Route search device, method, and program

Abstract

In one embodiment of the present invention, when searching for a route for visiting a plurality of geographically dispersed nodes, topology information representing the connection relationship between the plurality of nodes and parameter information including the number of users using each of the plurality of nodes and the travel times between the nodes are acquired in advance. In this state, ant colony optimization is used as a method for performing route optimization on the basis of the topology information and the parameter information, and matrix calculations are used for the route optimization using ant colony optimization. Selection of a route that corresponds to pheromone values by a plurality of imaginary agents, evaluation of the selected route, and an updating process of reflecting the result of the evaluation on the pheromone values are repeated in parallel a plurality of times by means of the matrix calculations, whereby an optimum route is obtained from a plurality of routes and the result is outputted as optimum route information.

Description

[Technical Field]

[0001] One aspect of the present invention relates to a route search device, method, and program used in the task of visiting communication buildings to restore power to communication equipment and restore network services, for example, when network services are stopped during a disaster. [Background technology]

[0002] For example, if a widespread power outage occurs due to a disaster such as an earthquake, and the power supply to the communication equipment in a communication building operated by a communication carrier runs out, users of that communication equipment will be unable to communicate. This situation is particularly undesirable if it occurs in a communication building that provides communication services to highly public buildings such as government offices, as it will seriously hinder disaster response operations.

[0003] To prepare for such an event, telecommunications carriers are considering measures to dispatch power supply vehicles to multiple telecommunications buildings in the affected area to restore power to each building.In this case, an important issue to be considered is how to determine the route for the power supply vehicles to minimize the time that telecommunications services are interrupted.

[0004] One method proposed for determining delivery routes during disasters is the method described in Non-Patent Document 1. This method uses a branch-and-bound method, which divides the problem to be solved into appropriate cases and finds an optimal solution by repeating the process of solving a relaxation problem to obtain an approximate value of the solution of the divided problem.

[0005] As another path search method, a method that applies ant colony optimization, as described in Non-Patent Document 2, for example, is also being considered. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Manami Ogawa et al., "Proposal of a Delivery Planning Algorithm Considering the Importance of Locations in Disasters," IEICE Technical Report, ICM Study Group, March 2021, pp. 120-124 [Non-patent document 2] M. Niu, R. Liu and H. Wang, “A Max-Min Ant System based on Decomposition for the Multi-Depot Cumulative Capacitated Vehicle Routing Problem”, 2021 IEEE Congress on Evolutionary Computation (CEC), 2021 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the route search method described in Non-Patent Document 1 has the problem that it uses a branch-and-bound method, which results in high calculation costs and takes a long time to find the optimal route. In addition, the method described in Non-Patent Document 2 has another problem in that it does not take into account the impact on users in the event of a disaster.

[0008] This invention has been made in light of the above circumstances, and aims to provide a technology that enables route searches to be performed efficiently with low calculation costs and that allows searches to be performed taking into account the impact on the user. [Means for solving the problem]

[0009] In order to solve the above problem, one aspect of a route search device or a route search method according to the present invention, when searching for a route to visit a plurality of geographically scattered nodes, acquires in advance topology information representing the connections between the plurality of nodes and parameter information including the number of users using each of the plurality of nodes and travel time between the nodes, applies ant colony optimization as a method for optimizing the route based on the topology information and the parameter information, and uses matrix calculations in the route optimization process using the ant colony optimization, and performs a predetermined number of parallel processes using the matrix calculations to select the route according to pheromone values ​​of a plurality of virtual agents, evaluate the selected route, and update the pheromone value to reflect the results of the evaluation, thereby finding an optimal route from among the routes, and generates and outputs optimal route information including the found optimal route and the corresponding result of the evaluation.

[0010] According to one aspect of the present invention, by applying ant colony optimization as a route optimization method, route search can be performed with a computational complexity of polynomial time, thereby significantly shortening the computation time and reducing the computational cost. Furthermore, a series of processes from route selection by ant colony optimization to its evaluation and pheromone update are performed in parallel for multiple ants using matrix calculations. Therefore, the time required for the computation can be significantly reduced compared to when the route search process is performed by sequential calculation. [Effects of the Invention]

[0011] That is, according to one aspect of the present invention, it is possible to provide a technique that enables a route search to be performed efficiently at low calculation cost and that allows the search to take into account the degree of impact on the user. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram showing an example of the hardware configuration of a route search device according to an embodiment of the present invention. [Figure 2]FIG. 2 is a block diagram showing an example of the software configuration of the route search device according to an embodiment of the present invention. [Figure 3] FIG. 3 is a flowchart showing an example of the overall processing procedure and processing contents executed by the control unit of the route search device shown in FIG. [Figure 4] FIG. 4 is a flowchart showing an example of the process procedure and process contents of the planning problem formulation process from the process procedure shown in FIG. [Figure 5] FIG. 5 is a flowchart showing an example of the processing procedure and processing content of the route optimization processing from the processing procedures shown in FIG. [Figure 6] FIG. 6 is a diagram for explaining an example of setting a planning problem as a technique related to the present invention. [Figure 7] FIG. 7 is a diagram showing an example of simultaneous optimization of VRP and network flow as a related art of the present invention. [Figure 8] FIG. 8 is a diagram for explaining an example of the definition of the influence level as a technique related to the present invention. [Figure 9] FIG. 9 is a diagram used to explain the outline of ant colony optimization applied in one embodiment of the present invention. [Figure 10] FIG. 10 is a diagram for explaining an example of a probabilistic route selection process used in the route optimization process shown in FIG. [Figure 11] FIG. 11 is a diagram for explaining an example of a route selection process using a matrix, which is used in the route optimization process shown in FIG. [Figure 12] FIG. 12 is a diagram for explaining an example of the calculation process of the influence degree used in the route optimization process shown in FIG. [Figure 13] FIG. 13 is a diagram for explaining an example of a pheromone update process using a matrix, which is used in the route optimization process shown in FIG. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0014] [One embodiment] (Related Technology) To facilitate understanding of the invention, first, a description will be given of techniques related to the route search method according to one embodiment.

[0015] (1) Problem setting Fig. 6 is a diagram showing an example of a network for which a problem is set up. As shown in Fig. 6, the network for which a route search is to be performed has a configuration in which multiple communication buildings (hereinafter referred to as nodes) V are connected to POIs (Points of Interface) that are connection points to ISPs (Internet Service Providers), and multiple users (user terminals) US are connected to each of these nodes v.

[0016] In Fig. 6, assume that all nodes have lost communication due to a lack of power (state 6a). In this state, when a mobile power supply vehicle MV departs from a base point (depo) (6b) and visits a node, the power supply of that node is restored. Then, when all nodes located on the route between the POI and the mobile power supply vehicle MV are restored (6c), the user US connected to that node can communicate.

[0017] Therefore, the number of users US with which communication is possible is determined by the route the mobile power supply vehicle MV takes to visit the nodes, and the requirement for problem setting is to maximize the number of users with which communication is possible, in other words, to minimize the number of users with whom communication is impossible.

[0018] (2) Simultaneous optimization of the Vehicle Routing Problem (VRP) and network flow FIG. 7 is a diagram illustrating an example of VRP and network flow optimization. Let V′ be the set of nodes excluding depo, and let Distance be the travel time between nodes i and j. ij ∈R +:i,j, the set of nodes i connected in the network is n(i), POI is sink, and the number of users corresponding to the amount of outflow at node i is Source. i Let ∈N. In this case, the path in the network flow is shown on the left side of FIG. 7, and the path in the VRP is shown on the right side of FIG.

[0019] (2-1) VRP Restrictions The VRP constraints are defined as follows: Now, the state where the vehicle has moved from node i to j is X ij = 1. Also, we assume that there is only one visit from node i to any other node and only one visit from any other node to i, and that there is only one visit from a depo to a node and only one visit from a node to a depo.

[0020] In this case, the visiting constraint in the travel constraint is

number

[0021] Also, the Miller-Tucker-Zemlin (MTZ) constraint specifies the order in which nodes i are visited by mtz. i Then,

number

[0022] (2-2) Flow Constraints The flow from node i to node j just before the vehicle arrives at node k is called flow ijk ∈R + , the source from node i just before the vehicle arrives at node k is ik ∈{0,1}, then source i If we multiply by , the number of connected users is calculated, then the flow constraint is

number

[0023] In equation (3), the left side represents the outflow from node i, and the first term on the right side represents the inflow to node i, so the outflow and inflow are equal.

[0024] (2-3) Relationship between visits and flow We assume that there is no source or flow at the node before the vehicle visits, and the state where node i has been visited just before visiting node k is called visited. ik If node i has not been visited, the outflow from node i is assumed to be 0.

[0025] In this case, the relationship between visits and flows is:

number

[0026] In this equation (4), the left side indicates the outflow from node i. ik If is 1, this relationship always holds.

[0027] (2-4) Visited constraint Assuming that the vehicle has already visited node i immediately before arriving at node j, there are two possible routes: a route from node i to node j directly, and a route from node i to node j via node k.

[0028] Also, since the product of the optimization variables is not a linear function, once converted to a linear equation, the constraints can be solved as a mixed integer linear program. That is, the visited constraints are

number

[0029] In this equation (5), visited_X ijk =visited ik X kj As, visited_X ijk ≦visited ik visited_X ijk ≦X kj visited_X ijk ≧visited ik +X kj -1 Convert to.

[0030] (3) Definition of impact The impact is defined as the sum of the number of users who cannot connect at each time. To calculate the impact, first count the number of users who cannot connect to the POI (= sink). Next, add up the number of users who cannot connect at each time, and the resulting value is the impact. In other words, impact = Σt (number of users who cannot connect (t)).

[0031] Figure 8 shows an example of the impact calculation results. In the figure, the dashed lines in the network topology indicate routes that cannot be connected, and the arrows indicate routes that can be connected. In this example, the number of users that can connect to the POI is 1,150, while the number of users that cannot be connected is 10.

[0032] (3-1) Objective function to minimize the impact The impact of adding up the number of users who cannot connect at each time is minimized. To achieve this, all routes are compared to find the route that minimizes the impact.

[0033] In this case, the objective function is

number

[0034] In this equation (6), the first term on the left side indicates the number of users who cannot connect, and the second term indicates the time when they cannot connect. kj )·X ij denotes the product of the linearly transformed optimization variables.

[0035] (overview) One embodiment of the present invention applies ant colony optimization as a method for optimizing a patrol route. Ant colony optimization is a route optimization method that focuses on the behavior that when ants find food and bring it back to the colony, if one ant finds a good (i.e., short) route from the colony to the food source, other ants are more likely to follow the same route by following the pheromones left by the previous ant, and a positive feedback effect will eventually result in all ants following one shortest route.

[0036] In one embodiment, the ant colony optimization is used to search for an optimal route when a mobile power supply vehicle visits multiple nodes that have lost communication and restores their communication facilities. That is, multiple virtual ants (also called agents) are introduced to visit each node, and the optimal route to visit each node is searched for by repeating the following process. Figure 9 shows an example of the route the ants take.

[0037] For example, first, multiple ants each select a route probabilistically according to the pheromone values ​​on the route, as shown in Figure 9, and patrol it. Then, the route selected by an ant that has completed patrolling the route is evaluated, and the pheromones on the route selected by the ant are updated based on the evaluation results. The above process is repeated a predetermined number of times, and finally, the route with the highest evaluation is output as optimal route information. Furthermore, the above route selection and pheromone update process is executed in parallel on a matrix for the routes of multiple ants.

[0038] Furthermore, when evaluating the above routes, the degree of impact that the selection of the route will have on users is calculated. The degree of impact is calculated from the number of users who cannot communicate and the time during which communication is unavailable. The degree of impact calculated for the ultimately obtained route with the highest evaluation is then included in the above optimal route information together with information representing that route and is output.

[0039] (Configuration example) The route search device SV according to an embodiment of the present invention is, for example, a server computer, and is located on the Web or in the cloud. Note that the route search device SV may also be a personal computer dedicated to the use of, for example, a local government or a delivery company.

[0040] 1 and 2 are block diagrams showing an example of the hardware configuration and software configuration of the route search device SV, respectively.

[0041] The route search device SV includes a control unit 1 that uses a hardware processor such as a central processing unit (CPU). A storage unit having a program storage unit 2 and a data storage unit 3, and a communication interface unit (hereinafter referred to as communication I / F unit) 4 are connected to the control unit 1 via a bus 5.

[0042] Under the control of the control unit 1, the communication I / F unit 4 uses a communication protocol defined by a network (not shown), including the Internet, to send and receive information data, for example, between a management terminal (not shown) used by a telecommunications carrier, or between the communication I / F unit 4 and an on-board terminal installed in a vehicle as a transportation medium.

[0043] The program storage unit 2 is configured, for example, by combining a non-volatile memory such as an HDD (Hard Disk Drive) or SSD (Solid State Drive) as a storage medium that can be written to and read at any time, with a non-volatile memory such as a ROM (Read Only Memory), and stores application programs necessary to execute various control processes according to one embodiment of the present invention, in addition to middleware such as an OS (Operating System).

[0044] The data storage unit 3 is, for example, a combination of a non-volatile memory such as an HDD or SSD as a storage medium that can be written to and read from at any time, and a volatile memory such as RAM (Random Access Memory), and its storage area includes an input information storage unit 31 and an optimal route information storage unit 32.

[0045] The input information storage unit 31 is used to store input information required for route search processing, which is input from, for example, a management terminal of a telecommunications carrier. The optimum route information storage unit 32 is used to store optimum route information generated by the control unit 1.

[0046] The control unit 1 includes, as processing functions according to an embodiment of the present invention, an input information acquisition processing unit 11, an optimization processing unit 12, an optimal route information generation processing unit 13, and a support information output processing unit 14. These processing units 11 to 14 are all realized by causing a hardware processor in the control unit 1 to execute an application program stored in the program storage unit 2.

[0047] Note that part or all of the processing units 11 to 14 may be realized using hardware such as an LSI (Large Scale Integration) or an ASIC (Application Specific Integrated Circuit).

[0048] The input information acquisition processing unit 11 acquires input information required for route search processing, for example, from a management terminal of a telecommunications carrier, and stores the acquired input information in the input information storage unit 31. In this example, network topology and parameter information are input as input information. The network topology is information that represents the connection relationships on the network of the telecommunications buildings (nodes) to be visited. The parameter information includes the travel time between nodes and the number of users connected to each node. An example of a method for acquiring travel time between nodes will be described in the operation example.

[0049] The optimization processor 12 applies ant colony optimization as an optimization algorithm and matrix calculations to the optimization process, thereby performing route selection and pheromone update processes for multiple ants in parallel to derive optimal routes. The optimization processor 12 also performs a process of calculating the influence of each selected route on the user.

[0050] The optimal route information generation processing unit 13 generates optimal route information including information representing the recommended optimal route and its influence based on the calculation results of the optimization processing unit 12, and stores the generated optimal route information in the optimal route information storage unit 32.

[0051] The support information output processing unit 14 generates support information for the telecommunications carrier or the vehicle driver, including the optimal route information stored in the above-mentioned optimal route information storage unit 32, and transmits the generated support information from the communication I / F unit 4 to the telecommunications carrier's management terminal or in-vehicle terminal to notify the telecommunications carrier or the vehicle driver.

[0052] (Example of operation) FIG. 3 is a flowchart showing an example of the processing procedure and processing contents of the route search processing executed by the control unit 1 of the route search device SV.

[0053] (1) Acquisition of input information For example, if a disaster occurs and a power outage occurs in a communication building under management, the communication carrier inputs, into a management terminal, a network topology representing the connection relationships on the network of each communication building (node) to be restored, location information of each node, and information representing the number of connected users for each node, to the route search device SV.Then, the input network topology, location information of each node, and information representing the number of users are transmitted from the management terminal to the route search device SV via a network (not shown).

[0054] In response to this, in step S10, the control unit 1 of the route search device SV, under the control of the input information acquisition processing unit 11, receives the input information sent from the management terminal via the communication I / F unit 4, and stores the received input information in the input information storage unit 31.

[0055] 4 is a flowchart showing an example of the processing procedure and processing content of the input information acquisition processing unit 11. First, in step S11, the input information acquisition processing unit 11 receives the network topology transmitted from the management terminal and stores it in the input information storage unit 31. Subsequently, in step S12, the input information acquisition processing unit 11 receives the number of users of each node, which is also transmitted from the management terminal, and stores the received number of users in the input information storage unit 31 as one piece of parameter information.

[0056] The input information acquisition processing unit 11 further receives position information, such as latitude and longitude information, of each of the nodes in step S13, and searches for an application in the map data based on the received latitude and longitude information in step S14. Then, in step S15, the input information acquisition processing unit 11 calculates a travel time between nodes that reflects the positional relationship of the nodes based on the map data, and stores the calculated travel time between nodes in the input information storage unit 31 together with the parameter information.

[0057] (2) The planning problem to be optimized In the example of the related art mentioned above, a planning problem was described in which VRP and network flow were simultaneously optimized. However, in one embodiment, the planning problem is not directly implemented because ant colony optimization is applied as the optimization method.

[0058] For example, VRP constraints and visited constraints are not directly implemented because they are naturally satisfied by route selection using ant colony optimization. Also, flow constraints do not need to be directly implemented because the maximum flow algorithm is used to evaluate the selected route.

[0059] On the other hand, the relationship between visits and flows is reimplemented by devising an impact calculation. Also, the objective function in ant colony optimization is an objective function that minimizes the impact.

[0060] (3) Route optimization When the process of acquiring the input information is completed, the control unit 1 of the route search device SV, under the control of the optimization processing unit 12, performs the route optimization process in step S20 as follows.

[0061] 5 is a flowchart showing an example of the processing procedure and processing content of the route optimization processing executed by the optimization processing unit 12. First, the optimization processing unit 12 initializes the value of the pheromone in step S21.

[0062] (3-1) Probabilistic route selection The optimization processing unit 12 then performs probabilistic route selection in step S22. The ants sequentially select the next node to visit according to the pheromone.

[0063] Now, let p be the probability that an ant at node i selects node j. ij and the pheromone is τ ij , the inverse of the distance between nodes is η ij Let Ω be the set of nodes adjacent to node i that have not yet been visited by an ant. The probability that an ant at node i selects node j is p ij teeth,

number

[0064] The optimization processing unit 12 performs probabilistic route selection for ants, similar to the case of using the known roulette wheel selection. That is, with probability p ij The cumulative value of P ij and the random number r generated for each ant a and the point where the magnitude changes is selected as the next node j to be visited.

[0065] FIG. 10 shows an example of this selection process. ij The row of indicates the sequence excluding visited nodes. In this example, P ij >r a The column where the difference is "1" is selected and "2" is output.

[0066] (3-2) Route selection using matrices In step S23, the optimization processing unit 12 performs the probabilistic route selection for the ants described above using a matrix. That is, the optimization processing unit 12 repeats the process of simultaneously selecting routes for, for example, A ants N times. Therefore, the calculation order is O(N) (N is the number of nodes). Incidentally, in the sequential calculation, the pheromone from the current node to the next node is calculated and compared for each of the A ants. Therefore, the calculation order is O(A·N 2 ) and becomes larger.

[0067] FIG. 11 shows an example of a route selection process using this matrix. In this example, the numerator (τ ij ) a (η ij ) β :R N×N , unvisited flag:{0,1} A×N Reflecting this, unvisited molecules are R A×N As a result, the probability of not visiting is [0,1] A×N And the probability of not visiting is [0,1] A×N Random number: [0,1] A By comparing with the ant path, the next destination is selected. At the same time, the unvisited flag is updated. By the above process, the ant path shown in the matrix in Figure 11: N A×N is obtained.

[0068] (3-3) Route evaluation Subsequently, in step S24, the optimization processing unit 12 evaluates the route selected by the ants by calculating the degree of influence that the route selected by the ants has on the user.

[0069] That is, the optimization processing unit 12 calculates in advance the communication volume from each node to the POI for each time point when the mobile power-supply vehicle visits each node. Also, the influence of each failure location occurrence pattern s is defined as f(s).

[0070] In this state, the optimization processor 12 assigns a number based on a power of 2 to each failure location pattern s. For example, if only nodes "2" and "4" are in operation, then 2 2 +2 4 = 20. In addition, the operating status (operating / non-operating) of each node is expressed using binary bits (1 / 0), for example, as [0 0 1 0 1].

[0071] Next, the optimization processing unit 12 calculates the influence from each ant's path. That is, the state at the time of visiting each node is calculated from the cumulative sum of powers of 2. Then, the influence E a An example of this calculation is shown in Figure 12. Thus, the influence is calculated as the number of users who can use the node when the route is selected, or the number of users who cannot use the node.

[0072] (3-4) Pheromone renewal After the above evaluation is completed, the optimization processing unit 12 updates the pheromone in the following manner in step S25.

[0073] That is, the optimization processing unit 12 first reflects the evaluation result for the route selected by the ant in (3-1) in the pheromones on the route.

[0074] Now, the evaporation rate is ρ and the evaluation value of ant is E a , the route that ant a took is T a , 1 is the indicator function, Q is a constant, then pheromone τ ij The update formula is

number

[0075] The optimization processing unit 12 also updates the pheromone using a matrix as follows. That is, the optimization processing unit 12 calculates the edges of the path from the indexes of the visited nodes, and calculates the indicator function 1((i, j)∈T a FIG. 13 shows an example of this calculation.

[0076] (3-5) Repetitive control of optimization process In step S26, optimization processor 12 determines whether the series of processes from route selection to route evaluation and pheromone update has been repeated a predetermined number of times (e.g., the number of nodes N). If the result of this determination is that the series of processes have not been repeated the predetermined number of times, optimization processor 12 returns to step S22 and executes the series of route optimization processes again in steps S22 to S26. Thereafter, optimization processor 12 similarly executes the series of processes repeatedly until the series of processes have been repeated the predetermined number of times.

[0077] On the other hand, when the number of repetitions of the above series of processes reaches a certain number, the optimization processing unit 12 outputs, in step S27, information representing the route, for example, information representing the order of visiting each node, and information representing the evaluation value of the route, for example, the objective function value of the influence, to the optimal route information generation processing unit 13, and ends the optimization processing.

[0078] (4) Generation of optimization information and output of support information When the optimum route information generation processing unit 13 receives information representing a route and its evaluation value from the optimization processing unit 12, in step S30 shown in Figure 3, it associates the received route and evaluation value with, for example, the administrator information of the requester, and stores this as optimum route information in the optimum route information storage unit 32.

[0079] In this case, the route with the highest evaluation value may be selected unconditionally as the optimal route information, or all routes with evaluation values ​​above a threshold value, or a certain number of the top routes selected from each of the routes, each including its evaluation value, may be included as the optimal route information.

[0080] Once the optimum route information is obtained, the control unit 1 of the route search device SV, under the control of the support information output processing unit 14, reads the optimum route information from the optimum route information storage unit 32 in step S40, generates support information for disaster recovery based on the read optimum route information, and transmits the generated support information from the communication I / F unit 4 to the management terminal that made the request or the on-board terminal of the mobile power supply vehicle.

[0081] As support information, the above-mentioned optimum route information may be transmitted as is, or alternatively, information displaying the route on map data may be generated and transmitted, in which case the travel distance or required time between nodes may be added. Furthermore, traffic information on the route may be acquired and reflected in the above-mentioned route.

[0082] (Actions and Effects) As described above, in one embodiment, ant colony optimization is applied as a route optimization method, and matrix calculations are applied to the route optimization process using this ant colony optimization. Then, by simultaneously selecting multiple virtual ant routes and repeating the evaluation and pheromone update process a certain number of times, an optimal route and its influence are searched for.

[0083] Therefore, according to one embodiment, the following effect can be achieved. In general, optimization processing using the branch and bound method requires a lot of time because the amount of calculation is exponential. In contrast, ant colony optimization requires a polynomial time amount of calculation, so the calculation time is significantly reduced. Therefore, according to one embodiment, it is possible to reduce the calculation cost.

[0084] In one embodiment, a series of processes from route selection by the Ant Colony Optimization to its evaluation and pheromone update are performed in parallel for multiple ants using matrix calculations. This reduces the computation order to O(N) (where N is the number of nodes). In contrast, in sequential calculations, the pheromone from the current node to the next node is sequentially calculated and compared for each of the multiple ants, so the computation order is O(A·N 2 ) and becomes larger.

[0085] [Other embodiments] In one embodiment, a series of processes from the process of acquiring input information to the process of outputting support information is executed on a single server device. However, all or part of the series of processes from the process of acquiring input information to the process of outputting support information may be distributed and processed by multiple server devices or information processing terminals such as a management terminal.

[0086] In addition, in one embodiment, a case where mobile power supply vehicles are dispatched to multiple communication buildings in the event of a disaster has been described as an example, but the present invention is not limited to this and can be applied to the supply of goods, and can also be applied as a support system for delivering goods to multiple stores during normal times, for example.

[0087] In addition, the processing procedures and processing contents of each processing unit provided in the route search device, the number of ants selected simultaneously, the configuration of the optimal route information and support information, etc. can be modified and implemented in various ways within the scope of the gist of this invention.

[0088] Although the embodiments of the present invention have been described in detail above, the above description is merely an example of the present invention in every respect. It goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. In other words, when implementing the present invention, specific configurations according to the embodiments may be appropriately adopted.

[0089] In short, this invention is not limited to the above-described embodiments, and in the implementation stage, the components can be modified and embodied without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiments. For example, some components may be omitted from all the components shown in the embodiments. Furthermore, components from different embodiments may be appropriately combined. [Explanation of symbols]

[0090] SV: Route search device 1...Control unit 2...Program memory section 3...Data storage unit 4...Communication I / F section 5...Bus 11...Input information acquisition processing unit 12...Optimization processing section 13...Optimal route information generation processing unit 14...Support information output processing unit 31...Input information storage unit 32...Optimal route information storage unit

Claims

1. A route search device that searches for a route for visiting a plurality of geographically scattered nodes, a first processing unit that acquires topology information that indicates a connection relationship between the plurality of nodes and parameter information that includes the number of users that use each of the plurality of nodes and travel time between the nodes; a second processing unit that applies ant colony optimization as a method for optimizing the routes based on the topology information and the parameter information, and uses matrix calculations in the process of optimizing the routes using the ant colony optimization, and that finds an optimal route from among the routes by repeatedly executing a predetermined number of processes in parallel using the matrix calculations: selecting the route according to pheromone values ​​of a plurality of virtual agents; evaluating the selected route; and updating the pheromone value to reflect the results of the evaluation; a third processing unit that generates and outputs optimal route information including the optimal route and the corresponding evaluation result; A route search device comprising:

2. The route search device according to claim 1 , wherein the second processing unit performs the evaluation by calculating a degree of influence on a user who uses the plurality of nodes.

3. The route search device according to claim 2 , wherein the degree of influence is determined by calculating the number of users who will be able to use the node or the number of users who will not be able to use the node when the route is selected.

4. 2. The path search device according to claim 1, wherein the second processing unit performs a process of selecting a second node to be visited next from a first node based on a probability calculated based on the pheromone value set between the first node and the second node, the inverse of the distance between the first node and the second node, and a set of unvisited nodes.

5. 2. The route search device according to claim 1, wherein the second processing unit updates the pheromone value by adding a value obtained by reflecting an evaporation rate to the pheromone value before updating to the result of the evaluation corresponding to the selected route.

6. A route search method executed by an information processing device for searching for a route for visiting a plurality of geographically scattered nodes, comprising: acquiring topology information representing a connection relationship between the plurality of nodes and parameter information including the number of users using each of the plurality of nodes and travel time between the nodes; a step of applying ant colony optimization as a method for optimizing the route based on the topology information and the parameter information, using matrix calculations in the route optimization process using the ant colony optimization, and repeatedly executing a predetermined number of steps in parallel using the matrix calculations to select the route according to pheromone values ​​of a plurality of virtual agents, evaluate the selected route, and update the pheromone value to reflect the results of the evaluation, thereby finding an optimal route from among the routes; generating and outputting optimal route information including the optimal route and the corresponding evaluation result; A route search method comprising:

7. 6. A program for causing a processor included in a route search device to execute at least one of the processes performed by first to third processing units included in the route search device according to claim 1.

Images