JP5213403B2 - Charge calculation network generation apparatus and charge calculation apparatus using the charge calculation network - Google Patents

Description

  The present invention relates to a fee calculation network generating device for calculating the lowest charge from the start point to the end point, and a fee calculation device using this fee calculation network, and in particular, between two points straddling a plurality of systems having different fee structures. The present invention relates to a charge calculation network generation apparatus suitable for the lowest charge calculation and a charge calculation apparatus using the charge calculation network.

  As an example of calculating the charge from the start point to the end point, for example, there is a fare calculation for a railway or the like. The fare calculation for railways, etc. is calculated by calculating the distance between the departure station and the arrival station from the route data indicating the connection relationship between the stations and the inter-station distance data when the information of the departure station and the arrival station is given. From the fare conversion table data, the calculated distance is converted into a fare, and the fare between the departure station and the arrival station is calculated (for example, Patent Document 1 and Patent Document 2).

By the way, in the case of fare calculation for railways, etc., the fare calculation between two given stations must be calculated assuming that the route with the lowest fare is used among the available routes between the two stations. . As such a cheapest fare calculation method, there is conventionally a method of calculating the cheapest fare using a physical network as shown in FIG. 29 is a network in which a station is a node, a track is an arc, and a node connection cost is a distance. In this physical network, a shortest path search algorithm is used to search for a shortest path. Is converted into a fare using distance-fare conversion table data to obtain the lowest fare.
JP-A-5-258137 JP 2000-11211 A

  However, for example, if the departure station and arrival station span multiple companies with different fee systems (different fees are set even for the same distance), a discount system applied to transfer between companies, etc. Therefore, there is no guarantee that the shortest distance route will be the cheapest fare. For this reason, when calculating the lowest fare using the physical network of FIG. 29, all the available routes between the two stations are obtained, the distances of all the routes are calculated, the calculated distances are converted into fare, A calculation process is performed to select the lowest fare from among them. Therefore, in a large-scale railway network such as the Tokyo metropolitan area, the number of search routes is enormous, and the calculation amount is enormous because it is necessary to perform distance calculation and distance-to-fare conversion for all search routes. There is a problem that it takes time to calculate the cheapest fare.

  The present invention has been made paying attention to the above problems, and aims to provide a fee calculation network that can reduce the number of search routes in the cheapest fee calculation processing and can reduce the calculation processing time and the calculation processing time. To do. In addition, a fee calculation apparatus using the fee calculation network is provided.

For this reason, the fee calculation network generation device according to the present invention described in claim 1 includes fee data defining a fee for each system of a plurality of systems having different fee systems, and connection relationship data indicating a connection relationship between the systems. The data storage unit for storing the data , and each point in each system of the plurality of systems as a node, each node is directly connected by an arc as necessary, and the cost of connection between the nodes connected by the arc is the fee Generate network data with charges based on data, use network data for each system as sub-network data, and generate connection arc data for connecting nodes of different sub-network data based on the connection relation data was constructed and a network data generator for the connection arc data and the sub-network data and billing network data And wherein the door.

According to a second aspect of the present invention, the network data generation unit directly connects all the nodes in each of the plurality of systems with arcs, and uses the connection cost between the nodes connected with the arcs as the fee data. the network data which was based fees, and configured to generate as the billing network data.
In such a configuration, since the cost of connection between nodes connected by arcs is used as a fee, the cheapest fee can be directly calculated by applying the shortest path search algorithm, and the amount of calculation can be reduced.

According to claim 3, the pre-kine Tsu network data generation unit, and configured to generate the billing network data costs of connection to the rates zero between the nodes before connection with Kia over click.

  In such a configuration, the cheapest price between two points across systems with different charge systems can be calculated by applying the shortest route search algorithm.

In the invention of claim 4, the network data generation unit regards each sub-network in the fee calculation network data as one upper node, and sets one upper arc between the upper nodes based on the connection relation data. To create a higher level network data with a connection cost between the higher level nodes connected by this higher level arc as 1, and layer the network data to start and end points across multiple systems with different fee systems When the above information is input, the cost value of the upper arc from the upper node including the start point of the upper network to the upper node including the end point is less than or equal to the predetermined cost value based on a predetermined prohibition condition in charge calculation. In addition, the contact pattern having a connection relationship such that the upper nodes do not overlap in the route from the start point to the end point Subordinate network data is selected, and subnetwork corresponding to each upper node in the selected contact pattern data and subnetwork data connecting the subnetworks are generated from the subnetwork data connecting the subnetworks. The lower network data is used as fee calculation network data.

In Claim 5, each said subnetwork data is comprised only by the node data connected by the said connection arc, and the arc data which connect between these nodes, The simple network data which consists of these subnetwork data and the said connection arc data is comprised. When the information of the starting point and the ending point that are generated and straddles a plurality of systems of different fee systems is inputted, the node data of the starting point and the ending point and the arc data connected to each of the starting point and the ending point are added to the simple network data. In addition, a simple fee calculation network data is generated.

The simple network comprising the sub-network data and the connection arc data, wherein each sub-network data includes only node data connected by the connection arc and arc data connecting the nodes. Generating simple data, and adding simple arc calculation network data in which each arc data connected to each of the start point and end point of each node data and start point and end point across a plurality of systems of different charge systems is added to the simple network data, A configuration may be adopted in which all combinations of the start point and the end point are generated in advance, and when the information on the start point and the end point is input, the simple fee calculation network data corresponding to the input information is selected.
In such a configuration, since a branch that cannot be reached to the end point is deleted, an unnecessary branch is not searched.

In the invention of claim 7, the charge calculation network data includes node data obtained by adding a dummy node for each node of each subnetwork, and a connection source or a connection destination of the arc connecting the nodes in each subnetwork. It has sub-network data composed of arc data with one end as a dummy node, and when the arc connection source is a dummy node, the connection destination of the connection arc that connects nodes in different sub-networks A connection arc data having a dummy node, and a connection arc data having a connection source of the connection arc connecting the nodes of different sub-networks as a dummy node when the connection destination of the arc is a dummy node; did.
With such a configuration, it is possible to prevent the calculation of a toll between two points straddling between unconnected systems, and to prevent a stopover route on the way.

In the invention of claim 8, the charge calculation network data includes discount arc data indicating a discount path that directly connects two predetermined points with charge discounts between different sub-networks, and the predetermined predetermined data Based on fee discount data between two points, the connection cost between nodes connected by the discount arc is set as a discounted fee.
With such a configuration, it becomes possible to easily calculate a charge between two points with a charge discount between systems having different charge systems.

The fee calculation apparatus according to the present invention described in claim 9 is an input unit for inputting information of a start point and an end point, and a fee calculation network generated by the charge calculation network generation device according to any one of claims 1 to 8. a network data storage for storing data, when the information of the start point and the end point is inputted, to construct a billing network based on billing network data of the network data storage unit, from the start of the billing network It is characterized by comprising a cheapest price calculation unit for calculating the lowest price to the end point and an output unit for outputting the calculated lowest price result.

  In such a configuration, when the information of the start point and the end point is input from the input unit, the cheapest charge calculation unit constructs a charge calculation network based on the charge calculation network data stored in the network data storage unit, and the charge calculation network The lowest price from the start point to the end point is calculated, and the lowest price result calculated from the output unit is output.

  As in claim 10, when the charge calculation network data is data generated by the charge calculation network generation device according to any one of claims 1 to 7, the cheapest charge calculation unit, Based on the minimum cost search unit that searches for a route that minimizes the cost of the arc from the start point to the end point in the fee calculation network, and the fee discount data that indicates the route to which the charge discount is applied and the discount amount, and the end point A maximum discount amount calculation unit that calculates the maximum value of the discount amount applied to the route up to and including the maximum discount amount calculated by the maximum discount amount calculation unit to the minimum cost fee searched by the minimum cost search unit Based on the search threshold setting unit for setting a search threshold and the fee calculation network, all routes where the cost of the arc is equal to or less than the search threshold are searched, and each route When there is a route to which a discount is applied to the searched route based on the fee discount data, a discount for the route calculated by the first fee calculation unit when a fee is calculated based on the fee discount data A second fee calculation unit that calculates a fee after applying the discount by subtracting the discount amount applied to the route from the pre-application fee, the fee calculated by the first fee calculation unit, and the second fee calculation And a cheapest charge selection unit that selects the lowest charge among the charges calculated by the department.

  Further, as in claim 11, when the charge calculation network data is data generated by the charge calculation network generation device according to claim 8, the cheapest charge calculation unit includes the charge calculation network in the charge calculation network. A minimum cost search unit that searches for a route that minimizes the cost of the arc from the start point to the end point is provided, and the cost of the route searched by the minimum cost search unit is the lowest price.

  If the charge is a railway fare as in claim 12, the fare of a complicated railway network can be calculated efficiently and in a short time.

According to the present invention, a fee calculation that can efficiently calculate a fee calculation between two points across a plurality of systems having different fee structures by applying a shortest route search algorithm efficiently and in a short time with a small amount of calculation processing. It becomes possible to provide a network. By using this fee calculation network, for example, it is possible to easily calculate a fee such as a complicated railway fare in a short time.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following embodiments, a case where the present invention is applied to railway fare calculation will be described as an example.
FIG. 1 is a block diagram showing a configuration of a first embodiment of a fee calculation apparatus according to the present invention.
In FIG. 1, a fee calculation apparatus 1 according to the present embodiment includes a departure station (hereinafter referred to as a departure station) and an arrival station (hereinafter referred to as a destination station) belonging to a railway company's fee system. Of course, the lowest fare is calculated between the departure station and arrival station across multiple systems with different fee systems. Further, the fee calculation device 1 has a network data generation function for fee calculation as will be described later, and also serves as a fee calculation network generation device. Here, the fare system is a fare calculation rule set for each railway company, and each railway company has its own fare system. In the present invention, even if each railway company has the same charge system, it is handled as a different charge system because each company is different, and the same railway company has two or more charge systems. Even if you have a credit card, it is handled as a separate fee structure.

The fee calculation device 1 includes an input unit 2, a data storage device 3, a network data generation unit 4, a minimum cost search unit 5, a search threshold setting unit 6, a cheapest fare calculation unit 7, and an output unit 8. And a storage device 9 . Here, the fee calculation device 1 is constituted by a computer, for example, and the functions of the network data generation unit 4, the minimum cost search unit 5, the search threshold setting unit 6 and the cheapest fare calculation unit 7 are provided as software. It is.

The input unit 2 is for inputting data of a departure station as a starting point and a destination station as an end point.
The data storage device 3 stores in advance fare triangle table data for each company, transit relationship data between the companies, transit discount table data, and the like, and corresponds to a data storage unit. The fare triangular table data defines the fare between two stations in the same company as shown in FIG. 2, and corresponds to fare data. The connection relation data defines station data that can be transferred to another company for each company and its connection relation. The transit discount table data is data that defines a route for applying transit discount between a plurality of companies having different fee systems as shown in FIG. 3 and the discount amount, and corresponds to fee discount data.

  The network data generation unit 4 generates network data for constructing a fee network as shown in FIG. 5 based on the data stored in the data storage device 3, and is connected to the sub network data generation unit 4A and the connection arc data. A generation unit 4B is provided. The sub-network generation unit 4A uses each station in the same system as a node, connects all nodes directly to each other by arcs, and weights the connections between the nodes connected by arcs based on the fare triangle table data in FIG. Subnetwork data for constructing the subnetworks N1 to N3 shown in FIG. 5 is generated for each fee system. In addition, if weighting is given to either a node or an arc, it is assumed that the connection between nodes connected by the arc is weighted (hereinafter referred to as cost). The connection arc data generation unit 4B generates a connection arc indicated by a dotted line in FIG. 5 that connects between nodes corresponding to contact stations between companies with different charge systems based on the connection relation data, and connects with the connection arc. Set the cost of connection between the nodes to be zero to zero. Note that the arc data that connects nodes within the same fee structure and the connection arc data that connects between different fee structures are the two that go from one node to the other, and from the other node to one node. Although it consists of arc data, in FIG. 5 it is shown by a single line for the sake of simplicity.

  The minimum cost search unit 5 uses a toll network constructed from the sub-network data and connection arc data, for example, between a departure station and a destination station input from the input unit 2 by a shortest route search algorithm such as Dijkstra method. Search for the route with the lowest cost. The fare of the route searched by the minimum cost search unit 5 does not take into account the transfer discount described later, but is the fare before the connection discount is applied.

  The search threshold setting unit 6 is a diagram to be described later based on the connection discount table data in FIG. 3 when there is a possibility that a route to which the connection discount is applied exists between the input departure station and the arrival station. 6, the maximum discount amount applied to the route between the departure station and the arrival station is calculated, and the maximum discount amount and the fare of the route with the minimum cost searched by the minimum cost search unit 5 are calculated. In total, the search threshold necessary for calculating the lowest fare between the departure station and the arrival station is set. Here, the search threshold value setting unit 6 has a function of a maximum discount amount calculation unit.

As described in the flowchart of FIG. 6 to be described later, the cheapest fare calculation unit 7 searches for all routes that have a fare less than or equal to the search threshold set by the search threshold setting unit 6 using a fee network. after the fare after the discount of the route it was determined if there is a transfer discount route in the route, select the cheapest fare in all of the search path, and calculates the fare as the lowest fare, the most Corresponds to the low price calculation unit . Here, the cheapest fare calculator 7 has functions of a first charge calculator, a second charge calculator, and a cheapest charge selector.

The output unit 8 outputs the result calculated by the lowest fare calculation unit 7 as the lowest fare data.
The storage device 9 stores the fee network data generated by the network data generating unit 4 as billing network data, which stores the various data required in the lowest fare calculation process temporarily This corresponds to a network data storage unit .

Next, each operation | movement of the production | generation of fee network data by the fee calculation apparatus 1 and calculation of the cheapest fare is demonstrated.
FIG. 4 is a flowchart for generating fee network data.
In step 1 (indicated by S1 in the figure, the same shall apply hereinafter), the network data generation unit 4 reads the fare triangle table data for each company and the transit relation data from the data storage device 3.
In step 2, the sub-network data generation unit 4A generates sub-network data for each different fee structure from the fare triangle table data of each company.
In step 3, the connection arc data generation unit 4B generates connection arc data for connecting sub-networks of different charge systems from the connection relation data.
In step 4, the generated fee network data including the sub-network data and connection arc data for each different fee structure is stored in the storage device 9 as fee calculation network data.

  Using such fee network data, each station in the same system as shown in FIG. 5 is set as a node (af, gl, mr in the figure), and all nodes are arcs (solid lines in the figure). ) Between the sub-networks N1 to N3 for each of the different fee systems 1 to 3 that are directly connected to each other and charge the connection cost between the nodes, and between the nodes of the sub-networks N1 to N3 (in FIG. 5, d, g, and d A fee network consisting of connection arcs (dotted lines in the figure) connecting m, k and o) is constructed.

FIG. 6 is a flowchart showing calculation of the lowest fare by the generated fee network. Here, a case will be described in which the departure station and the arrival station span different companies. In this case, if there is a route to which the transit discount is applied between the input departure station and the arrival station, the lowest fare searched by the minimum cost search unit 5 is not necessarily the lowest.
In step 11, the process waits until the arrival / departure station data is input from the input unit 2.
In step 12, the minimum cost search unit 5 reads the charge network data from the storage device 9, and uses the shortest route search algorithm in the charge network of FIG. And stored in the storage device 9, for example.
In step 13, the search threshold setting unit 6 reads the transfer discount table data from the data storage device 3, obtains the maximum discount amount between the departure station and the arrival station, and stores it in the storage device 9, for example.

A method for calculating the maximum discount amount will be described with reference to FIG. In FIG. 7, a case where there are four systems with different fee systems and up to a discount between three fee systems will be described.
In FIG. 7, a route from the node c to the node u will be described as an example. The maximum discount amount obtained here is the maximum discount amount between fee systems. For example, possible combinations of discounts between fee systems on all routes from c to u are as follows (a) to (e).
(A) One discount between two fee structures (between fee structures 1, 2; one between fee structures 2, 3; one between fee structures 3, 4)
(B) Two discounts between two fee systems (combination of two of the above three fee systems)
(C) 3 discounts between 2 fee systems (all between the above 3 fee systems)
(D) One discount between three fee systems (one between fee systems 1 to 3 or between fee systems 2 to 4)
(E) One discount between three fee systems + one discount between two fee systems (between fee systems 1 to 3 and between fee systems 3 and 4, or between fee systems 1 and 2 and between fee systems 2 to 4 One)

The maximum discount amount between the respective fee systems can be obtained from the maximum discount amount between the fee systems of the transit relation data stored in the data storage device 3.
For example, if the maximum discount between two fee systems is 20 yen, and the maximum discount between three fee systems is 30 yen, (a) is 20 yen, (b) is 40 yen, (c) is 60 yen, (d ) 30 yen, (e) will be 50 yen, and the maximum discount can be set at 60 yen. In the fee calculation network of FIG. 7, for example, a route (c → d → g → i → k → o → r → m → v → u) as shown by a solid line is conceivable. The route is not limited to this.
This maximum discount amount can be calculated from the following equation (1).

Here, d _i represents the maximum discount amount between i fee systems, x _i represents the number of discounts between i fee systems, and N represents the maximum number of fee systems to be used. Further, a constraint condition for the number of discounts applied between i fee systems is defined as shown in Equation (2).
The maximum discount amount in the fee network does not change unless the maximum discount amount changes between the fee systems or the maximum number of fee systems to which the discount is applied. .
In the description of FIG. 7, there is a case where there are four fee systems and up to a discount between three fee systems, but the present invention is not limited to this, and the number of systems may be any number. Discounts between fee systems are not limited to discounts between two fee systems and discounts between three fee systems.

In step 14, the search threshold = the lowest fare before application + the maximum discount amount is set from the lowest fare and the maximum discount amount obtained in steps 12 and 13, respectively. That is, when a maximum discount amount is applied, a search threshold is set as an upper limit value of a fare that is less than or equal to the lowest fare before application, and only a fare route before applying a transit discount that is less than or equal to the search threshold is searched. . This eliminates the need to search for a fare route exceeding the search threshold when obtaining the lowest fare, and the number of searched routes can be greatly reduced. This search threshold value is stored in the storage device 9, for example.
In step 15, a search is performed for all routes of the pre-discount fare that are below the search threshold by the toll network and routes to which the discount is applied in all the searched routes.
In step 16, the connection discount amount of each discount application route searched in step 15 is calculated based on the connection discount table data, and stored in the storage device 9, for example.
In step 17, the fare before connection discount application for each discount application route obtained in step 15 is subtracted from the connection discount amount obtained in step 16 to calculate the fare after the connection discount on the discount application route.
In step 18, it is determined whether or not the discounted fare calculation has been completed for all connection discount application routes. If the determination is YES, the process proceeds to step 19.
In step 19, the cheapest one of the lowest fares calculated in step 12 and all fares of all discount application routes calculated in step 17 is output as the lowest fares via the output unit 8.

  By using the fee network of the first embodiment, the nodes are directly connected by arcs, and the calculation cost between the nodes is the fee. Therefore, compared to the case of using the conventional physical network, the distance to the fare is reduced. There is no need to perform conversion processing. In addition, even when the departure and arrival stations cross between different fee systems, the fare can be obtained by adding the fare calculated by each fee system, and using the shortest route search algorithm such as Dijkstra method as the shortest route search problem Can be calculated. Therefore, especially when the departure station and the arrival station cross between different fee systems, the number of search routes and the amount of calculation can be greatly reduced, and the lowest fare can be calculated in a very short time.

Next, a second embodiment of the present invention will be described.
The present embodiment is a network that enables the cheapest fare calculation between the departure station and the arrival station across different fee systems to be performed at a higher speed than in the first embodiment.
FIG. 8 is a block diagram showing the configuration of the second embodiment of the fee calculation apparatus according to the present invention. In addition, the same code | symbol is attached | subjected to the same element as 1st Embodiment, and description is abbreviate | omitted.
In FIG. 8, the fee calculation apparatus 10 of the present embodiment is configured by adding a simple network data generation unit 11 to the fee calculation apparatus 1 of the first embodiment.
The simple network data generation unit 11 removes nodes other than the nodes connected by the connection arc from the subnetwork data of the fee network data generated by the network data generation unit 4, and node data connected by the connection arc And sub-network data composed only of arc data connecting these nodes, and simple network data (simple network data) composed of these sub-network data and connection arc data are generated and stored in the storage device 9.
Instead of removing nodes other than the nodes connected by the connection arc from the subnetwork data of the fee network data generated by the network data generation unit 4, the simple network data generation unit 11 starts from FIG. The network data (simple network data) consisting of the sub-network data consisting of the node data connected by the connection arc and the arc connecting the nodes and the connection arc data as shown in FIG. .

Next, each operation | movement of the production | generation of the simple network data by the fee calculation apparatus 10 and calculation of the cheapest fare is demonstrated.
FIG. 9 is a flowchart of generation of simple network data.
In step 21, the network data generation unit 4 generates the fee network data described in the first embodiment, which includes the same subnetwork data and connection arc data as in the first embodiment, and stores them in the storage device 9. The fee network data generation procedure is generated as shown in the flowchart of FIG. 4 of the first embodiment.
In step 22, the simple network data generation unit 11 reads the fee network data from the storage device 9.
In step 23, the simple network data generation unit 11 connects the nodes connected by connection arcs and the arc-only subnetwork data connecting these nodes and the nodes between the subnetworks based on the read data. Simple network data composed of connection arcs is generated (first method).
In step 24, the simple network data generated in step 23 is stored in the storage device 9.
In addition, in the simple network data generated in step 23, each network data corresponding to all combinations of the departure station and the arrival station across the sub-networks having different charge systems is generated in advance. Each network data may be stored in the storage device 9 as the execution simple network data to be described later for performing the charge calculation (second method).

  FIG. 10 shows an example of a simple network using simple network data generated based on the fee network data of FIG.

FIG. 11 is a flowchart showing calculation of the lowest fare using a simple network.
In step 31, the system waits until the arrival / departure station data is input from the input unit 2.
In step 32, execution simple network data is generated as simple charge calculation network data for actually performing the fare calculation using the simple network. Specifically, in the simple network data generation unit 11, it is composed of subnetwork data composed only of nodes connected by connection arcs, arcs connecting these nodes, and connection arcs connecting nodes between subnetworks. When the simplified network data to be stored is stored (the first method), the input departure station and arrival station data (for example, data of the node a of the departure station and the node r of the arrival station described later) and the departure station Charge arc data (for example, data between nodes a and d, between nodes m and r, and between nodes or), which will be described later, is added for each of the arc data connected to each of the destination stations. It is generated as execution simple network data (not shown in FIG. 9) and temporarily stored in the storage device 9.
Further, when the execution simple network data for all combinations of the departure station and the arrival station is stored in the storage device 9 in advance (second method), the input departure station and arrival station data (for example, a departure station described later) The execution simple network data (not shown in FIG. 9) corresponding to the node a and the node r of the arrival station are selected and temporarily stored in another location of the storage device 9.

  Thereafter, in steps 33 to 40, the lowest fare is calculated and output in the same manner as in the first embodiment using the execution simple network as shown in FIG. 12 constructed by the execution simple network data generated in step 32. The execution simple network shown in FIG. 12 is connected to the simple network shown in FIG. 10 for each charge for connecting the node a at the departure station, the node r at the arrival station, the nodes ad, the nodes mr, and the nodes o-r. An arc is added. Note that the operations of Steps 33 to 40 are the same as Steps 12 to 19 of the flowchart of FIG.

  By using an execution simple network that uses such a simple network, the number of branches (search) when finding a route in the fare calculation process can be greatly reduced, so the fare between the departure station and the arrival station across different fare systems Calculations can be performed even faster, and the fare calculation processing time can be shortened.

Next, a third embodiment of the present invention will be described.
In the fee network of the first embodiment shown in FIG. 5, when trying to calculate the cheapest fare between the departure station and the arrival station belonging to different fee systems, a route inconsistent with the rules defined for the fare calculation is obtained. Includes possibilities.

Here, the following two routes conflict with the rule.
One is an inaccessible route. This will be described with reference to the fee network in FIG. FIG. 13 shows that fee systems A and C can be communicated by nodes d and f, and fee systems B and C can be communicated by nodes e and f, but fee systems A and C are defined as incapable of communication. This is the case for a fee network. Here, “contactable” means that connections can be made without going through the ticket gates, in other words, it may be calculated as a single route, and “not connectable” means that connections cannot be made without going through the ticket gates once. In other words, it means that it cannot be calculated as a single route. Therefore, for example, the fare calculation from the point a of the fee system A to the point b of the fee system B is a route that should not be calculated as a route originally according to the rules defined in the fare calculation. In the toll network, since the cost of connection of the route part d → f → e of the route a → d → f → e → b is zero, There is a possibility that the fare is calculated as a route from point a to point b from the route d → e → b.

  The other is a stopover route. This will be described with reference to the fee network in FIG. For example, in the same fare system A, the fare from point a to point d is determined on the assumption that the user gets on the way from point a to d without getting off halfway. It is not supposed to pass through the ticket gate. Therefore, as a rule for calculating the fare, the fare calculation between a and d should not be calculated on the a-m-d route to get off at the point m. However, in the toll network of FIG. 13, for example, when the connection cost between a and d is equal to the cost of adding the connection cost between a and m and the connection cost between m and d, the point a to the point d When the fare calculation up to is performed, there is a possibility of obtaining the amd route.

The third embodiment generates a network that can avoid finding a transit route that contradicts the rules defined for fare calculation as described above. Hereinafter, the network is a direct network.
FIG. 14 is a block diagram showing the configuration of the third embodiment of the fee calculation apparatus according to the present invention. In addition, the same code | symbol is attached | subjected to the same element as 1st Embodiment, and description is abbreviate | omitted.
In FIG. 14, the charge calculation device 20 of the present embodiment is configured by adding a direct network data generation unit 21 to the charge calculation device 1 of the first embodiment.

  The direct network data generation unit 21 adds dummy nodes to all nodes (hereinafter referred to as original nodes) in each sub-network generated by the network data generation unit 4. Furthermore, for arcs connecting between nodes in each sub-network (hereinafter referred to as original nodes), the original node of the arc source that is the connection source is changed to a dummy node corresponding to the original node, and is the connection destination. The original node is the arc destination. For connection arcs connecting sub-networks with different charge systems, change the arc source that is the connection source to the original node, and change the source node of the arc destination that is the connection destination to a dummy node corresponding to the source node. Generate arc data. Then, the direct network data including the fee network data in which the arc data (including the connection arc data is changed) and the added dummy node data is stored in the storage device 9 as the fee calculation network data.

Next, each operation | movement of the production | generation of the direct network data by the charge calculation apparatus 20 and calculation of the cheapest fare is demonstrated.
FIG. 15 is a flowchart of generation of direct network data.
In step 51, the network data generator 4 generates the fee network data of the first embodiment and stores it in the storage device 9. The fee network data generation procedure is shown in the flowchart of FIG. 4 of the first embodiment.
In step 52, the direct network data generation unit 21 reads fee network data from the storage device 9.
In step 53, dummy nodes are added to all nodes of the read fee network data.
In step 54, one of the arc data of the fee network data is retrieved.
In step 55, it is determined whether or not the extracted arc data is a connection arc. If it is not a connection arc, it is determined as arc data for connecting nodes in the sub-network. Is changed to a dummy node. On the other hand, if it is a connection arc, in step 57, the original node of the arc destination is changed to a dummy node.
In step 58, it is determined whether or not all arc data of the fee network data has been taken out. If the determination is YES, the process proceeds to step 59, where the sub-network data with the dummy node added and the changed arc data are directly specified. As network data, this direct network data is stored in the storage device 9 as fee calculation network data.

  FIG. 16 shows an example of a direct network constructed using direct network data generated based on the fee network data of FIG. In the figure, a 'to f' and m 'to o' indicate dummy nodes. When calculating the fare, each dummy node is treated as a departure station node, and the original node is treated as a destination station node.

  For example, when the fare calculation from the point a of the fee system A to the point b of the fee system B is performed using such a direct network, the dummy node a ′ becomes the departure station node, and the route a ′ → d → f ′. Although the fee systems A and C are connected, the connection destination of the dummy node f ′ is only the original node in the same fee system C, and the node of the fee system B is not the connection destination. A route from fare system C to fare system B is not configured. In order to go from the fee system A to the fee system B, it is a network that can only be reached through a ticket gate. Therefore, it is possible to avoid the problem of an inaccessible route that is one of the contradictions in the rules described above.

For example, when the fare calculation from the point a to the point d is performed in the same fee system A, the dummy node a ′ becomes the departure node, and the connection destination of the dummy node a ′ is the original nodes m and d. Since the original node m is treated as a destination station node, there is no connection destination. Therefore, only the route having the dummy node a ′ as the departure station and the former node d as the arrival station is provided, and the problem of a stopover route that is a contradiction in the second rule described above can be avoided.
As for the connection source / connection destination of the arc, when each source node is a source station node and each dummy node is a destination station node, the connection source / connection destination of the above embodiment is reversed and each sub node For arcs that connect between original nodes in the network, the connection source is the original node, the arc destination that is the connection destination is changed to a dummy node, and for connection arcs that connect between sub-networks with different pricing systems, The arc source that is the connection source may be changed to a dummy node, and the arc destination that is the connection destination may be the original node.

The cheapest fare calculation process by this direct network is the same as the flowchart of FIG. 6 using the charge network of the first embodiment. However, the calculation method of the maximum discount amount by the search threshold setting unit 6 is different.
Hereinafter, a method for calculating the maximum discount amount in the direct network will be described with reference to FIG.

In FIG. 7, the same route from the node c to the node u as in the charge network will be described as an example. The charge network of the first embodiment can select a route using a plurality of arcs in the same charge system, but the direct network can use only one arc in the same charge system. For this reason, a route to which the maximum discount amount may be applied in the direct network is a route (c → d → g → k → o → m → v → u) indicated by a dotted line in the figure. Combinations of discounts between fee systems that may be applied to this route are the following (a) to (c).
(A) One discount between two fee structures (between fee structures 1, 2; one between fee structures 2, 3; one between fee structures 3, 4)
(B) Two discounts between two fee systems (combination of two of the above three fee systems)
(C) One discount between three fee systems (one between fee systems 1 to 3 or between fee systems 2 to 4)

The maximum discount amount between each fee structure can be obtained from the maximum discount amount between the charge structures of the transit relation data stored in the data storage device 3, and is the same as in the case of the charge network described above. If the maximum discount amount is 20 yen, and the maximum discount amount between three fee systems is 30 yen, (a) will be 20 yen, (b) will be 40 yen, (c) 30 yen, and the maximum discount will be 40 yen. .
This maximum discount amount can be calculated from Equation (1) of Equation 1 as in the case of the charge network. The constraint condition of the number of discounts applied between i fee systems is as shown in Equation (3) of Equation 2.

Since it is clear that the maximum discount amount in the fee network is larger than the maximum discount amount in the direct network, it may be used as the maximum discount amount for calculating the upper limit value in the direct network.
Further, in the description of FIG. 7, there is a case where there are four fee systems and up to a discount between three fee systems, but the present invention is not limited to this, and the number of systems may be any number, Further, the discount between the fee systems is not limited to the discount between the two fee systems and the discount between the three fee systems.

Next, a fourth embodiment of the present invention will be described.
The fourth embodiment generates a network that can obtain the lowest fare as a shortest route search problem even when a transit discount system exists between different fee systems. Hereinafter, this network is referred to as a discount network.
FIG. 17 is a block diagram showing a configuration of the fourth embodiment of the fee calculation apparatus according to the present invention. In addition, the same code | symbol is attached | subjected to the same element as 1st Embodiment, and description is abbreviate | omitted.
In FIG. 17, the charge calculation device 30 of the present embodiment removes the search threshold setting unit 6 and the cheapest fare calculation unit 7 from the charge calculation device 1 of the first embodiment, and changes the connection discount applied arc data generation unit 31. It is configured by adding.

  The connection discount application arc data generation unit 31 applies the connection discount application based on the fare triangle table data of each company as shown in FIG. 2 and the connection discount table data as shown in FIG. Generate arc data. The connection discount application arc connects the nodes corresponding to the departure station and the arrival station of the route to which the connection discount is applied, and the connection cost between the nodes connected by the connection discount application arc is The fare after discount is applied by subtracting the discount amount from the fare before discount for that route. The route to which the transfer discount is applied and its departure / arrival station data can be set from the departure / arrival data and connection station data of the transfer discount table data. Can be set.

Next, operations of generating discount network data and calculating the lowest fare by the fee calculation device 30 will be described.
FIG. 18 is a flowchart of generating discount network data.
In step 61, fee network data is generated by the network data generating unit 4 and stored in the storage device 9. The fee network data generation procedure is shown in the flowchart of FIG. 4 of the first embodiment.
In step 62, the connection discount application arc data generation unit 31 generates connection discount application arc data based on the fare triangle table data of each company and the connection discount table data.
In step 63, discount network data is generated by combining the charge network data and the transfer discount application arc data generated in step 62.
In step 64, the discount network data generated in step 63 is stored in the storage device 9 as fee calculation network data.

  FIG. 19 shows an example of a discount network using discount network data. In FIG. 19, the thick lines in the diagram connecting between node b of fee structure 1 and node n of fee structure 3 and between node i of fee structure 2 and node p of fee structure 3 are connected. It is a discount application arc.

FIG. 20 is a flowchart of calculating the lowest fare using a discount network.
In step 71, the process waits until the arrival / departure station data is input from the input unit 2. When the input / output station data is input, the process proceeds to step 72.
In step 72, the minimum cost search unit 5 reads the discount network data from the storage device 9, and searches the discount network for the cheapest fare route between the departure station and the arrival station input using the shortest route search algorithm.
In step 73, the route fare searched in step 72 is output through the output unit 8 as the lowest fare result.
If such a discount network is used, the search result of the minimum cost search unit 5 becomes the lowest fare as it is, and there is an advantage that the calculation amount can be greatly reduced.

Next, a fifth embodiment of the present invention will be described.
When the network of the present invention is applied to railway fare calculation, (b) a route that passes multiple times of the same fare system, and (b) a route that connects five or more fare systems (the current railway fare calculation rule is fare calculation within 4 company lines) (C) There is a possibility that prohibited routes such as multiple passes at the same point may be included in the calculation target. However, (c) is a problem that occurs only in the case of a direct network.

For example, consider the case where the fare calculation between points pf is performed in the network configuration of FIG. In this case, the route of p → n → q → r → u → v → l → i → k → h → f is a route spanning the 5 fee system, and p → n → q → t → w → u → r The route q → g → f is a route that passes through the same point (point q) and the same fee system multiple times, and the route p → n → q → t → w → u → r → h → f has the same charge. A route that passes through the system multiple times.
In order to solve such a problem, it is necessary to search for a route by the k-shortest route search algorithm repeatedly until an uninhibited route is searched. However, when the number of nodes is large, the calculation load is large, and k may become large depending on the shape of the network, which is inefficient.

  In the present embodiment, the above-mentioned problems are solved and the lowest fare is efficiently calculated by adopting a network form in which the network is simplified into a higher level network and a detailed lower level network. is there.

FIG. 22 is a block diagram showing a configuration of the fifth embodiment of the fee calculation apparatus according to the present invention. In addition, the same code | symbol is attached | subjected to the same element as 1st Embodiment, and description is abbreviate | omitted.
In FIG. 22, the fee calculation device 40 of the present embodiment has a configuration of the fee calculation device 1 of the first embodiment, the upper network data generation unit 41, the contact pattern data generation unit 42, and the lower network data generation unit 43. It is configured by adding. The data storage device 3 stores in advance the prohibition condition data (A) to (C) described above.

  As shown in FIG. 23, the upper network data generation unit 41 regards each sub-network (each fee structure) having a different charge system in the charge network as one node (assumed to be an upper node), and based on the transfer relation data. Simplified higher-level network data is generated in which the higher-level nodes (sub-networks) are directly connected to each other by one arc (referred to as higher-level arc). Here, the cost of connection between the upper nodes connected by the upper arc of the upper network is set to 1. Thus, by calculating the route up to cost 3 by the shortest route search algorithm, the searched route can be limited within the four fee system, and the search for the route across the five fee system can be prevented.

  The contact pattern data generation unit 42 is based on the upper network data and the prohibition condition data in FIG. 24 for each pair of charge systems corresponding to the number of combinations of the departure station and arrival station nodes across different charge systems. The contact pattern data as shown is generated in advance. The upper arcs that connect the fee systems have direction as shown by the arrows in the figure. FIG. 24 shows an example of a contact pattern when the first station is a point p and the destination station is a point f in the network of FIG. 21. In this case, there are two contact patterns (A) and (B). . These contact pattern data are such that the cost value of the high-order arc from the high-order node including the start point of the high-order network to the high-order node including the end point is not more than a predetermined cost value (for example, the above-mentioned cost 3) and the high-order node in the route from the start point to the end point The connection pattern data satisfying the prohibition condition has a connection relationship that does not overlap.

  The lower network data generation unit 43 replaces each upper node of the contact pattern with the subnetwork data of the charge network, and the detailed lower network data satisfying the prohibition condition as shown in FIG. 25 in which these subnetworks are connected by a connection arc. Is generated. 25A corresponds to the contact pattern of FIG. 24A, and FIG. 25B corresponds to the contact pattern of FIG.

  The lower network form may be the charge network form of the first embodiment shown in FIG. 25, the simple charge calculation network form, the direct network form, or the discount network form of the second to fourth embodiments. Moreover, you may make it combine at least 2 network form of these simple charge calculation networks, a direct network, and a discount network.

FIG. 26 is a flowchart of generation of network data having a hierarchical structure according to the fifth embodiment.
In step 81, fee network data is generated by the network data generating unit 4 and stored in the storage device 9. The fee network data generation procedure is shown in the flowchart of FIG. 4 of the first embodiment.
In step 82, the upper network data generator 41 generates upper network data as shown in FIG.
In step 83, the contact pattern data generation unit 42 generates contact pattern data for each pair of fee systems corresponding to the number of combinations of the departure station and arrival station nodes, based on the upper network data and the prohibition condition data. To do.
In step 84, the lower-layer network data generation unit 43 generates detailed lower-layer network data from the selected contact pattern data and toll network sub-network data based on the input information on the departure station and destination station.
In step 85, the generated data of the upper network, the contact pattern, and the lower network are stored in the storage device 9.

FIG. 27 is a flowchart of calculating the lowest fare using a hierarchical network, and an example will be described in which the point p is input as the departure station and the point f is input as the arrival station in the network of FIG.
In step 91, the process waits until the arrival / departure station data is input from the input unit 2.
In step 92, a contact pattern corresponding to the inputted station information is selected. Here, if the departure station is a point p and the arrival station is a point f, the communication pattern from the prohibition condition is a charge system 4 → charge system 1 → charge system 2 communication pattern shown in FIG. 2) of the charge system 4 → charge system 5 → charge system 2 shown in FIG.
In step 93, a subordinate network is generated from the selected contact pattern. In this case, the lower networks shown in FIGS. 25A and 25B are generated.

In step 94, the lowest cost fare in each lower network generated in step 93 is calculated by the minimum cost search unit 5 using the shortest route search algorithm. Here, if there is no possibility that the connection discount is applied, the fare obtained by the minimum cost search unit 5 may be set as the lowest fare. When there is a possibility that a transfer discount is applied, the search threshold is set in the same manner as in the first embodiment, and the cheapest fare below that is set as the cheapest fare. Further, if a discount network in which a transit discount application arc is set as a lower network, the fare calculated by the minimum cost search unit 5 can be made the cheapest.
In step 95, the lowest fare of each lower network calculated in step 94 is compared, and the cheaper fare is output as the lowest fare via the output unit 8.

In the above example, the subnetwork of the fee network is used as the lower network of the hierarchical network. However, as the lower network form, the simplified fee calculation network of the second embodiment, the third embodiment Each of the direct network and the discount network of the fourth embodiment may be applied, or two of these may be combined, or all three may be combined. When there is a transit discount route, except for the case where the lower network is a discount network, after the fare calculation in step 94, the processing after step 13 in FIG. The fare will be calculated.
In addition, when the departure station and arrival station information is input, the contact pattern data generation unit 42 has a connection relation that satisfies a predetermined prohibition condition in charge calculation according to the departure station and arrival station information. Only the system may be selected from the upper network data to generate the contact pattern. In this case, it is not necessary to store a large number of contact pattern data.

  In this way, it is prohibited if the network configuration is hierarchized into a simplified upper network and a detailed lower network whose route is restricted based on prohibition conditions, and the lowest fare is calculated by the lower network. Routes are no longer calculated and the number of routes sought is greatly reduced. Therefore, the above-mentioned problems can be avoided and the lowest fare can be calculated efficiently in a short time.

  In each of the above-described embodiments, the simple charge calculation network, the direct network, and the discount network are individually described. However, a simple charge calculation network and a direct network are added by adding a dummy node and a transit discount application arc to the simple charge calculation network. And a discount network may be combined. Further, a combination of a simple fee calculation network and a direct network, a combination of a simple fee calculation network and a discount network, and a combination of a direct network and a discount network are also conceivable.

The fare calculation network of the present invention is not limited to application to railway fare calculation. For example, the following application is possible.
(1) Application to calculation of parking charges and power charges for parking lots, etc., where the charge setting per unit time varies depending on the time of day.
(2) Application to calculation of transportation / delivery charges such as land transportation, maritime transportation, air transportation, etc. with different fee settings depending on weight and distance. In this case, land transportation, sea transportation, and air transportation may straddle each other as different systems.
(3) Application to toll road toll calculation.
(4) Application to insurance premium calculation. For example, an insurance company may be selected for each insurance content.
(5) Application to calculation of taxes.
(6) Application to calculation of the amount of welfare payment that minimizes the amount of welfare payment without reducing the number of people who receive welfare protection. For example, it can be used for calculation for determining the amount of payment for welfare so that the take-up at the time of work (the loss of welfare) does not become the take-up when receiving welfare.
(7) Application to billing calculation of application usage by SaaS (software as a service). For example, it may be possible to calculate a fee such as increasing the discount rate when used for a long time, or discounting a combination of specific functions.

The sub-network in the above-described embodiment has a configuration in which all nodes in the system are directly connected to each other by arc. However, depending on the application target of the fee calculation network, it is not necessary to directly connect all the nodes. As long as it is necessary to connect directly with an arc as necessary, the cost of connection between nodes connected by the arc is charged. As such an example, the case where it applies to the above-mentioned parking charge calculation can be considered, for example.
For example, as shown in FIG. 28, the charge calculation network configuration used for the calculation of the parking charge for a parking lot with a charge structure in which 300 yen is added for 3 hours after warehousing and 100 yen is added every 30 minutes thereafter. A calculation network configuration may be used.
The fee calculation network in FIG. 28 is centered on a node with a parking time of 0 minutes, arcs are arranged radially from the node as a connection source, and the connection destination of each arc is a node indicating the parking time from warehousing and connected by an arc. In this configuration, the cost between the nodes is set as a parking fee according to the parking time of the arc destination node. Here, the setting interval of the parking time is arbitrary, such as a minute unit or a second unit, and may be set appropriately according to the charge system.
And, by connecting nodes with a parking time of 0 minutes to other sub-networks (for example, railway fare at a station where a certain parking lot is located) with a connection arc, a charge calculation network that spans multiple sub-networks can be configured. It becomes possible.

  In the present invention, even if each company has the same fee system, it can be handled as different fee systems because each company is different, and the same company has two or more fee systems. Can be handled as different fee systems, so that it is possible to flexibly respond to changes in the fee structure and to set various fee networks.

The block diagram which shows the structure of 1st Embodiment of the fee calculation apparatus which concerns on this invention. Figure showing an example of fare triangle table data Figure showing an example of transit discount table data Charge network data generation flowchart of the first embodiment Diagram showing an example of a fee network The lowest fare calculation flowchart using the charge network Illustration of the maximum discount calculation method The block diagram which shows the structure of 2nd Embodiment of the fee calculation apparatus which concerns on this invention. Simple network data generation flowchart of the second embodiment Diagram showing an example of a simple network The lowest fare calculation flowchart using a simple network Figure showing an example of a simple execution network Explanatory diagram of problems when calculating the lowest fare on the fee network The block diagram which shows the structure of 3rd Embodiment of the fee calculation apparatus which concerns on this invention. Direct network data generation flowchart Diagram showing an example of a direct network The block diagram which shows the structure of 4th Embodiment of the fee calculation apparatus which concerns on this invention. Discount network data generation flowchart Diagram showing an example of a discount network The lowest fare calculation flowchart using a discount network Explanatory diagram of problems when applying the toll network to railway fare calculation The block diagram which shows the structure of 5th Embodiment of the fee calculation apparatus which concerns on this invention. Diagram showing an example of the upper network Figure showing contact pattern example A diagram showing an example of a subordinate network Hierarchical network data generation flowchart The lowest fare calculation flowchart using a hierarchical network The figure which shows an example of the charge calculation network applied to the parking charge calculation of the parking lot Diagram showing an example of a physical network

Explanation of symbols

1, 10, 20, 30, 40 Charge calculation device 2 Input unit 3 Data storage device 4 Network data generation unit 5 Minimum cost search unit 6 Search threshold setting unit 7 Lowest fare calculation unit 8 Output unit 9 Storage device 11 Simple network data Generation unit 21 Direct network data generation unit 31 Transfer discount application arc data generation unit 41 Upper network data generation unit 42 Contact pattern generation unit 43 Lower network data generation unit

Claims (12)

A data storage unit for storing charge data defining charges for each system of a plurality of systems of different charge systems and connection relation data indicating a connection relation between each system ;
A network in which each point in each system of the plurality of systems is a node, each node is directly connected by an arc as necessary, and the connection cost between the nodes connected by the arc is a fee based on the fee data Data is generated, network data for each system is set as subnetwork data, connection arc data for connecting nodes of different subnetwork data based on the connection relation data is generated, and the subnetwork data and the connection arc are generated. A network data generator that uses the data as charge calculation network data;
A charge calculation network generating device characterized by comprising: The network data generation unit is configured to connect all nodes in each of the plurality of systems directly to each other by an arc, and network data in which the cost of connection between the nodes connected by the arc is a fee based on the fee data The charge calculation network generation device according to claim 1, wherein the charge calculation network data is generated as the charge calculation network data. Before kine Tsu network data generation unit, according to the cost of the connections between the nodes are pre-connected with Kia over click to claim 1 or 2 and configured to generate the billing network data and Rates zero Charge calculation network generator. The network data generation unit regards each sub-network in the fee calculation network data as one upper node, and connects the upper nodes with one upper arc based on the connection relation data. The network data is hierarchized by generating upper network data with a connection cost between the upper nodes connected by 1 as 1,
From the upper node including the start point of the upper network to the upper node including the end point based on a predetermined prohibition condition in charge calculation when information on the start point and the end point across multiple systems of different charge systems is input A contact pattern having a connection relationship such that the cost value of the upper arc is equal to or less than a predetermined cost value and the upper node does not overlap in the route from the start point to the end point is selected from the upper network data,
Lower network data satisfying the prohibition condition is generated from each subnetwork corresponding to each upper node in the selected contact pattern data and each connection arc data connecting between the subnetworks, and the lower network data is charged The fee calculation network generation device according to claim 1, wherein the fee calculation network generation device is calculation network data. Each subnetwork data is composed of only node data connected by the connection arc and arc data connecting these nodes, and generates simple network data composed of the subnetwork data and the connection arc data,
When starting point and ending point information across multiple systems with different fee systems is input, node data of starting point and ending point and arc data connected to each of starting point and ending point are added to the simple network data. The fee calculation network generation device according to claim 1, wherein the fee calculation network generation device is configured to generate simple fee calculation network data. Each subnetwork data is composed of only node data connected by the connection arc and arc data connecting these nodes, and generates simple network data composed of the subnetwork data and the connection arc data,
Simple fee calculation network data in which each arc data connected to each of the start and end points and each node data of start and end points across a plurality of systems of different charge systems is added to the simple network data, all of the start and end points In any one of claims 1 to 4, a simple charge calculation network data corresponding to the input information is selected when the start point and end point information is input. The fee calculation network generation device described. The fee calculation network data includes node data obtained by adding a dummy node for each node of each subnetwork, and a connection node or a connection destination of the arc connecting the nodes in each subnetwork. With sub-network data composed of arc data and
When the connection source of the arc is a dummy node, connection arc data having a connection destination of the connection arc connecting between nodes of different sub-networks as a dummy node,
The connection arc data according to any one of claims 1 to 6, wherein when the connection destination of the arc is a dummy node, the connection arc data includes a connection source of the connection arc that connects nodes of different sub-networks as a dummy node. Charge calculation network generation device described in 1. The fee calculation network data includes discount arc data indicating a discount path that directly connects two predetermined points with fee discounts between different sub-networks, and the predetermined fee discount data between the two predetermined points based on, billing network generating device according to any one of claims 1 to 7 which is configured to fee discounted cost of connections between nodes connected by the discount arc. An input unit for inputting information of the start point and the end point;
A network data storage unit for storing fee calculation network data generated by the fee calculation network generation device according to any one of claims 1 to 8;
When the start point and end point information is input, a charge calculation network is constructed based on the charge calculation network data in the network data storage unit, and the lowest charge from the start point to the end point of the charge calculation network is calculated. A low price calculator,
An output unit for outputting the calculated lowest price result;
A charge calculation device comprising: When the fee calculation network data is data generated by the fee calculation network generation device according to any one of claims 1 to 7,
The cheapest price calculation unit
A minimum cost search unit that searches for a route that minimizes the cost of the arc from the start point to the end point in the fee calculation network;
A maximum discount amount calculation unit for calculating a maximum value of the discount amount applied to the route from the start point to the end point based on a route to which the charge discount is applied and fee discount data indicating the discount amount;
A search threshold setting unit that sets a search threshold by adding the maximum discount amount calculated by the maximum discount amount calculation unit to the minimum cost fee searched by the minimum cost search unit;
Based on the fee calculation network, a first fee calculation unit that searches all routes for which the arc cost is equal to or less than the search threshold and calculates a fee for each route;
When there is a route to which a discount is applied to the searched route based on the fee discount data, the discount amount applied to the route from the pre-discount fee for the route calculated by the first fee calculation unit A second charge calculation unit that calculates a charge after applying the discount by subtracting
A cheapest charge selection unit for selecting the lowest charge among the charge calculated by the first charge calculation unit and the charge calculated by the second charge calculation unit;
The charge calculation device according to claim 9, wherein When the fee calculation network data is data generated by the fee calculation network generation device according to claim 8,
The cheapest charge calculation unit includes a minimum cost search unit that searches for a route that minimizes the cost of the arc from the start point to the end point in the charge calculation network, and calculates the cost of the route searched by the minimum cost search unit. The charge calculation apparatus according to claim 9 , wherein the charge calculation apparatus is configured to be the lowest charge.   The fee calculation device according to any one of claims 9 to 11, wherein the fee is a railway fare.

Images