Embodiment 1. Generally, in traffic, a bottleneck is a location that obstructs traffic flow. That is, a bottleneck is a location with low traffic capacity. Furthermore, bottleneck capacity is the traffic capacity at the bottleneck. For example, bottlenecks may arise at intersections, junctions, merging / lane reduction points, curves, narrow roads, uphill sections, road construction / traffic control points, land entrances, parking lot entrances, and within parking lots. Here, the traffic capacity of the path is determined by the relatively small capacity of the bottleneck. That is, if the upstream bottleneck capacity is smaller than the downstream capacity, there is a high probability that congestion will occur at the upstream bottleneck but not at the downstream bottleneck. Therefore, in the traffic model described in this embodiment, it is assumed that bottlenecks in the path are as follows: Figure 10 As shown, it is set as the downstream end. Therefore, in the traffic model, the distance between the bottleneck and the designated location P is ignored. In addition, in the traffic model, it is assumed that the vehicle travels freely from the reference point S to the vicinity of the designated location P (bottleneck) and stays at the designated location P (bottleneck).

[0083] Here, an example of a traffic model is shown. Let T be the time required for path k at each time t. k (t), let the outflow traffic volume be d k (t). In this case, the required time T k (t) is represented by Equation 1 below. It should be noted that, in the following explanations of mathematical expressions, for convenience, mathematical expressions with a horizontal bar “-” above the letter X are sometimes written as “X…” - ".

[0084]

Mathematical Formula 1

[0085]

[0086] Here, q - k,t This represents the bottleneck capacity at time t for path k. It should be noted that, if using... Figure 10As described above, in the traffic model involved in this embodiment, a bottleneck is set at the downstream end of path k (a designated location P).

[0087] In addition, T - k,t It is the free-flow travel time at time t for path k. As mentioned above, T - k,t It can be preset by the parameter setting unit 128. Furthermore, as shown in Equation 2 below, A k (t) represents the cumulative inflow traffic volume at time t for path k. It should be noted that t' represents the timed processing time prior to time t. For example, the time T required to calculate the traffic volume at 5-minute intervals... k (t) and outflow traffic volume d k In the case of (t), t' = t - 5 [points].

[0088]

Mathematical Formula 2

[0089]

[0090] It should be noted that in Equation 2, the boundary condition is set as A. k (-T - k,0 ) = 0. Additionally, a k (t) represents the inflow traffic volume at time t along path k. This a k (t) corresponds to the inflow traffic volume calculated through the processing in S110 (S116). Therefore, a k (t) Input to the traffic model.

[0091] Additionally, as shown in Equation 3 below, D k (t) represents the cumulative outflow traffic volume at time t for path k.

[0092]

Mathematical Expression 3

[0093]

[0094] It should be noted that in Equation 3, the boundary condition is set to D. k (0) = 0. Furthermore, as shown in Equation 4 below, d k (t) represents the outflow traffic volume at time t for path k.

[0095]

Mathematical Expression 4

[0096]

[0097] In addition, a k (tT - k,t This indicates that the free-flowing travel time T was traced back from time t.- k,t The traffic volume flowing from the reference point S into path k. Here, in this traffic model, it is assumed that the travel time from the reference point S to the designated location P (bottleneck) is free-flowing. Therefore, a k (tT - k,t This corresponds to the traffic volume of vehicles that arrive at the bottleneck (designated location P) at time t. Similarly, A k (tT - k,t This indicates that the free-flowing travel time T was traced back from time t. - k,t The cumulative traffic volume flowing from the base point S into path k. And, A k (tT - k,t This corresponds to the cumulative traffic volume of vehicles that arrive at the bottleneck (designated location P) at time t.

[0098] Additionally, the top row on the right side of Equation 4 indicates A. k (tT - k,t ) = D k In the case of (t), d k (t)=a k (tT - k,t ). And, A k (tT - k,t ) = D k The case of (t) means that there is no delay (waiting queue) at the bottleneck (the designated location P). Therefore, at time (tT) - k,t Inflow traffic a into path k k (tT - k,t All of them flow out from path k at time t. Therefore, d k (t)=a k (tT - k,t It should be noted that, in this case, at time (tT) - k,t Inflow traffic a into path k k (tT - k,t At time t, the bottleneck capacity q has not been exceeded. - k,t Therefore, the inflow traffic volume a k (tT - k,t All of them flow out from path k at time t.

[0099] Additionally, the lower row of Equation 4 indicates A k (tT - k,t )>D k In the case of (t), d k (t)=q - k,t Furthermore, A k (tT - k,t )>D k The case of (t) implies that a bottleneck (waiting queue) has occurred at the bottleneck (the designated location P). Therefore, it is not at time (tT) - k,t Inflow traffic a into path k k (tT - k,t All traffic flows out of path k at time t, and a portion (or, if the waiting queue is large, all) remains stuck at the bottleneck. Furthermore, the traffic volume corresponding to the stuck waiting queue exceeds the bottleneck capacity q. - k,t In the case of outflow traffic volume d k (t) becomes the bottleneck capacity -q k,t .

[0100] Furthermore, in Equation 1 above, "A" k (tT - k,t )-D k "(t)" represents the size of the waiting queue (number of vehicles). Therefore, the second term on the right represents the time spent in the waiting queue (delay time). Thus, Equation 1 means the time T required for path k at time t. k (t) is the free-flowing travel time T. - k,t Total of the time spent in the area.

[0101] For each path, the traffic model generation unit 130 inputs the inflow traffic volume calculated by the processing in S110 into the traffic model represented by Equations 1 to 4 above. Then, the traffic model generation unit 130 calculates the predicted time T required at each time t (each time period) as shown in Equation 1 by sequentially performing Equations 1 to 4 starting from t=0. k (t). Then, the traffic model generation unit 130 calculates the predicted time T. k (t) and the actual value of the required time ( Figure 8 The curve is fitted to the figure below. Then, the predicted time T is calculated. k(t) Adjust the bottleneck capacity q in a way that fits the actual value of the required time. - k,t (Bottleneck capacity parameter). It should be noted that, for example, the predicted time T required at a certain time t... k (t) If the actual value of the required time deviates significantly from the actual value, adjust the bottleneck capacity q corresponding to that time t. - k,t That's all.

[0102] Figure 11 This is a diagram illustrating the bottleneck capacity adjusted during the generation of the traffic model involved in Implementation Method 1. Additionally, Figure 12 This is a graph illustrating the predicted time calculated by the traffic model involved in Implementation 1. Figure 11 and Figure 12 These represent the bottleneck capacity and required time for path R1, respectively. Here, in Figure 11 In the graph, the horizontal axis represents time, and the vertical axis represents the bottleneck capacity value. Additionally, Figure 11 The dashed line represents the bottleneck capacity over time. Additionally, in Figure 12 In the diagram, the horizontal axis represents time, and the vertical axis represents the required travel time. Additionally, in... Figure 12 In the diagram, solid lines represent the predicted time required over time, while dashed lines represent the actual time required over time. Additionally, dotted lines represent the free-flowing travel time over time.

[0103] For example, Figure 12 The traffic model involved can be used to Figure 11 The bottleneck capacity shown is a parameter. For example... Figure 12 As illustrated, the predicted time calculated by the traffic model generated by the traffic model generation unit 130 largely and accurately reproduces the actual time required. Furthermore, as... Figure 11 As illustrated, the bottleneck capacity is not constant over all time periods, but rather a value that varies for each time period. In other words, the bottleneck capacity is a parameter for each path at each time period. By configuring the bottleneck capacity to be different for each time period, the traffic model can calculate a predicted value of the required time that accurately reproduces the actual required time. Therefore, in this embodiment, the traffic model defined by the above formula can be used to predict the required time T of path k with high accuracy. k (t) and outflow traffic volume d k (t).

[0104] Furthermore, in this embodiment, the required time T for path k can be predicted with high accuracy using a traffic model. k(t), thus enabling the calculation of the predicted time required to cause a change in inflow traffic volume with high accuracy. Therefore, in this embodiment, it is possible to calculate the change in the predicted time required to cause a change in inflow traffic volume with high accuracy. This will be described in detail below.

[0105] Figure 5 This is a flowchart illustrating the process of calculating the change in the predicted time required to change the inflow traffic volume using the traffic model according to Embodiment 1. Here, "change in inflow traffic volume" is performed for the evaluation of the situation where measures (action changes) to reduce traffic congestion have been implemented. That is, the traffic prediction device 100 uses the inflow traffic volume that takes into account the change in inflow traffic volume when a certain measure is implemented to calculate the change in the time required to implement that measure.

[0106] Steps S132 to S140 are respectively with Figure 3 The processing of S102 to S110 is essentially the same, therefore, the explanation is omitted. It should be noted that the processing of generating the traffic model ( Figure 3 The inflow traffic volume of each generated path is... Figure 5 In cases where this process is also used, the processes S132 to S140 can be omitted.

[0107] The change setting unit 140 sets the change in inflow traffic volume (step S142). Specifically, the change setting unit 140 sets the change in inflow traffic volume corresponding to the envisioned measure. In an example measure, there is a possibility of converting X% of employee vehicles (target vehicles) traveling on the evaluation target's route to bus commuting (transportation mode conversion) such as local buses. Additionally, in another example measure, there is a possibility of converting X% of employee vehicles (target vehicles) traveling during peak traffic periods to off-peak periods (off-peak commuting). For example, in the case of the measure "converting 30% of employee vehicles to bus commuting," the change setting unit 140 can reduce the inflow traffic volume by 30% for all time periods. That is, the change setting unit 140 sets the inflow traffic volume to 0.7 times for all time periods. Furthermore, for example, in the case of the measure "converting 30% of employee vehicles to bus commuting," the change setting unit 140 can also reduce the inflow traffic volume by 30% during congested periods (described later). Additionally, for example, in the case of a measure that "30% of employee vehicles passing through peak hours will be converted to time periods outside of peak hours", the change setting unit 140 can also shift 30% of the inflow traffic volume during peak hours to time periods outside of peak hours.

[0108] The traffic state prediction unit 150 calculates the predicted time required (step S150). That is, for each path, the traffic state prediction unit 150 uses the predicted time required for the inflow traffic volume calculated in S140 and the predicted time required under conditions of change in the inflow traffic volume to calculate the predicted time required for each time period before the implementation of the measure. Specifically, in the processing of S120, the parameters of the traffic model (free-flow travel time and bottleneck capacity) are determined. Then, the traffic state prediction unit 150 calculates the predicted time required for each time period before the implementation of the measure by inputting the inflow traffic volume for each time period calculated through the processing of S140 into the traffic model after the parameters are determined. Similarly, the traffic state prediction unit 150 calculates the predicted time required for each time period under conditions of change in the inflow traffic volume (that is, when the measure is implemented) by inputting the inflow traffic volume for each time period considering the changes set through the processing of S142 into the traffic model after the parameters are determined.

[0109] The change calculation unit 160 calculates the change in required time (step S160). That is, for each path, the change calculation unit 160 calculates the change in the predicted value of the required time under conditions that cause a change in inflow traffic volume. Specifically, the change calculation unit 160 calculates the difference between the predicted value of the required time relative to the inflow traffic volume before the change (before the measure was implemented) and the predicted value of the required time under conditions that cause a change in inflow traffic volume (when the measure was implemented). For example, the change (difference) could be the difference between the peak value of the predicted value of the required time before the measure was implemented and the peak value of the predicted value of the required time when the measure was implemented. Alternatively, for example, the change (difference) could also be the difference between the average value of the predicted value of the required time before the measure was implemented and the average value of the predicted value of the required time when the measure was implemented. Furthermore, for example, the change (difference) could also be the difference between the length of the congestion period under the predicted value of the required time before the measure was implemented and the length of the congestion period under the predicted value of the required time when the measure was implemented. Here, the congestion period can be the period during which the predicted value of the required time exceeds the free-flowing travel time.

[0110] Alternatively, for example, the change (difference) could also be the difference between the area defined by the curve of the predicted time required before the measure is implemented and the area defined by the curve of the predicted time required at the time the measure is implemented. Here, "the area defined by the curve of the predicted time required" could be the value obtained by integrating the predicted time required over the entire time period (that is, the area between the curve of the predicted time required and the horizontal axis). Alternatively, "the area defined by the curve of the predicted time required" could also be the area between the curve of the predicted time required and the curve of free-flowing travel time.

[0111] Furthermore, if the traffic prediction device 100 has not completed the processing of S142 to S160 for all measures (No in step S170), the processing returns to S142. Then, the traffic prediction device 100 processes the remaining measures that were not completed in S142 to S160. On the other hand, if the traffic prediction device 100 has completed the processing of S142 to S160 for all measures (Yes in S170), the processing ends.

[0112] Figures 13-16 This is a graph comparing the predicted time required before and during the implementation of the measures involved in Implementation Method 1. Figures 13-16 In the diagram, the dashed line represents the predicted time required over time under the condition of no intervention (i.e., a conversion rate of 0%). The solid line represents the predicted time required over time under the condition of intervention. The dotted-dash line represents the free-flowing travel time over time. The horizontal axis represents time, and the vertical axis to the left represents travel time (required time) [seconds].

[0113] Figures 13-14 This shows the predicted time required when route R1 implements the "mode-of-transport" measure. Figures 13-14 In the diagram, the points on the quadrilateral represent the number of vehicles switching from private car commuting to public bus commuting at each time point over time, and the vertical axis on the right represents the number of conversions. Figure 13 The diagram shows a comparison of the time required to implement the measure "convert 5% of employee vehicles to bus commuting" with the time required before implementing the measure. Figure 14 This shows a comparison of the time required to implement the measure "convert 30% of employee vehicles to bus commuting" with the time required before implementing the measure. Figure 13 As illustrated, with a conversion rate of 5%, the change in required time (travel time) during peak hours is -0.4 minutes. On the other hand, as... Figure 14 As illustrated, with a conversion rate of 30%, the change in required time (travel time) during peak hours is -2.1 minutes. Therefore, managers can effectively evaluate initiatives such as "converting 5% of employee vehicles to bus commuting" and "converting 30% of employee vehicles to bus commuting".

[0114] Figures 15-16 This shows the predicted time required for path R1 to implement "off-peak commuting" measures. Figure 15 The diagram shows a comparison of the time required to implement the measure "shifting 5% of employee vehicles that travel during peak hours to outside of peak hours" with the time required before implementing the measure. Figure 16This shows a comparison of the time required to implement the measure "shifting 30% of employee vehicles traveling during peak hours to outside peak hours" with the time required before implementing the measure. Here, in Figure 15 and Figure 16 In the example, the peak period is set as one hour from 6:40 to 7:40. Furthermore, for employee vehicles corresponding to the first half of the peak period (6:40 to 7:10), they are required to depart 15 minutes earlier, and for employee vehicles corresponding to the second half of the peak period (7:10 to 7:40), they are required to depart 15 minutes later.

[0115] like Figure 15 As illustrated, with a conversion rate of 5%, the change in required time (travel time) during peak hours is -0.4 minutes. On the other hand, as... Figure 16 As illustrated, with a conversion rate of 30%, the change in required time (travel time) during peak hours is -2.4 minutes. Therefore, managers can effectively evaluate measures such as "converting 5% of employee vehicles traveling during peak hours to outside peak hours" and "converting 30% of employee vehicles traveling during peak hours to outside peak hours."

[0116] When a facility manager (system administrator) predicts traffic flow for vehicles associated with a given location P, they sometimes want to predict how much congestion will be alleviated and how much the travel time will be reduced by changes in vehicle behavior. To predict traffic conditions, traffic volume data is needed; however, obtaining infrastructure data such as traffic counter information is often difficult.

[0117] The traffic prediction system 1 of this embodiment is configured to calculate the inflow traffic volume for each path based on traffic volume data at a designated location P and the actual value of the required time. Furthermore, the traffic prediction system 1 of this embodiment is configured to calculate, for each path, the predicted value of the required time for the calculated inflow traffic volume and the predicted value of the required time under conditions of change in the inflow traffic volume using a pre-generated traffic model. Moreover, the traffic prediction system 1 of this embodiment is configured to calculate, for each path, the change in the predicted value of the required time under conditions of change in inflow traffic volume. With this structure, even without obtaining infrastructure data, it is possible to predict changes in the required time under conditions of change in inflow traffic volume. Therefore, the traffic prediction system 1 of this embodiment can efficiently predict changes in the required time for traversing a path when traffic volume changes along the path due to changes in movement.

[0118] Furthermore, the traffic prediction system 1 according to this embodiment is configured to obtain the actual value of the required time using detection data (actual driving data) obtained from multiple detection vehicles traveling on the path. Here, the detection data is obtained from the detection vehicles 20 that can be managed by the system administrator. Therefore, the system administrator can easily obtain the detection data. Thus, the traffic prediction system 1 according to this embodiment can easily obtain the actual value of the required time.

[0119] Furthermore, the traffic prediction system 1 according to this embodiment is configured to: obtain actual values ​​of the required time for each of multiple paths, calculate the inflow traffic volume, calculate the predicted value of the required time, and calculate the predicted value of the required time under conditions that cause changes in the inflow traffic volume. Moreover, the traffic prediction system 1 according to this embodiment is configured to: calculate the amount of change in the predicted value of the required time under conditions that cause changes in the inflow traffic volume for each of the multiple paths. With this structure, it is possible to predict which path will have a large change in required time. Therefore, it is possible to determine the path where the effect of action change is high. Thus, action change measures can be actively implemented on the path where the effect of action change is high.

[0120] Furthermore, the traffic prediction system 1 according to this embodiment is configured to calculate the inflow traffic volume of each of the multiple paths using the modal share of traffic volume for each path in traffic volume data, calculated based on the number of detector vehicles traveling on multiple paths respectively. With this structure, the inflow traffic volume of each path can be calculated even without obtaining infrastructure data such as traffic counter information.

[0121] Furthermore, the traffic prediction system 1 according to this embodiment is configured to: calculate the cumulative outflow traffic volume for each of multiple paths using modal share, calculate the cumulative inflow traffic volume using the actual values ​​of the cumulative outflow traffic volume and the required time, and calculate the inflow traffic volume based on the cumulative inflow traffic volume. With this structure, the inflow traffic volume for each path can be calculated even without obtaining infrastructure data such as traffic counter information.

[0122] Furthermore, in the traffic prediction system 1 according to this embodiment, the traffic model is configured to be generated by adjusting the parameters of the traffic model in a manner that reproduces the actual value of the required time when the inflow traffic volume is input into the traffic model. With such a structure, the predicted value of the required time can be calculated with high accuracy based on the inflow traffic volume.

[0123] Furthermore, in the traffic prediction system 1 according to this embodiment, the parameters of the traffic model are configured to include the bottleneck capacity of each path at each time. With this structure, it is possible to calculate the predicted time that accurately reproduces the actual time required. Therefore, in this embodiment, it is possible to predict the required time and outflow traffic volume with high accuracy.

[0124] (Implementation Method 2)

[0125] Next, regarding implementation method 2, while referring to the appendix... Figure 1 The following description is provided for clarity. To simplify the explanation, the following descriptions and figures have been appropriately omitted and simplified. Furthermore, in each figure, the same reference numerals are used to label the same elements, and repeated descriptions have been omitted as needed. It should be noted that the system structure related to Embodiment 2 is different from... Figure 1 The system architecture shown is essentially the same, therefore description is omitted. Implementation 2 differs from Implementation 1 in that it involves the ability to obtain infrastructure data.

[0126] Figure 17 This is a diagram showing the structure of the traffic prediction device 100 according to Embodiment 2. Additionally, Figures 18-19 This is a flowchart illustrating the traffic prediction method executed by the traffic prediction system 1 according to Embodiment 2. Similar to Embodiment 1, the traffic prediction device 100 according to Embodiment 2 has a CPU 102, ROM 104, RAM 106 and interface unit 108 as the main hardware structure.

[0127] Similar to Embodiment 1, the traffic prediction device 100 in Embodiment 2, as a component of the representation function, includes a route information storage unit 110, a traffic volume data acquisition unit 112, a detection data acquisition unit 114, a route share calculation unit 116, and a required time acquisition unit 118. Furthermore, the traffic prediction device 100 in Embodiment 2 includes a traffic volume calculation unit 120, a parameter setting unit 128, and a traffic model generation unit 130. Additionally, the traffic prediction device 100 in Embodiment 2 includes a change setting unit 140, a traffic state prediction unit 150, and a change calculation unit 160. The traffic prediction device 100 in Embodiment 2 also includes an infrastructure data acquisition unit 210, a mixing rate calculation unit 212, and an overall traffic volume calculation unit 220.

[0128] In addition, Figure 18 In the flowchart shown, the processes S202 to S210 are respectively related to... Figure 3 The processes shown in S102 to S110 are essentially the same, so their description is omitted. The infrastructure data acquisition unit 210 acquires infrastructure data (step S212). Specifically, the infrastructure data acquisition unit 210 acquires infrastructure data collected by equipment managed by road managers (e.g., administrative agencies) that are different from the system manager. Infrastructure data, for example, is data collected from traffic counters (traffic counter data).

[0129] More specifically, the infrastructure data acquisition unit 210 acquires infrastructure data representing traffic volume before and after a designated location P in the same direction as the travel direction of vehicles on each path. For example, regarding... Figure 6 For the illustrated path R1, the infrastructure data acquisition unit 210 acquires infrastructure data (traffic counter data) from a traffic counter located on the side of the reference point S1 relative to the designated location P and measuring the traffic volume in the same direction as the travel direction of path R1. Additionally, the infrastructure data acquisition unit 210 acquires infrastructure data (traffic counter data) from a traffic counter located on the side opposite to the reference point S1 relative to the designated location P and measuring the traffic volume in the same direction as the travel direction of path R1.

[0130] The mixing rate calculation unit 212 calculates the mixing rate of the target vehicle relative to the overall traffic volume (step S214). Here, "overall traffic volume" refers to the overall traffic volume of each route. That is, the overall traffic volume for each route refers to the traffic volume related to vehicles that combine the target vehicle and other general vehicles. Furthermore, the mixing rate represents the traffic volume (number of vehicles entering the parking lot per unit time) related to the target vehicle relative to the overall traffic volume. Specifically, for each route, the mixing rate calculation unit 212 calculates the mixing rate for each time period using the ratio of the number of parking lot entries per unit time represented by traffic volume data to the traffic volume represented by infrastructure data obtained in S212 as the mixing rate. Here, if using Figure 20 As will be described later, the mixing rate changes over time.

[0131] The overall traffic volume calculation unit 220 calculates the overall traffic volume (step S216). Specifically, the overall traffic volume calculation unit 220 uses the mixing ratio calculated in S214 and the inflow traffic volume calculated in S210 to calculate the overall traffic volume. More specifically, for each time period, the overall traffic volume calculation unit 220 calculates the overall traffic volume flowing into each path by dividing the inflow traffic volume by the mixing ratio.

[0132] Figure 20 This is a graph illustrating the mixing rate and overall traffic volume involved in Implementation Method 2. Figure 20 The diagram illustrates the traffic volume and mixing rate for path R1. Figure 20 In the graph shown, the horizontal axis represents time, the left vertical axis represents traffic volume (number of vehicles every 5 minutes), and the right vertical axis represents the mixing rate of employee vehicles. Additionally, Figure 20 The points represented by the quadrilateral indicate the inflow traffic volume of path R1 over time. It should be noted that... Figure 20 The point shown by the quadrilateral corresponds to Figure 9 The inflow traffic volume is shown (represented by circular dots). Additionally, Figure 20The circles represent the points indicating the overall traffic volume of path R1 over time. Additionally, Figure 20 The dashed line represents the rate of employee vehicles entering the system over time.

[0133] exist Figure 20 In the example, the employee vehicle mix-in rate for route R1 increases from approximately 5% to 30% during the time period from 6:00 to 7:00. Additionally, the employee vehicle mix-in rate for route R1 increases by approximately 30% during the time period from 7:00 to 8:45, peaking at approximately 35% at 8:15. Furthermore, the employee vehicle mix-in rate for route R1 decreases from approximately 30% to 5% during the time period from 8:45 to 9:15. The mix-in rate calculation unit 212 calculates the overall traffic volume represented by the circular dots for each time period by dividing the inflow traffic volume represented by the quadrilateral dots by the mix-in rate represented by the dashed lines.

[0134] In Embodiment 2, the traffic model generation unit 130 generates a traffic model for each path (step S220). Here, in Embodiment 2, the traffic model generation unit 130 uses the overall traffic volume calculated in S216 to generate the traffic model. In Embodiment 2, the traffic model is generated by adjusting its parameters to reproduce the actual value of the required time obtained through the processing in S208 when the overall traffic volume flowing into the path, calculated through the processing in S216, is input into the traffic model. Therefore, the parameter setting unit 128 can set the bottleneck capacity for the overall traffic volume. Furthermore, the traffic model generation unit 130 can generate a traffic model with the bottleneck capacity corresponding to the overall traffic volume as a parameter.

[0135] Figure 19 This is a flowchart illustrating the process of calculating the change in the predicted time required to cause a change in inflow traffic volume using the traffic model according to Embodiment 2. Similar to S132-S140 of Embodiment 1, the traffic prediction device 100 according to Embodiment 2 performs essentially the same process as S202-S216 (step S230). It should be noted that the process of generating the traffic model (…) Figure 18 The generated inflow traffic volume (overall traffic volume) for each path is... Figure 19 In cases where this process is also used, the S230 process can be omitted.

[0136] In Implementation 2, the change setting unit 140 sets the change in the overall traffic volume flowing into the path (step S242). Specifically, the change setting unit 140 sets the change in the overall traffic volume corresponding to the envisioned measure. Here, the target of the measure is employee vehicles (target vehicles). Therefore, the change setting unit 140 in Implementation 2 sets the change based on the ratio of the number of parking lot occupants (facility occupants) per unit time to the overall traffic volume. Specifically, the change setting unit 140 in Implementation 2 sets the change by multiplying the overall traffic volume flowing into each path by the mixing rate and the conversion rate at each time (time period).

[0137] The traffic state prediction unit 150 in Implementation Method 2 calculates the predicted time required (step S250). That is, for each path, the traffic state prediction unit 150 uses the predicted time required for the overall traffic volume calculated in S216 and the predicted time required when the overall traffic volume changes to calculate the time required for each path using a traffic model pre-generated in S220. Specifically, in the processing of S220, the parameters of the traffic model (free-flow travel time and bottleneck capacity) are determined. Then, the traffic state prediction unit 150 calculates the predicted time required for each time period before the implementation of the measure by inputting the overall traffic volume for each time period calculated through the processing of S216 into the traffic model after the parameters are determined. Similarly, the traffic state prediction unit 150 calculates the predicted time required for each time period when the overall traffic volume changes (that is, when the measure is implemented) by inputting the overall traffic volume for each time period considering the changes set through the processing of S242 into the traffic model after the parameters are determined.

[0138] In Embodiment 2, the change calculation unit 160 calculates the change in required time (step S260). That is, for each path, the change calculation unit 160 calculates the change in the predicted time required to cause a change in the overall traffic volume flowing into that path. The method for calculating the change is essentially the same as in Embodiment 1, so its explanation is omitted. Furthermore, if the traffic prediction device 100 in Embodiment 2 has not completed the processing of S242 to S260 for all measures (no in step S270), the processing returns to S242. Then, the traffic prediction device 100 performs the processing of S242 to S260 for other measures that have not been processed. On the other hand, if the traffic prediction device 100 has completed the processing of S242 to S260 for all measures (yes in S270), the processing ends.

[0139] The traffic prediction device 100 according to Embodiment 2 can predict changes in the required time under the condition that measures have been taken, taking into account the overall traffic volume. Therefore, it is possible to make predictions that more closely correspond to the actual state of the path. On the other hand, in this embodiment, as described in Embodiment 1, traffic prediction under the condition of action change can be made even without obtaining infrastructure data. That is, in this disclosure, it is possible to predict changes in the required time when action change is performed even without obtaining infrastructure data.

[0140] (Modified Example)

[0141] It should be noted that the present invention is not limited to the above-described embodiments, and appropriate modifications can be made without departing from the spirit of the invention. For example, Figures 3-5 and Figures 17-19 The order of the steps in the flowchart shown can be changed appropriately. Additionally, Figures 3-5 and Figures 17-19 The flowchart shown may have more than one step, or it may be omitted.

[0142] Furthermore, while the above embodiments process multiple paths separately, the structure is not limited to this. That is, the above processing can also be performed on a single path. However, by processing multiple paths separately as described above, it is possible to determine which path has the highest effectiveness for the action change.

[0143] Furthermore, in Implementation 2, the overall traffic volume is calculated by multiplying the inflow traffic volume by the mixing ratio, but this structure is not limited to this. Alternatively, the overall traffic volume of the outflow traffic volume can be calculated by dividing the outflow traffic volume by the mixing ratio, and the overall traffic volume of the outflow traffic volume can be calculated using the overall traffic volume of the outflow traffic volume. Figure 4 The process shown is used to calculate the overall traffic volume of inflow traffic.

[0144] Furthermore, in the examples above, programs can be stored and supplied to a computer using a wide variety of types of non-transitory computer-readable media. Non-transitory computer-readable media include a wide variety of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., floppy disks, magnetic tapes, hard disks), optical-magnetic recording media (e.g., optical discs), CD-ROMs, CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAM). Additionally, programs can also be supplied to a computer using a wide variety of transient computer-readable media. Examples of transient computer-readable media include electrical signals, optical signals, and electromagnetic waves. Transient computer-readable media can supply programs to a computer via wired or wireless communication paths such as wires and optical fibers.

[0145] As will be apparent from this disclosure as described herein, embodiments of the present disclosure may be varied in many ways. These variations should not be considered as departing from the spirit and scope of the present disclosure, and all such modifications, which will be apparent to those skilled in the art, are intended to be included within the scope of the appended claims.

Claims

1. A traffic prediction system, comprising: The traffic volume data acquisition unit acquires traffic volume data that shows the relationship between time and traffic volume at a specified location; The required time acquisition unit acquires, for each of the multiple paths to the specified location, the actual value of the required time for a vehicle to travel from a reference point along the path to the specified location. The traffic volume calculation unit uses the traffic volume data and the actual value of the required time to calculate the inflow traffic from the reference location to the path; The traffic condition prediction unit, with respect to the route, uses a pre-generated traffic model to calculate the predicted time required for the calculated inflow traffic volume and the predicted time required under conditions that change the inflow traffic volume. and The change calculation unit calculates, with respect to the path, the change in the predicted value of the required time under the condition that the inflow traffic volume changes. The traffic model is generated by adjusting the parameters of the traffic model in a manner that reproduces the actual value of the required time when the inflow traffic volume is input into the traffic model. The parameters of the traffic model include the bottleneck capacity of each of the multiple paths at each time step, and the value of the bottleneck capacity varies for each time step. The traffic model reproduces the actual value of the required time, including the dwell time, which is obtained by dividing a first value by the bottleneck capacity, the first value being obtained by subtracting the cumulative outflow traffic from the cumulative inflow traffic volume.

2. The traffic prediction system according to claim 1, wherein, The required time acquisition unit uses actual driving data obtained from multiple specific vehicles traveling on the path to obtain the actual value of the required time.

3. The traffic prediction system according to claim 1, wherein, The required time acquisition unit uses actual driving data obtained from multiple specific vehicles traveling on multiple said paths to obtain the actual value of the required time. The traffic volume calculation unit uses the modal share of the traffic volume for each path in the traffic volume data to calculate the inflow traffic volume for each of the plurality of paths, the modal share of each path being calculated based on the number of specific vehicles traveling on each of the plurality of paths.

4. The traffic prediction system according to claim 3, wherein, The traffic volume calculation unit uses the modal share to calculate the cumulative outflow traffic volume, which is the sum of the outflow traffic volume from the traffic volume data that flows out from multiple of the aforementioned paths. The traffic volume calculation unit uses the actual values ​​of the cumulative outflow traffic volume and the required time to calculate the cumulative inflow traffic volume, i.e., the cumulative inflow traffic volume. The traffic volume calculation unit calculates the inflow traffic volume at each time point based on the cumulative inflow traffic volume.

5. A traffic prediction method, wherein, Obtain traffic volume data that represents the relationship between time and traffic volume at a specified location. For each of the multiple paths leading to the specified location, obtain the actual time required for a vehicle to travel from a reference point along the path to the specified location. The inflow traffic from the reference point to the path is calculated using the traffic volume data and the actual value of the required time. Regarding the aforementioned path, a pre-generated traffic model is used to calculate the predicted time required for the calculated inflow traffic volume, as well as the predicted time required under conditions that change the inflow traffic volume. Regarding the stated path, calculate the change in the predicted time required, assuming a change in the inflow traffic volume. The traffic model is generated by adjusting the parameters of the traffic model in a manner that reproduces the actual value of the required time when the inflow traffic volume is input into the traffic model. The parameters of the traffic model include the bottleneck capacity of each of the multiple paths at each time step, and the value of the bottleneck capacity varies for each time step. The traffic model reproduces the actual value of the required time, including the dwell time, which is obtained by dividing a first value by the bottleneck capacity, the first value being obtained by subtracting the cumulative outflow traffic from the cumulative inflow traffic volume.

6. A computer-readable medium storing a program that causes a computer to perform the following steps: Obtain traffic volume data that represents the relationship between time and traffic volume at a specified location; For each of the multiple paths leading to the specified location, obtain the actual value of the time required for a vehicle to travel from a reference point on the path to the specified location. The inflow traffic from the reference location to the path is calculated using the traffic volume data and the actual value of the required time. Regarding the route, a pre-generated traffic model is used to calculate the predicted time required for the calculated inflow traffic volume and the predicted time required under different inflow traffic volumes. and Regarding the stated path, calculate the change in the predicted time required, assuming a change in the inflow traffic volume. The traffic model is generated by adjusting the parameters of the traffic model in a manner that reproduces the actual value of the required time when the inflow traffic volume is input into the traffic model. The parameters of the traffic model include the bottleneck capacity of each of the multiple paths at each time step, and the value of the bottleneck capacity varies for each time step. The traffic model reproduces the actual value of the required time, including the dwell time, which is obtained by dividing a first value by the bottleneck capacity, the first value being obtained by subtracting the cumulative outflow traffic from the cumulative inflow traffic volume.

Images