Traffic volume prediction device, traffic volume prediction method, and program

Abstract

To provide a traffic volume prediction device, a traffic volume prediction method, and a program that are capable of appropriately predicting a traffic volume on a route by using an OD matrix that is automatically calibrated.SOLUTION: A traffic volume prediction device comprises: an input value reception unit that receives a current destination matrix, information on a road network, information on a railway network, an input value containing at least one of an actual value of a traveling speed of an automobile on a road and an actual value of the number of users at each station in the railway; a traffic volume prediction unit that predicts a traffic volume of each road in the road network and the railway network by using a characteristic that increases a time of movement for the railway road in accordance with the number of users on the basis of the current destination matrix, information on the road network, and information on the railway network; a movement time error calculation unit that calculates a movement time error indicating an error between a time of movement calculated from the actual value of the traveling speed of the automobile on the road and the time of movement for each road calculated from the predicted value of the traffic volume of each road in the road network; and a current destination matrix calibration unit that corrects the current destination matrix on the basis of the movement time error.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a traffic volume prediction device, a traffic volume prediction method, and a program.

Background Art

[0002] When considering a new railway construction plan or the like, the number of travelers when the railway is constructed is predicted, and the profitability of the business plan is examined. In that case, it is necessary to grasp the traffic volume of existing transportation agencies, and the traffic volume of existing transportation agencies is also predicted.

[0003] For example, Patent Document 1 below discloses a traffic flow simulation device that simulates the driving of automobiles based on an origin-destination (OD) matrix, which is information on the origin and destination of automobiles and the like, and predicts traffic flow.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the traffic flow simulation device described in Patent Document 1 above, it takes time to set the OD matrix, which is a prerequisite for simulation, and the OD matrix cannot be appropriately corrected according to the measured values.

[0006] In view of the above problems, an object of the present disclosure is to provide a traffic volume prediction device, a traffic volume prediction method, and a program that can appropriately predict the traffic volume of each route using an automatically calibrated OD matrix.

Means for Solving the Problems

[0007] In order to solve the above-described problems and achieve the object, a traffic volume prediction apparatus according to the present disclosure includes: an input value reception unit that receives an input value including at least one of a current location - destination matrix, information on a road network, information on a railway network, an actually measured value of the traveling speed of an automobile on a road, and an actually measured value of the number of users at each station on a railway; a traffic volume prediction unit that predicts the traffic volume of each route in the road network and the railway network by using a characteristic of increasing the travel time of a railway route according to the number of users, based on the current location - destination matrix, the information on the road network, and the information on the railway network; a travel time error calculation unit that calculates a travel time error indicating an error between the travel time obtained from the actually measured value of the traveling speed of an automobile on a road and the travel time of each route obtained from the predicted value of the traffic volume of each route in the road network; and a current location - destination matrix correction unit that corrects the current location - destination matrix based on the travel time error.

[0008] In order to solve the above-described problems and achieve the object, a traffic volume prediction method according to the present disclosure includes: a step of receiving an input value including at least one of a current location - destination matrix, information on a road network, information on a railway network, an actually measured value of the traveling speed of an automobile on a road, and an actually measured value of the number of users at each station on a railway; a step of predicting the traffic volume of each route in the road network and the railway network by using a characteristic of increasing the travel time of a railway route according to the number of users, based on the current location - destination matrix, the information on the road network, and the information on the railway network; a step of calculating a travel time error indicating an error between the travel time obtained from the actually measured value of the traveling speed of an automobile on a road and the travel time of each route obtained from the predicted value of the traffic volume of each route in the road network; and a step of correcting the current location - destination matrix based on the travel time error.

[0009] In order to solve the above-described problems and achieve the object, the program according to the present disclosure causes a computer to execute steps of: receiving an input value including at least one of a current location - destination matrix, information on a road network, information on a railway network, a measured value of the traveling speed of an automobile on a road, and a measured value of the number of users of each station on a railway; predicting the traffic volume of each route in the road network and the railway network by using a characteristic of increasing the travel time of a railway route according to the number of users, based on the current location - destination matrix, the information on the road network, and the information on the railway network; calculating a travel time error indicating an error between the travel time obtained from the measured value of the traveling speed of an automobile on a road and the travel time of each route obtained from the predicted value of the traffic volume of each route in the road network; and correcting the current location - destination matrix based on the travel time error.

Advantages of the Invention

[0010] According to the present disclosure, it is possible to provide a traffic volume prediction device, a traffic volume prediction method, and a program that can appropriately predict the traffic volume of each route by using an automatically calibrated OD matrix.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. Note that the present disclosure is not limited by the embodiments described below.

[0013] (Configuration of Traffic Volume Prediction System) First, the traffic volume prediction system 1 according to the present disclosure will be described with reference to FIG. 1. FIG. 1 is a diagram showing a configuration example of the traffic volume prediction system according to the present disclosure. As shown in FIG. 1, the traffic volume prediction system 1 according to the present disclosure includes a traffic volume prediction device 100, a server device 200, and a network N.

[0014] The traffic volume prediction device 100 is an information processing device that executes calculation processes and the like for predicting the traffic volume on various routes. The traffic volume prediction device 100 may be realized by, for example, a PC (Personal Computer), a WS (Work Station), or a computer having server functions. Note that the traffic volume prediction device 100 receives input of operation information from a user, for example, and performs processing based on the operation information.

[0015] The server device 200 is an information processing device that executes processes for realizing various arithmetic processes and functions. The server device 200 may be realized by, for example, a PC, a WS, or a computer having server functions. Note that the server device 200 performs processing based on information transmitted from the traffic volume prediction device 100 via the network N, for example.

[0016] The network N connects the traffic volume prediction device 100 and the server device 200 so that they can communicate with each other by wire or wirelessly. When the network N is wired, it may be realized by Ethernet (registered trademark) defined in IEEE802.3. When the network N is wireless, it may be realized by a wireless LAN (Local Area Network) defined in IEEE802.11.

[0017] (Configuration of Traffic Volume Prediction Device) (First Embodiment) Next, the traffic volume prediction device 100 according to the present disclosure will be described with reference to FIG. 2. FIG. 2 is a diagram showing a configuration example of the traffic volume prediction device according to the present disclosure. As shown in FIG. 2, the traffic volume prediction device 100 according to the present disclosure includes a communication unit 110, a storage unit 120, a control unit 130, an input unit 140, and a display unit 150. These configurations will be described in order below.

[0018] The communication unit 110 is responsible for transmitting and receiving various types of information, etc. with external devices via wired or wireless means. In the case of wired communication, it may be realized by a NIC (Network Interface Card) equipped with an interface such as a wired LAN terminal. In the case of wireless communication, it may be realized by a wireless LAN defined by IEEE802.11 or the like.

[0019] The storage unit 120 is a storage device that stores various types of information. The storage unit 120 includes a main storage device and an auxiliary storage device. The main storage device may be realized by semiconductor memory elements such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, etc. Also, the auxiliary storage device may be realized by, for example, a hard disk, SSD (Solid State Drive), optical disk, etc.

[0020] As shown in FIG. 2, the storage unit 120 includes a road network information storage unit 121, a railway network information storage unit 122, and a prediction program storage unit 123. Hereinafter, the information stored by these components will be described in order.

[0021] The road network information storage unit 121 stores road network information indicating information related to the road network. The road network information is information related to a road model represented by edges that simulate the smallest sections of roads delimited by intersections, etc. and nodes that simulate intersections, road width change points, etc. In this case, the following may be set for the edges: the distance of the delimited road, the speed limit of the automobiles traveling on it, the number of lanes of the road, the presence and number of right and left turn lanes, etc. Also, in this case, the position coordinates on the map may be set for the nodes. Note that the road network information may store information related to a road model that simulates an existing road network, or may be information related to a road model that simulates a non-existing road network (for example, a road to be newly constructed).

[0022] The railway network information storage unit 122 stores railway network information indicating information about the railway network. Here, the railway may include conventional railways, subways, suspended or straddle monorails, guided track new transportation systems (AGT: Automated Guideway Transit), light rail transit (LRT), cable cars, self-propelled ropeways, and the like. The railway network information is information about a railway model represented by edges indicating railway lines and nodes indicating stations where people board and alight. In this case, the following may be set for the edges: the distance between the node of the departure station and the node of the arrival station, the traveling direction of the railway, the running speed of the railway, the number of passengers that can be carried on the railway, and the like. The railway network may simulate the routes of existing railways. Also, for the nodes in this case, the position coordinates on the map of the station may be set. Note that the railway network information may be information about a railway model that simulates an existing railway network, or may be information about a railway model that simulates a non-existing railway network (for example, a newly constructed railway).

[0023] The prediction program storage unit 123 stores information about programs used for predicting traffic volumes in road networks and railway networks. The programming language of the programs used for predicting traffic volumes in road networks and railway networks is not limited, and for example, it may be described in a programming language such as FORTRAN or C language. Note that an executable file obtained by compiling a program described in a programming language may also be stored.

[0024] Next, returning to FIG. 2, the control unit 130 will be described. The control unit 130 is a controller that executes various arithmetic processes, processes for realizing functions, and the like. The control unit 130 is realized by executing various programs stored in the storage unit 120 using a RAM as a work area by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. Further, the control unit 130 may be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0025] As shown in FIG. 2, as functions realized by the execution of a program stored in the storage unit 120, the circuit configuration, etc., the control unit 130 includes an input value reception unit 131, a traffic volume prediction unit 132, a railway traffic volume prediction unit 133, a travel time error calculation unit 134, a current location - destination matrix calibration unit (OD matrix calibration unit) 135, and a railway current location - destination matrix calibration unit (railway OD matrix calibration unit) 136. Note that the control unit 130 may execute these processes by one CPU, or may include a plurality of CPUs and execute these processes in parallel by the plurality of CPUs. Hereinafter, these configurations will be described in sequence.

[0026] The input value receiving unit 131 receives input values used for traffic volume prediction from the user. For example, the input value receiving unit 131 receives an input value including at least one of a road origin-destination matrix (road OD matrix), a railway origin-destination matrix (railway OD matrix), an actual measured value of the driving speed of automobiles on the road, and an actual measured value of the number of users at each railway station. Note that the input value receiving unit 131 does not necessarily need to receive the road OD matrix and the railway OD matrix separately, and may receive an integrated matrix of the road OD matrix and the railway OD matrix as the OD matrix. Here, the method of representing the route from the origin to the destination in the OD matrix will be described with reference to FIG. 3. FIG. 3 is a diagram for explaining the method of representing the route from the origin to the destination in the OD matrix according to the present disclosure. As shown in FIG. 3, in the representation of the route from the origin to the destination in the OD matrix, the map is divided by a plurality of grids, identifiers are assigned to the divided grids, and the origin and the destination are represented by the identifiers. For example, as shown in FIG. 3, for the traffic volume moving from the origin A to the destination B, the notation Tab may be used.

[0027] Next, a specific example of the OD matrix according to the present disclosure will be described with reference to FIG. 4. FIG. 4 is a diagram showing an example of the OD matrix according to the present disclosure. As shown in FIG. 4, in the OD matrix according to the present disclosure, for the route T from the origin to the destination, the identifiers of the origin and the destination are represented by subscripts according to the above-described notation method. For example, in FIG. 4, for the traffic volume from the origin A to the destinations B, C, and D, the notations Tab, Tac, and Tad are respectively used, and numerical values indicating the number of movers on the routes from the origin A to the destinations B, C, and D are assigned. Then, with the rows as the origins and the columns as the destinations, the number of movers on these routes is arranged in a matrix and shown.

[0028] The traffic volume prediction unit 132 predicts the traffic volume of each route in the road network and the railway network by using the characteristic of increasing the travel time of the railway route according to the number of users, based on the current location - destination matrix, information on the road network, and information on the railway network. Specifically, it predicts the traffic volume of each route using StrUE (Strategic User Equilibrium). StrUE is a method of predicting the traffic volume of each route based on the premise that travelers select routes so that the expected value of the cost of travel, such as travel time, is minimized. First, the modeling of the route will be described with reference to FIG. 5.

[0029] FIG. 5 is a diagram showing an example of the modeling of a route according to the present disclosure. In FIG. 5, it is shown that a node N1 indicating the origin, a node N4 indicating the destination, and nodes N3 and N4 are arranged between them. Also, an edge E1 connecting node N1 to node N2, an edge E2 connecting node N1 to node N3, an edge E3 connecting node N2 to node N3, an edge E4 connecting node N2 to node N4, and an edge E5 connecting node N3 to node N4 are shown. In this way, each route is modeled by a directed graph.

[0030] Next, the prediction of the traffic volume using StrUE will be described with reference to FIG. 6. FIG. 6 is a flowchart showing the flow of the traffic volume prediction process according to the present disclosure. The traffic volume prediction process according to the present disclosure will be described along the flow shown in FIG. 6.

[0031] First, the ratio ξi of users selecting route i is set (step S101). Next, the traffic volume pa of each edge is calculated (step S102). Here, assuming d is the total traffic volume from the origin to the destination, the traffic volume p1 of edge 1 is represented by the following formula (1).

[0032]

Equation

[0033] Next, the cost ca of each edge is calculated (step S103). From the traffic volume p1 of edge 1, the travel time (a part of the cost) c1 of edge 1 is represented by the following formula (2) using the standard BPR (Bureau of Public Roads) function for road traffic.

[0034]

Equation

[0035] Next, the cost ca of each edge is calculated (step S104). Next, ξi is selected so as to minimize the total expected value of the cost (step S105).

[0036] That is, in traffic volume prediction using StrUE, the traffic volume of each route in the road network is determined so as to minimize the cost.

[0037] Such processing will be described using mathematical formulas. Let p be the ratio of passing through edge a, D be the total traffic volume, and pD be the traffic volume passing through edge a. Then, the objective function is represented by the following formula (3). Also, Φ(D) in formula (3) is the distribution of the traffic volume from the same origin to the same destination. In the following formula (3), it shows that the integral is taken from 0 to paD with respect to the traffic volume pD passing through edge a, which is a variable of the cost Ca(pD).

[0038]

Equation

[0039] Note that the cost Ca(pD) in the above formula (3) is represented by the following formula (4).

[0040]

Equation

[0041] t ais a function representing the travel time of edge a when the traffic volume is pD. VOT (Value of Time) is a value coefficient for time. σ a is the variation in the travel time of edge a. VOR (Value of Reliability) is a value coefficient for time variation. f a is the toll. t a is the travel time of edge a when the traffic volume is pD. When edge a is a road, it is derived by the BPR function, which is a function of general road traffic volume. The BPR function takes the travel time t ij0 from node i to node j under uncongested conditions, the traffic volume xij from node i to node j, the time traffic capacity cij from node i to node j, and parameters α and β, and is represented by the following formula (5).

[0042] [Number]

[0043] The railway traffic volume prediction unit 133 (which may be the traffic volume prediction unit 132) predicts the traffic volume of each route in the railway network based on the railway OD matrix and information on the railway network. As will be described later, the railway traffic volume prediction unit 133 does not necessarily need to be provided separately from the traffic volume prediction unit 132. It may receive the combined OD matrix of the road OD matrix and the railway OD matrix as an input value and have the traffic volume prediction unit 132 execute the processing of the railway traffic volume prediction unit 133. That is, the processing of the railway traffic volume prediction unit 133 to be described hereinafter may be implemented as the processing of the traffic volume prediction unit 132. The railway traffic volume prediction unit 133 predicts the traffic volume of each route in the railway network using the above-described StrUE in the same manner as the traffic volume prediction unit 132 described above. When the traffic volume of each route in the railway network is predicted by the railway traffic volume prediction unit 133 (which may be the traffic volume prediction unit 132), the travel time t ij (x ijIt calculates ). That is, the railway traffic volume prediction unit 133 (which may be the traffic volume prediction unit 132) calculates the travel time of a route on the railway as a function of the number of people boarding the railway vehicle, and calculates the travel time of the route on the railway.

[0044]

Number

[0045] In the above formula (6), t ij f is the travel time of the non-congested route from station i to station j, and x ij is the traffic volume of the route from station i to station j. μ j is the boarding capacity of station j.

[0046] Also, when the traffic volume of each route in the railway network is predicted, the railway traffic volume prediction unit 133 (which may be the traffic volume prediction unit 132) calculates the number of users at each station. The number of users at each station includes the number of entrants to each station and the number of exits from each station. The number of entrants to each station is calculated by summing up all the entry numbers from other specific stations to the target station based on the traffic volume of each route. The number of exits from each station is calculated by summing up all the exit numbers from the target station to other specific stations based on the traffic volume of each route.

[0047] Note that it is not essential to separately provide the traffic volume prediction unit 132 and the railway traffic volume prediction unit 133. The traffic volume prediction unit 132 may be made to execute the processing of the railway traffic volume prediction unit 133. That is, an OD matrix combining the road OD matrix and the railway OD matrix is received as an input value, and in the combined road and railway OD matrix, when the edge is a road, the cost of formula (5) is used, and when the edge is a railway, the cost of formula (6) is used to predict the traffic volume of each route. When the calibrated integrated OD matrix is obtained, since the traffic volumes of the road and the railway can be known by Str, the integrated OD matrix may be separated into a road OD matrix and a railway OD matrix.

[0048] The travel time error calculation unit 134 calculates a travel time error indicating the error between the travel time obtained from the actual driving speed of the vehicle and the travel time of each route obtained from the predicted traffic volume of each route in the road network. First, the travel time error calculation unit 134 calculates the actual travel time of each route based on the measured value of the actual driving speed of the vehicle on the road and the distance of each route. Then, the travel time error calculation unit 134 calculates the error by taking the difference between the travel time calculated by the traffic volume prediction unit 132 and the travel time obtained from the actual driving speed for each route in the road network.

[0049] The OD matrix correction unit 135 corrects the road current location destination matrix based on the travel time error. Specifically, the OD matrix correction unit 135 corrects the road OD matrix using a genetic algorithm. Here, the genetic algorithm will be described with reference to FIG. 7. FIG. 7 is a flowchart showing the processing flow of the genetic algorithm according to the present disclosure.

[0050] First, N individuals are randomly generated in the current generation (step S201). Next, the fitness of each individual in the current generation is calculated respectively (step S202). Next, it is determined whether the fitness (in the case where the individual is an OD matrix, it is the travel time error) satisfies the condition (step S203). If the fitness does not satisfy the condition (in the case of the OD matrix, for example, the travel time error is less than or equal to a predetermined threshold) (step S203: No), two individuals are selected from the current generation (step S204). Next, an individual of the next generation is generated by crossing the two selected individuals (step S205). The crossover process in step 205 does not necessarily have to be executed, and not all individuals in the next generation have undergone crossover and mutation. The individuals in the current generation may be left as they are. The crossover process will be described in detail later. Next, an individual of the next generation is generated by mutation (step S206). Next, it is determined whether the number of individuals in the next generation has reached N (step S207). If the number of individuals in the next generation has reached N (step S207: Yes), the process returns to step S202 and the subsequent processing is executed. If the number of individuals in the next generation has not reached N (step S207: No), the process returns to step S204 and the subsequent processing is executed.

[0051] Next, the crossover process in the genetic algorithm in step S205 will be described with reference to FIG. 8. FIG. 8 is a diagram for explaining the crossover process in the genetic algorithm processing according to the present disclosure. As shown in FIG. 8, in the crossover process, genes of two individuals selected from N individuals in the current generation (parent generation) are compared with the errors in the travel times of the paths, which are the individual values of the OD matrix, of the two individuals, and the genes with the smaller error are selected to generate the genes of the next generation (offspring generation). In FIG. 8, the error in the travel time of the path OD from the current location A to the destination B of the individual Parent1 in the current generation (parent generation) is 9%, while the error in the travel time of the path OD from the current location A to the destination B of the individual Parent2 in the current generation (parent generation) is 1%. Therefore, it is shown that the genes of the individual Parent2 are selected as the genes of the path OD from the current location A to the destination B of the next generation (offspring generation). In this way, for each path, the genes with the smaller error in the travel time between the two individuals are selected to generate the genes of the next generation (offspring generation). That is, the OD matrix calibration unit 135 generates the individuals of the offspring generation by selecting the genes of the individuals of the offspring generation from the genes of the individuals of the parent generation based on the travel time error using the genetic algorithm, and corrects the road current location destination matrix by selecting the individuals with the travel time error below a predetermined threshold as the corrected road current location destination matrix. AB The error in the travel time of the path OD from the current location A to the destination B of the individual Parent1 in the current generation (parent generation) is 9%, while the error in the travel time of the path OD from the current location A to the destination B of the individual Parent2 in the current generation (parent generation) is 1%. Therefore, it is shown that the genes of the individual Parent2 are selected as the genes of the path OD from the current location A to the destination B of the next generation (offspring generation). In this way, for each path, the genes with the smaller error in the travel time between the two individuals are selected to generate the genes of the next generation (offspring generation). That is, the OD matrix calibration unit 135 generates the individuals of the offspring generation by selecting the genes of the individuals of the offspring generation from the genes of the individuals of the parent generation based on the travel time error using the genetic algorithm, and corrects the road current location destination matrix by selecting the individuals with the travel time error below a predetermined threshold as the corrected road current location destination matrix. AB The error in the travel time of the path OD from the current location A to the destination B of the individual Parent1 in the current generation (parent generation) is 9%, while the error in the travel time of the path OD from the current location A to the destination B of the individual Parent2 in the current generation (parent generation) is 1%. Therefore, it is shown that the genes of the individual Parent2 are selected as the genes of the path OD from the current location A to the destination B of the next generation (offspring generation). In this way, for each path, the genes with the smaller error in the travel time between the two individuals are selected to generate the genes of the next generation (offspring generation). That is, the OD matrix calibration unit 135 generates the individuals of the offspring generation by selecting the genes of the individuals of the offspring generation from the genes of the individuals of the parent generation based on the travel time error using the genetic algorithm, and corrects the road current location destination matrix by selecting the individuals with the travel time error below a predetermined threshold as the corrected road current location destination matrix. AB The error in the travel time of the path OD from the current location A to the destination B of the individual Parent1 in the current generation (parent generation) is 9%, while the error in the travel time of the path OD from the current location A to the destination B of the individual Parent2 in the current generation (parent generation) is 1%. Therefore, it is shown that the genes of the individual Parent2 are selected as the genes of the path OD from the current location A to the destination B of the next generation (offspring generation). In this way, for each path, the genes with the smaller error in the travel time between the two individuals are selected to generate the genes of the next generation (offspring generation). That is, the OD matrix calibration unit 135 generates the individuals of the offspring generation by selecting the genes of the individuals of the offspring generation from the genes of the individuals of the parent generation based on the travel time error using the genetic algorithm, and corrects the road current location destination matrix by selecting the individuals with the travel time error below a predetermined threshold as the corrected road current location destination matrix.

[0052] The railway OD matrix calibration unit (railway current location destination matrix calibration unit) 136 generates a railway current location destination matrix by extracting movement information on the railway network from the current location destination matrix, calculates an error in the number of station users indicating the error between the number of station users based on the predicted value of traffic volume for each route in the railway network and the actual measured value of the number of station users on the railway, and calibrates the railway current location destination matrix based on the calculated error in the number of station users. Specifically, the railway OD matrix calibration unit 136 corrects the railway OD matrix by the method described below. First, the railway OD matrix calibration unit 136 calculates the error between the actual number of station users on the railway and the number of station users calculated by the railway traffic volume prediction unit 133. Next, the correction amount TD of the number of visitors to the station is calculated using the following equations (7), (8), (9), and (10). onij , the correction amount TD for the number of participants at the station offij Calculate.

[0053] OP ij is the proportion of passengers moving from station i to station j based on the unadjusted railway OD matrix, TD ij If we define the number of people moving from station i to station j, then OP ij is required.

[0054]

number

[0055] Then, if EGi is the error in the number of visitors to station i, the correction amount of the railway OD matrix based on the error in the number of visitors to station i is calculated by the following equation (8): TD onij is required.

[0056]

number

[0057] Also, D.P. ij is the ratio of people moving from station i to people moving to station j based on the unadjusted railway OD matrix. As mentioned above, TD ij If we define the number of people moving from station i to station j, then the DPij is required.

[0058]

Number

[0059] And, assuming EGj is the error in the number of departing passengers at station j, the correction amount TD of the railway OD matrix based on the error in the number of departing passengers at station j is obtained by the following formula (10). offij is required.

[0060]

Number

[0061] And, assuming the correction amount for the value in the i-th row and j-th column of the railway OD matrix is TD ij then TD ij is obtained by the following formula (11).

[0062]

Number

[0063] The method for calculating the correction amount of the railway OD matrix described above will be described with reference to FIG. 9. FIG. 9 is a diagram for explaining the correction of the railway OD matrix according to the present disclosure. In the above-described formulas (7) and (8), as shown in FIG. 9, the ratio between the passengers departing from station A and arriving at station B and the passengers departing from station A and arriving at station C is maintained, and the correction amount is obtained so that the measured value of the number of arriving passengers at station A matches. Similarly, in the above-described formulas (9) and (10), as shown in FIG. 9, the ratio between the passengers departing from station A and arriving at station B and the passengers departing from station C and arriving at station B is maintained, and the correction amount is obtained so that the measured value of the number of departing passengers at station B matches.

[0064] The input unit 140 receives various operation information from the user. For example, the input unit 140 may receive various operations from the operator via the display surface (e.g., the display unit 150) by a touch panel. Also, the input unit 140 may receive various operations from the user by various buttons, a keyboard, or a mouse.

[0065] The display unit 150 displays various information. The display unit 150 may display, for example, a GUI (Graphical User Interface) for receiving operations related to various processes from the user, the results of traffic volume prediction simulations, etc. The display unit 150 may be realized by a liquid crystal display, an organic EL (Electro Luminescence) display, a micro LED (Light Emitting Diode) display, etc. Also, the display unit 150 may be a touch panel of various methods such as a capacitance method.

[0066] (Regarding the traffic volume prediction method) Next, the processing of the first embodiment of the traffic volume prediction device according to the present disclosure will be described with reference to FIG. 10. FIG. 10 is a flowchart showing the flow of the processing of the first embodiment of the traffic volume prediction device according to the present disclosure. The processing of the first embodiment of the traffic volume prediction device according to the present disclosure will be described along the flow shown in FIG. 10.

[0067] First, the traffic volume prediction device 100 generates an initial value of the OD matrix (S301). By this processing, an OD matrix (O101) with the initial value set is generated. Next, based on the OD matrix (O101), the road network (I101), and the railway network (I102) before the opening of the subway, an alternating traffic volume prediction process is executed (step S302). By this processing, the road traffic volume (O102) and the railway traffic volume (O103) are output. Next, based on the road traffic volume (O102), the travel time is calculated (step S303). In calculating the travel time based on the road traffic volume, the travel time may be calculated by the above-described BPR function. Also, based on the railway traffic volume (O103), the number of users at each station is calculated (step S304).

[0068] According to this, based on the travel time based on the predicted value of road traffic volume and the error of the travel time obtained from the measured values of the actual driving speeds of each route of the automobile, an appropriately corrected road current location destination matrix can be generated. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict traffic volume by using an automatically calibrated OD matrix.

[0069] (Configuration of traffic volume prediction device) (Second Embodiment) Next, a second embodiment of the traffic volume prediction device 100 according to the present disclosure will be described. FIG. 11 is a diagram showing a configuration example of the second embodiment of the traffic volume prediction device according to the present disclosure. As shown in FIG. 11, the traffic volume prediction device 100 according to the present disclosure includes a communication unit 110, a storage unit 120, a control unit 130, an input unit 140, and a display unit 150. Among the configurations of the second embodiment of the traffic volume prediction device 100, the communication unit 110, the storage unit 120, the input unit 140, and the display unit 150 are the same as the communication unit 110, the storage unit 120, the input unit 140, and the display unit 150 of the first embodiment of the traffic volume prediction device 100, respectively.

[0070] Among the configurations of the second embodiment of the traffic volume prediction device 100, the configuration different from that of the first embodiment of the traffic volume prediction device 100 is the control unit 130. As shown in FIG. 11, the control unit 130 of the second embodiment of the traffic volume prediction device 100 has an additional similar zone correction unit 137 with respect to the control unit 130 of the first embodiment of the traffic volume prediction device 100. Therefore, in the description of the configuration of the second embodiment of the traffic volume prediction device 100, the description of the same configuration as that of the first embodiment of the traffic volume prediction device 100 described above is omitted, and the similar zone correction unit 137, which is a configuration different from that of the first embodiment of the traffic volume prediction device 100, will be described.

[0071] The similar zone correction unit 137 extracts similar zones in the existing railway network and road network for the new railway construction plan zone, and corrects the railway OD matrix based on the ratio of railway users to automobile users in the extracted similar zones. First, based on the information of the road OD matrix and the railway OD matrix, the similar zone correction unit 137 uses the number of outflows per unit population (morning), the number of outflows per unit population (daytime), the number of inflows per unit population (evening), and the number of inflows per unit population (night) as feature quantities, and extracts zones similar to the zone where the station to be predicted is installed based on the least squares error. The extraction of similar zones will be described with reference to FIG. 12. FIG. 12 is a diagram for explaining the extraction of similar zones according to the present disclosure. For example, assume that new railway construction is planned in the area shown by Metro Zone 1 in FIG. 12. In this case, the similar zone correction unit 137 extracts similar zones based on the least squares error using the number of outflows per unit population (morning), the number of outflows per unit population (daytime), the number of inflows per unit population (evening), and the number of inflows per unit population (night) as feature quantities based on the calculation results of the OD matrix. In FIG. 12, it is shown that the area south of Metro Zone 1 has been extracted as a zone similar to Metro Zone 1.

[0072] Also, in the extraction of similar zones, the similar zone correction unit 137 extracts similar zones based on the number of outflows per unit population and the number of inflows per unit population in a specific time period. Note that the feature quantities for extracting similar zones may be changed based on the population difference between daytime and nighttime. For example, if the population during the day is less than the population at night, the number of outflows per unit population (morning), the number of outflows per unit population (daytime), the number of inflows per unit population (evening), and the number of inflows per unit population (night) may be used as feature quantities as described above. If the population during the day is more than the population at night, the number of inflows per unit population (morning), the number of inflows per unit population (daytime), the number of outflows per unit population (evening), and the number of outflows per unit population (night) may be used as feature quantities. Thereby, the prediction accuracy outside the suburban area can be improved.

[0073] When the similar zone is extracted, the similar zone correction unit 137 predicts the OD matrix of train users based on the number of movers in the metro zone, the train utilization rate in the similar zone, and the number of movers from the metro zone to the zones to which other train stations belong. Specifically, as shown in FIG. 13, the correction amount of the railway OD matrix is calculated. FIG. 13 is a diagram showing the calculation formula of the correction amount of the railway OD matrix based on the similar zone according to the present disclosure. The number of railway outflows from metro zone i to other station zones j is Demand zoneij Production is the product of the total number of outflows and the ratio of the number of train-using outflows (similar zone). i Let proportion be the ratio of the movement to zone j among the movements from metro zone i to other station zones. Then, according to the formula shown in the upper part of FIG. 13, the number of railway outflows from metro zone i to other station zones j can be calculated. Also, the number of railway inflows from other station zones j to metro zone i is Demand i,j Attraction is the product of the total number of inflows and the ratio of the number of train-using inflows (similar zone). j,zonei Let proportion be the ratio of the movement from zone j among the movements from other station zones to metro zone i. Then, according to the formula shown in the lower part of FIG. 13, the number of railway inflows from other station zones j to metro zone i can be calculated. The similar zone correction unit 137 uses the calculated number of railway outflows from metro zone i to other station zones j as Demand i and the number of railway inflows from other station zones j to metro zone i as Demand j,i to correct the railway OD matrix as the correction amount of the railway OD matrix. zoneij and correct the railway OD matrix with the calculated number of railway inflows from other station zones j to metro zone i as Demand j,zonei as the correction amount of the railway OD matrix.

[0074] (Regarding the traffic volume prediction method) Next, the processing of the second embodiment of the traffic volume prediction device according to the present disclosure will be described with reference to FIG. 14. FIG. 14 is a flowchart showing the flow of the processing of the second embodiment of the traffic volume prediction device according to the present disclosure. The processing of the second embodiment of the traffic volume prediction device according to the present disclosure will be described along the flow shown in FIG. 14. Among the processing of the second embodiment of the traffic volume prediction device according to the present disclosure shown in FIG. 14, steps S402, S403, and S404 are the same as steps S302, S303, and S304 of the processing of the first embodiment of the traffic volume prediction device according to the present disclosure shown in FIG. 10, so the description thereof will be omitted.

[0075] First, the corrected OD matrix (I201) is corrected based on the movement amounts in similar zones (step S401). As a result, a corrected OD matrix (O201) is generated. For the next steps S402, S403, and S404, since they are the same as steps S302, S303, and S304 of the processing of the first embodiment of the traffic volume prediction device according to the present disclosure shown in FIG. 10, the description thereof will be omitted. Next, the travel time after the opening of the subway for each route is calculated from the road travel speed after the opening of the subway. Then, the error between the travel time when using the road calculated in step S403 and the travel time after the opening of the subway for each route from the road travel speed after the opening of the subway is calculated (step S405). As a result, the error (O206) in the travel time on roads other than the subway is calculated. Also, the error between the number of users of each station calculated in step S404 and the number of users of each station after the opening of the subway is calculated (step S405). As a result, the error (O207) in the number of users of each station on railways including the subway is calculated. Thereafter, based on the calculated errors, the road OD matrix and the railway OD matrix are corrected as described above, so that the road OD matrix and the railway OD matrix after the opening of the subway can be generated.

[0076] According to the processing of the second embodiment of the traffic volume prediction device according to the present disclosure described above, the number of users of a newly constructed railway can be appropriately predicted. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict the traffic volume using an automatically calibrated OD matrix.

[0077] (Configuration of Server Device) Next, the configuration of the server device 200 according to the present disclosure will be described. The server device 200 according to the present disclosure may include a part or all of the configurations of the first and second embodiments of the traffic volume prediction device 100 according to the present disclosure, and may undertake a part or all of the processes of the first and second embodiments of the traffic volume prediction device 100 according to the present disclosure. For example, if a storage area is required for the road network information storage unit 121 or the railway network information storage unit 122, a storage medium with a large storage capacity may be provided in the server device 200, and the first and second embodiments of the traffic volume prediction device 100 may access the server device 200 via the network N to obtain the road network information and railway network information of the parts necessary for traffic volume prediction calculation. Further, in the first and second embodiments of the traffic volume prediction device 100, for example, only the processes that require arithmetic processing capabilities such as traffic volume prediction calculation processing may be executed by the server device 200.

[0078] As a result, traffic volume prediction calculation can be executed without using an information processing device with high performance such as arithmetic processing capabilities and storage capacity as the traffic volume prediction device 100. Also, the same road network information and railway network information can be shared among multiple traffic volume prediction devices 100, and traffic volume prediction calculation can be executed in parallel by multiple traffic volume prediction devices 100.

[0079] (Hardware Configuration) The traffic volume prediction device 100 and the server device 200 according to the above-described embodiments are realized by a computer 1000 having a configuration as shown in FIG. 15, for example. FIG. 15 is a hardware configuration diagram showing an example of a computer that realizes the functions of the traffic volume prediction device and the server device according to the present disclosure. The computer 1000 has a form in which an output device 1010, an input device 1020 are connected, and an arithmetic device 1030, a primary storage device 1040, a secondary storage device 1050, an output IF (Interface) 1060, an input IF 1070, and a network IF 1080 are connected by a bus 1090.

[0080] The arithmetic unit 1030 operates based on programs stored in the primary storage device 1040 and the secondary storage device 1050, programs read from the input device 1020, etc., and executes various processes. The primary storage device 1040 is a memory device that primarily stores data used by the arithmetic unit 1030 for various operations, such as a RAM. Also, the secondary storage device 1050 is a storage device that stores data used by the arithmetic unit 1030 for various operations and various databases, and is realized by a ROM (Read Only Memory), an HDD (Hard Disk Drive), a flash memory, etc.

[0081] The output IF 1060 is an interface for transmitting information to be output to an output device 1010 that outputs various types of information, such as a monitor and a printer, and is realized by a connector of a standard such as USB (Universal Serial Bus), DVI (Digital Visual Interface), HDMI (registered trademark) (High Definition Multimedia Interface). Also, the input IF 1070 is an interface for receiving information from various input devices 1020, such as a mouse, a keyboard, and a scanner, and is realized by, for example, USB or the like.

[0082] Note that the input device 1020 may be a device that reads information from an optical recording medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Also, the input device 1020 may be an external storage medium such as a USB memory.

[0083] The network IF 1080 receives data from other devices via the network N and sends it to the arithmetic unit 1030, and also sends data generated by the arithmetic unit 1030 via the network N to other devices.

[0084] The arithmetic unit 1030 controls the output device 1010 and the input device 1020 via the output IF 1060 and the input IF 1070. For example, the arithmetic unit 1030 loads a program from the input device 1020 or the secondary storage device 1050 onto the primary storage device 1040 and executes the loaded program.

[0085] For example, when the computer 1000 functions as the traffic volume prediction device 100, the arithmetic unit 1030 of the computer 1000 realizes the function of the control unit 130 of the traffic volume prediction device 100 by executing the program loaded on the primary storage device 1040.

[0086] (Configuration and Effect) The traffic volume prediction device 100 according to the present disclosure includes an input value reception unit 131 that receives an input value including at least one of a current location - destination matrix, information on a road network, information on a railway network, a measured value of the running speed of automobiles on a road, and a measured value of the number of users at each station on a railway, a traffic volume prediction unit 132 that predicts the traffic volume of each route in the road network and the railway network using a characteristic of increasing the travel time of a railway route according to the number of users based on the current location - destination matrix, information on the road network, and information on the railway network, a travel time error calculation unit 134 that calculates a travel time error indicating the error between the travel time obtained from the actual running speed of an automobile and the travel time of each route obtained from the predicted value of the traffic volume of each route in the road network, and a current location - destination matrix correction unit 135 that corrects the current location - destination matrix based on the travel time error.

[0087] According to this configuration, it is possible to generate an appropriately corrected current location - destination matrix based on the error between the travel time based on the predicted value of the traffic volume of each route in the road network and the travel time obtained from the measured value of the actual running speed of an automobile. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict the traffic volume of each route using the automatically calibrated OD matrix.

[0088] The current location - destination matrix correction unit 135 of the traffic volume prediction device 100 according to the present disclosure has each element of the current location - destination matrix as a gene, and selects a gene with a small error in the travel time of the route corresponding to the element to preferentially leave it in the offspring generation, and corrects the road current location - destination matrix by selecting it as the current location - destination matrix using a genetic algorithm.

[0089] According to this configuration, a current location - destination matrix can be generated using a genetic algorithm to correct the current location - destination matrix. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict the traffic volume of each route using the automatically corrected OD matrix.

[0090] The traffic volume prediction device 100 according to the present disclosure generates a railway current location - destination matrix by extracting movement information by the railway network from the current location - destination matrix, calculates each station user number error indicating the error between the predicted value of the traffic volume of each route in the railway network and the measured value of the number of users of each station in the railway, and further includes a railway current location - destination matrix correction unit (railway OD matrix correction unit) 136 that corrects the railway current location - destination matrix based on the calculated each station user number error.

[0091] According to this configuration, based on the measured value of the actual number of users of each station, the error in the number of entrants and the number of exits of each station can be calculated, and the railway current location - destination matrix can be corrected based on the calculated error. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict the traffic volume of each route using the automatically corrected OD matrix.

[0092] The traffic volume prediction device 100 according to the present disclosure further includes a similar zone correction unit 137 that corrects the railway current location - destination matrix based on the ratio of railway users and automobile users in the similar zones extracted for the new railway construction plan zone in the railway, from the existing railway network and road network.

[0093] According to this configuration, the number of users of the railway can be predicted by the new construction of the railway. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict the traffic volume of each route using an automatically calibrated OD matrix.

[0094] The similar zone correction unit 137 of the traffic volume prediction device 100 according to the present disclosure extracts similar zones based on the number of outflows per unit population and the number of inflows per unit population in a specific time period.

[0095] According to this configuration, similar zones can be extracted based on the number of outflows per unit population and the number of inflows per unit population in a specific time period, so that similar zones can be appropriately extracted. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict the traffic volume of each route using an automatically calibrated OD matrix.

[0096] The traffic volume prediction method according to the present disclosure includes a step of predicting the traffic volume of each route in a road network and a railway network using the characteristic of increasing the travel time of a railway route according to the number of users based on a current location - destination matrix, information on the road network, information on the railway network, the current location - destination matrix, and the information on the road network and the railway network; a step of calculating the error between the travel time obtained from the actual driving speed and the travel time of each route obtained from the predicted value of the road traffic volume; and a step of correcting the current location - destination matrix based on the error.

[0097] According to this configuration, it is possible to generate an appropriately corrected current location - destination matrix based on the error between the travel time based on the predicted value of the traffic volume of each route in the road network and the travel time obtained from the measured value of the actual driving speed of an automobile. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict the traffic volume of each route using an automatically calibrated OD matrix.

[0098] The program according to the present disclosure causes a computer to execute steps of predicting the traffic volume of each route in a road network and a railway network by using the characteristic of increasing the travel time of a railway route according to the number of users, based on a current location - destination matrix, information on the road network, and information on the railway network; calculating an error between the travel time obtained from the actual driving speed and the travel time of each route obtained from the predicted value of the road traffic volume; and correcting the current location - destination matrix based on the error.

[0099] According to this configuration, an appropriately corrected current location - destination matrix can be generated based on the travel time based on the predicted value of the traffic volume of each route in the road network and the error between the travel time obtained from the measured value of the actual driving speed of an automobile. Therefore, it is possible to provide a traffic volume prediction device 100 that can appropriately predict the traffic volume of each route by using the automatically calibrated OD matrix.

[0100] As described above, although the embodiments of the present disclosure have been described, the embodiments are not limited by the content of this embodiment. Further, the above - described constituent elements include those that can be easily assumed by those skilled in the art, those that are substantially the same, and those within the so - called equivalent range. Furthermore, the above - described constituent elements can be combined as appropriate. Furthermore, various omissions, substitutions, or changes of the constituent elements can be made without departing from the gist of the above - described embodiments.

Explanation of Reference Numerals

[0101] 100 Traffic volume prediction device 110 Communication unit 120 Storage unit 121 Road network information storage unit 122 Railway network information storage unit 123 Prediction program storage unit 130 Control unit 131 Input value reception unit 132 Traffic volume prediction unit 133 Railway traffic volume prediction unit 134 Travel time error calculation unit 135 OD Matrix Calibration Unit 136 Railway OD Matrix Calibration Unit 137 Similar Zone Correction Unit 140 Input Unit 150 Display Unit 200 Server Device N Network

Claims

1. An input value reception unit that receives an input value including at least one of a current location - destination matrix, information on a road network, information on a railway network, a measured value of the driving speed of automobiles on a road, and a measured value of the number of users at each station on a railway; A traffic volume prediction unit that predicts the traffic volume of each route in the road network and the railway network by using a characteristic of increasing the travel time of a railway route according to the number of users based on the current location - destination matrix, information on the road network, and information on the railway network; A travel time error calculation unit that calculates a travel time error indicating the error between the travel time obtained from the measured value of the driving speed of automobiles on a road and the travel time of each route obtained from the predicted value of the traffic volume of each route in the road network; A current location - destination matrix correction unit that corrects the current location - destination matrix based on the travel time error, and comprises a traffic volume prediction device. Traffic volume prediction device.

2. The current location - destination matrix correction unit corrects the current location - destination matrix by using a genetic algorithm characterized by having each element of the current location - destination matrix as a gene and preferentially leaving in the offspring generation a gene with a small error in the travel time of the route corresponding to the element. The traffic volume prediction device according to Claim 1.

3. Generates a railway current location - destination matrix by extracting movement information by the railway network from the current location - destination matrix, calculates an error in the number of users at each station indicating the error between the number of users at each station based on the predicted value of the traffic volume of each route in the railway network and the measured value of the number of users at each station on the railway, and further comprises a railway current location - destination matrix correction unit that corrects the railway current location - destination matrix based on the calculated error in the number of users at each station. The traffic volume prediction device according to Claim 1 or 2.

4. Further comprises a similar zone correction unit that extracts similar zones in the existing railway network and road network for a newly planned construction zone of a railway, and corrects the railway current location - destination matrix based on the ratio of railway users and automobile users in the extracted similar zones. The traffic volume prediction device according to Claim 1 or 2.

5. Further comprises a similar zone correction unit that extracts similar zones in the existing railway network and road network for a newly planned construction zone of a railway, and corrects the railway current location - destination matrix based on the ratio of railway users and automobile users in the extracted similar zones. The traffic volume prediction device according to Claim 3.

6. The similar zone correction unit extracts similar zones based on the number of outflows per unit population and the number of inflows per unit population in a specific time period. The traffic volume prediction device according to claim 4.

7. The similar zone correction unit extracts similar zones based on the number of outflows per unit population and the number of inflows per unit population in a specific time period. The traffic volume prediction device according to claim 5.

8. A step of receiving an input value including at least one of a current location - destination matrix, information on a road network, information on a railway network, an actual measured value of the traveling speed of automobiles on a road, and an actual measured value of the number of users at each railway station; A step of predicting the traffic volume of each route in the road network and the railway network by using the characteristic of increasing the travel time of railway routes according to the number of users, based on the current location - destination matrix, information on the road network, and information on the railway network; A step of calculating a travel time error indicating the error between the travel time obtained from the actual measured value of the traveling speed of automobiles on a road and the travel time of each route obtained from the predicted value of the traffic volume of each route in the road network; A step of correcting the current location - destination matrix based on the travel time error, and including: A traffic volume prediction method.

9. A step of receiving an input value including at least one of a current location - destination matrix, information on a road network, information on a railway network, an actual measured value of the traveling speed of automobiles on a road, and an actual measured value of the number of users at each railway station; A step of predicting the traffic volume of each route in the road network and the railway network by using the characteristic of increasing the travel time of railway routes according to the number of users, based on the current location - destination matrix, information on the road network, and information on the railway network; A step of calculating a travel time error indicating the error between the travel time obtained from the actual measured value of the traveling speed of automobiles on a road and the travel time of each route obtained from the predicted value of the traffic volume of each route in the road network; A step of correcting the current location - destination matrix based on the travel time error; A program for causing a computer to execute.

Images