JP6280082B2 - Route selection device, route selection method, and program - Google Patents

Description

  The present invention relates to a route selection device, a route selection method, and a program.

  In Non-Patent Documents 1 and 2, evaluation is performed on an evaluation scale related to facilities constituting a network based on a disaster area shape or a disaster parameter. Based on the evaluation result, it is possible to evaluate the characteristics at the time of the disaster for the facilities constituting the current network. In addition, for example, it is possible to conduct an impact assessment on whether the evaluation scale value will improve when the equipment that makes up the network is changed to a new equipment. It can be evaluated whether or not the change is advantageous.

Hiroshi Saito, “Network” and “Shape” (Part 2) -Shape of Network Resistant to Surface Damage-, Operations Research Society of Japan, 2014 Autumn Research Conference, August 2014 Hiroshi Saito, Network survivability in face of disaster, Operations Research Society of Japan, 2014 Autumn Research Conference, August 2014 Tsukasa, Yodogawa, distance attenuation formula of maximum acceleration and maximum speed considering fault type and ground conditions, Architectural Institute of Japan, 523, 1999.

  However, the above-mentioned non-patent documents do not give any suggestions as to what route should be selected when it is desired to reduce the influence of an earthquake that is expected to occur as much as possible for the facility to be constructed.

  The present invention has been made in view of the above points, and an object of the present invention is to support selection of a route having a relatively low influence of a disaster on a facility construction destination.

Therefore, in order to solve the above-described problem, the route selection device includes, for each adjacent node among a plurality of nodes constituting the network, a first candidate route for each candidate route included in the set of candidate routes between the adjacent nodes in the network An index value and a second index value indicating the magnitude of the effect of the earthquake are calculated for each predetermined earthquake that is assumed, and the first index value and the second index value for each earthquake are calculated. Based on the extraction unit that extracts a subset from the set and the candidate route combination group generated by selecting one candidate route from each of the subsets, included in each combination The sum of the first index values of each candidate route and the candidate route included in each combination between the nodes are used for each node that is not necessarily adjacent in the network. And a third index value of which is calculated, and a selection unit for selecting one combination based on.

  It is possible to support the selection of a route having a relatively low influence by a disaster at a facility construction destination.

It is a figure which shows the hardware structural example of the route selection apparatus in embodiment of this invention. It is a figure for demonstrating a candidate path | route set. It is a figure which shows the function structural example of the route selection apparatus in embodiment of this invention. It is a flowchart for demonstrating an example of the process sequence of a route selection process. It is a figure which shows an example of the grid of a grid between adjacent nodes. It is a figure which shows an example of the search result of the path | route which passes along the road network or railway network of the vicinity of a certain lattice point. It is a flowchart for demonstrating an example of the process sequence of the calculation process of the cutting probability regarding one earthquake of one candidate path | route.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating a hardware configuration example of a route selection device according to an embodiment of the present invention. The path selection device 10 in FIG. 1 includes a drive device 100, an auxiliary storage device 102, a memory device 103, a CPU 104, an interface device 105, and the like that are mutually connected by a bus B.

  A program that realizes processing in the route selection apparatus 10 is provided by a recording medium 101 such as a CD-ROM. When the recording medium 101 storing the program is set in the drive device 100, the program is installed from the recording medium 101 to the auxiliary storage device 102 via the drive device 100. However, the program need not be installed from the recording medium 101 and may be downloaded from another computer via a network. The auxiliary storage device 102 stores the installed program and also stores necessary files and data.

  The memory device 103 reads the program from the auxiliary storage device 102 and stores it when there is an instruction to start the program. The CPU 104 executes a function related to the route selection device 10 according to a program stored in the memory device 103. The interface device 105 is used as an interface for connecting to a network.

In the present embodiment, the route selection apparatus 10 uses a physical route (hereinafter referred to as a “traffic route”) that makes the influence of an earthquake relatively low with respect to equipment that constitutes a network passing through a predetermined node. .) Is selected (hereinafter referred to as “route selection process”). For example, the equipment constituting the network is equipment including a network cable and a device (router or the like) for connecting the network cables. The node A, facility device for connecting a network cable is accommodated (Building), and the like. However, a device that connects a network cable may be grasped as a node. Further, the physical route refers to a series of geographical positions (that is, routes in which the geographical position is specified) where a network cable or the like is laid or constructed. In addition, the component of the equipment which comprises a network, such as a network cable and a router, is hereafter called "equipment element".

  In this embodiment, the candidate route set for adjacent nodes i and j is R (i, j), the shortest route among the candidate routes is Wmin (i, j), and the route length of the shortest route is Lmin (i, j ), One candidate route in the candidate route set R (i, j) is W (i, j), the route length of the candidate route W (i, j) is L (W (i, j)), or The probability that the candidate route W (i, j) is disconnected due to the influence of the equipment failure due to the earthquake a (where a indicates an assumed earthquake name) is denoted as d_a (W (i, j)). Adjacent nodes i and j are two nodes with no other nodes between them. The candidate route set between the adjacent nodes i and j is a set of routes searched as candidates for physical routes connecting the node i and the node j.

  FIG. 2 is a diagram for explaining a candidate route set. FIG. 2 shows that there are three candidate routes between node i and node j. Each candidate route is searched so as to pass on a geographical element that can be a construction destination of facilities constituting the network, such as a road network and a railway network.

  In the following, the former is referred to as an “overall route” in order to avoid confusion between the entire physical route of the construction target network and the physical route for each candidate route W (i, j).

In the present embodiment, the probability that the node u and the node v are disconnected due to the equipment failure due to the earthquake a (hereinafter referred to as “departure / departure”) for the communication of the two nodes u and v that are not necessarily adjacent to each other. Is expressed as d_a (u, v). The maximum probability of disconnection between landing points when all candidate routes W (i, j) are given to all earthquakes between all adjacent nodes is referred to as the worst probability of disconnection between landing points d_A (W). That is, for the set S of departure / arrival point sets, the set A of assumed earthquakes
d_A (W) = max_ {a∈A, (u, v) ∈S} d_a (u, v) (1)
It is.

  The disconnection probability d_a (u, v) between the arrival and departure points for a certain earthquake a and the arrival and departure points u and v becomes the worst disconnection probability d_A (W) between the arrival and departure points. These landing points u and v are called the worst landing points, and this earthquake a is called the worst earthquake.

In the present embodiment, the disconnection probability and the worst disconnection probability d_A (W) between arrival and departure points are examples of an index value or a scale indicating the magnitude of the influence of the earthquake on the facilities constituting the network. The simple “cut probability” means the cut probability between adjacent nodes. In other words, in the present embodiment, a physical path that minimizes the worst disconnection probability d_A (W) between the arrival and departure points is selected for the entire target network. As a result, it is possible to support the selection of a route having a relatively low influence by the disaster for the facility construction destination. Note that disconnection between node u and node v means that network communication between node u and node v is difficult due to damage or failure of any part of the equipment that makes up the network between node u and node v. The situation that becomes.

Therefore, the problem to be solved in the present embodiment is the following optimization problem using the worst disconnection probability d_A (W) between landing points.
min_ {W (i, j) ∈R (i, j), (i, j) ∈E} d_A (W) (2)
subj.Σ _ {(i, j) ∈E} L (W (i, j)) ≦ (1 + c) Σ _ {(i, j) ∈E} Lmin (i, j) (3)
Here, E is a set of adjacent node pairs, and Equation (3) means a condition that the sum of physical paths between adjacent nodes (hereinafter referred to as “total path length”) is equal to or less than the allowable cost c. To do. The allowable cost c is a constant that determines the upper limit of the route length of the entire route (hereinafter referred to as “total route length”) arbitrarily set in the network design, and is a value larger than 0.0, for example. In the present embodiment, the allowable cost c is determined based on the sum of the shortest path lengths between adjacent nodes.

  That is, in the present embodiment, among the combinations of candidate routes W (i, j) whose total route length is equal to or less than the allowable cost c, the combination that minimizes the disconnection probability d_A (W) between the worst departure and arrival points is It is selected as the optimum overall route (hereinafter referred to as “optimum route”).

In addition, the following formula | equation (4) and formula (3) which replaced the role of said formula | equation (2) and formula (3) may be solved.
minΣ _ {(i, j) ∈E} L (W (i, j)) (4)
subj.d_A (W) ≦ c '(5)
In order to solve the above optimization problem, the route selection device 10 has a functional configuration as shown in FIG. FIG. 3 is a diagram illustrating a functional configuration example of the route selection device according to the embodiment of the present invention. In FIG. 3, the route selection apparatus 10 includes a problem decomposition unit 11, a multipurpose problem solving unit 12, a cutting probability calculation unit 13, a combination unit 14, and the like. Each of these units is realized by processing that one or more programs installed in the route selection apparatus 10 cause the CPU 104 to execute. The route selection device 10 also uses a connection information storage unit 121, a geographic information storage unit 122, an equipment information storage unit 123, an earthquake information storage unit 124, and the like. The connection information storage unit 121, the geographic information storage unit 122, the facility information storage unit 123, and the earthquake information storage unit 124 are, for example, a storage that can be connected to the auxiliary storage device 102 or the route selection device 10 in FIG. It can be realized using an apparatus or the like.

  The problem resolution unit 11 converts the optimization problem defined by the equations (2) and (3) (hereinafter referred to as “parent problem”) into an optimization problem (hereinafter referred to as “children”) for each pair of adjacent nodes i and j. It is decomposed into "Problem C (i, j)"). Specifically, the problem decomposition unit 11 generates a multi-purpose optimization problem represented by the following equation (6) for each child problem C (i, j).

min_ {W (i, j) ∈R (i, j)} (L (W (i, j)), d_a1 (W (i, j)),…, d_an (W (i, j)))) (6)
Here, A = {a1, ..., an}. That is, the earthquakes to be examined are a1, ..., an.

  That is, the child problem C (i, j) is an optimization problem for the purpose of minimizing the path length and minimizing the cutting probability for each earthquake a for the physical path between the adjacent nodes i, j.

Each objective function of the child problem C (i, j) is part of the objective function or condition of the parent problem. Therefore, the optimal solution of the parent problem is a combination of Pareto optimal solutions selected one by one from each Pareto optimal solution set obtained for each child problem C (i, j). The Pareto optimal solution is a solution that does not have a solution that improves all objective function values for a problem having a plurality of purposes . For example, among candidate route sets R (i, j) of adjacent nodes i, j, L (W (i, j)), d_a1 (W (i, j)),..., D_an (W (i, j) A candidate path W (i, j) in which any of the above is the smallest, or a candidate path W (i, j) whose result of combining these objective functions is relatively small is included in the Pareto optimal solution set. Therefore, the Pareto optimal solution set of the child problem C (i, j) is a subset of the candidate route set R (i, j) between the adjacent nodes i, j, where the shortest route Wmin (i, j) is , L (W (i, j)) is minimized, so that it is included in the Pareto optimal solution set of the child problem C (i, j).

  Information indicating the connection relationship between each node (which node is adjacent to which node) is stored in the connection information storage unit 121. Information about each earthquake a belonging to the earthquake set A is stored in the earthquake information storage unit 124. For example, the earthquake information storage unit 124 stores the probability of occurrence of each earthquake a, the epicenter area, the depth (the depth of the epicenter), and the epicenter magnitude, and further calculates the assumed seismic intensity of each area. An amplification factor based on the evaluation formula and the structure of the ground surface is stored. Alternatively, the earthquake information storage unit 124 may store the estimated seismic intensity of each region in advance.

  The multipurpose problem solving unit 12 executes processing for obtaining a Pareto optimal solution set of Expression (6) for each child problem C (i, j).

  The cutting probability calculation unit 13 calculates a cutting probability d_a (W (i, j)) for each earthquake a for each candidate route W (i, j) for each adjacent node i, j. Each calculated d_a (W (i, j)) is used by the multipurpose problem solving unit 12.

  The combination unit 14 selects a Pareto optimal solution (candidate route W (i, j)) one by one from each Pareto optimal solution set for each child problem C (i, j), and generates a candidate route W ( For each combination of i, j), the worst probability of disconnection between arrival and departure points d_A (W) and the total route length are evaluated to obtain a solution to the parent problem.

  The geographic information storage unit 122 stores geographical information such as a road network and a traffic network in a region where a facility constituting the network is constructed. The facility information storage unit 123 stores information related to facilities that make up the network. For example, the equipment information storage unit 123 stores information on equipment that configures or intends to configure a physical route, such as a pipe type.

  Hereinafter, a processing procedure executed by the route selection device 10 will be described. FIG. 4 is a flowchart for explaining an example of a processing procedure of route selection processing.

  In step S101, the decomposition unit identifies each adjacent node (i, j) based on the information stored in the connection information storage unit 121, and for each adjacent node (i, j), the child problem C (i, j j) is generated. That is, Expression (6) is generated for each adjacent node (i, j).

  The following steps S102 to S105 are executed for each child problem C (i, j).

  In step S102, the multipurpose problem solving unit 12 generates each candidate route W (i, j) that satisfies W (i, j) ∈R (i, j) for the child problem C (i, j) to be processed. (Step S102). The candidate route set R (i, j) is configured by geographical elements that can physically install a network such as a road network and a railway network. Each candidate route W (i, j) may be generated as follows, for example.

  First, the multipurpose problem solving unit 12 generates grids (lattices) at appropriate intervals, centering on an intermediate point between the nodes i and j.

  FIG. 5 is a diagram illustrating an example of a grid pattern between adjacent nodes. The distance between the grids depends on the distance between adjacent nodes i and j. For example, if node i is a certain point in Tokyo and node j is a certain point in Yokohama, the distance is It may be about 5 km. Further, one direction orthogonal to the grid eyes may be the north-south direction, and the other direction may be the east-west direction, or the other directions may be orthogonal. In addition, the geographical position information (for example, latitude and longitude) of each node is stored in the facility information storage unit 123, for example.

  The multi-objective problem solving unit 12 uses the adjacent node i as a constraint condition that each grid point (intersection of orthogonal lines) on the grid passes through one point on the road network or railway network in the vicinity of the grid point. , j is searched for a route. The searched route is a route configured by a road network or a railroad network. This is because the facilities that make up the network are generally installed along roads and railways. The geographical information of the road network and the railway network is stored in the geographical information storage unit 122, for example.

  FIG. 6 is a diagram illustrating an example of a search result of a route passing through a road network or a railway network in the vicinity of a certain grid point. FIG. 6 shows an example of a route search result when the grid point P1 is a processing target and one point passing on the road network or the railroad network is P2 in the vicinity of the grid point P1. The search result is one candidate route W (i, j). For the route search, for example, a route search technique in a navigation system or the like may be used.

  By executing such route search for each lattice point, a plurality of search results, that is, a plurality of candidate routes W (i, j) are generated. Since each candidate route W (i, j) is searched based on mutually different constraint conditions, it can be expected that the candidate routes W (i, j) are separated from each other to some extent. This means that candidate paths W (i, j) having different influences on an earthquake whose epicenter is a certain point can be searched for. The route search may be performed under a constraint condition based on two or more grid points. Further, the candidate route W (i, j) may be generated by a method other than the above.

  Note that the grid can be generated indefinitely, but actually, the value of the candidate route W (i, j) deviating from the shortest route Wmin (i, j) is hardly recognized. This is due to the following reason.

  An equipment failure due to an earthquake is less likely to occur at a seismic intensity of 5 or less, for example. Therefore, even if equipment failure on the route is avoided by a route far from the epicenter, if a detour is made to an area having a seismic intensity of 5 or less, the need for searching for a route farther than that is low. The region with a seismic intensity of 5 or higher is a limited range, so it does not spread indefinitely. Further, even when a region having a seismic intensity of 5 or more is very wide due to a huge earthquake, for example, the region to be targeted for the candidate route W (i, j) is further limited for the following reason.

  The cutting probability d_a (W (i, j)) due to an earthquake a is given as a positive weighted sum of the path lengths in the area of each seismic intensity scale of this earthquake a as a first approximation. Since the disconnection probability d_a (Wmin (i, j)) of the shortest path Wmin (i, j) is the same, in order to satisfy d_a (W (i, j)) <d_a (Wmin (i, j)) The path length L (W (i, j)) cannot take a very large value (in order that the disconnection probability becomes smaller than the shortest path).

  Note that the candidate route W (i, j) between the adjacent nodes i and j may be given as input information. In this case, for each adjacent node i, j, information indicating the candidate path W (i, j) of the adjacent node i, j may be input at the start of FIG. 4, or the auxiliary storage device 102 or the like in advance. May be stored.

  Returning to FIG. Subsequent to step S102, the multipurpose problem solving unit 12 calculates a path length L (W (i, j)) for each candidate path W (i, j) (step S103). The route length L (W (i, j)) may be calculated together with the search for the candidate route W (i, j).

  Subsequently, the cutting probability calculation unit 13 calculates a cutting probability d_a (W (i, j)) for each earthquake a included in the earthquake set A for each candidate route W (i, j) (step S104). That is, d_a1 (W (i, j)),..., D_an (W (i, j)) is calculated for each candidate route W (i, j). The details of the method for calculating the cutting probability d_a (W (i, j)) will be described later.

  Subsequently, the multipurpose problem solving unit 12 is based on each path length L (W (i, j)) calculated in step S102 and each cutting probability d_a (W (i, j)) calculated in step S103. Thus, by solving Equation (6), a Pareto optimal solution set for the child problem C (i, j) to be processed is obtained (step S105).

  The above steps S102 to S105 are executed for each child problem C (i, j), so that the Pareto optimal solution is obtained for each child problem C (i, j) (that is, for each pair of adjacent nodes i, j). A set is obtained.

  Subsequently, the combination unit 14 selects one element (hereinafter referred to as a candidate route W * (i, j)) from each of the Pareto optimal solution sets for each child problem C (i, j). All combinations of candidate routes W * (i, j) generated by the above are generated (step S106).

  Subsequently, the combination unit 14 calculates, for each combination, the total route length (that is, the total route length) of each W * (i, j) belonging to the combination (step S107).

  Subsequently, the combination unit 14 arranges the combinations in ascending order of the total path length (step S108). That is, the alignment is performed so that the combination having the shortest total path length is the head.

  Subsequently, the combination unit 14 substitutes 1 for the variable k, substitutes the constant MAX for the variable m, and substitutes 0 for the variable Rmin (step S109). The variable k is a variable for identifying a combination to be processed. The variable m is a variable for storing the minimum value of the disconnection probability d_A (W) between the worst landing points. The constant MAX is a theoretical maximum value of the worst disconnection probability d_A (W) between landing points. The variable Rmin is a variable for storing the order of combinations (value of k) corresponding to the minimum value of the disconnection probability d_A (W) between the worst departure and arrival points.

  Subsequently, the combination unit 14 determines whether or not the total route length (k) of the k-th combination (hereinafter referred to as “combination (k)”) is equal to or less than the allowable cost c (step S110). The allowable cost c is (1 + c) Σ _ {(i, j) εE} Lmin (i, j). That is, step S110 is processing corresponding to the condition of Expression (3).

  When the total path length (k) is equal to or less than the allowable cost c (Yes in step S110), the combination unit 14 uses the worst-case disconnection point disconnection probability d_A (W) of the combination (k) based on the equation (1). (Step S111). That is, based on the disconnection probability d_a (W (i, j)) for each earthquake a for each candidate route W * (i, j) belonging to the combination (k), the worst disconnection probability d_A (W) between arrival and departure points is Calculated.

  Subsequently, the combination unit 14 determines whether or not the calculated worst probability of disconnection between arrival and departure points d_A (W) is less than the variable m (step S112). When the worst disconnection probability d_A (W) between the landing points is less than the variable m (Yes in step S112), the combination unit 14 substitutes the worst disconnection probability d_A (W) between the landing points into the variable m, and the variable The value of k is substituted for variable Rmin (step S113). If the worst disconnection point d_A (W) is greater than or equal to the variable m (No in step S112), step S113 is not executed.

  Subsequently, the combination unit 14 determines whether or not the combination (k) is the last combination (step S114). When the combination (k) is not the last combination (No in Step S114), the combination unit 14 adds 1 to the variable k (Step S115), and repeats Step S110 and subsequent steps. When the combination (k) is the last combination (Yes in step S114), the combination unit 14 sets the entire route constituted by the candidate route W * (i, j) group belonging to the Rmin-th combination as the optimum route. Select (step S117). That is, the optimal route is a solution of Equation (2), which is a parent problem.

  If the total path length (k) of the combination (k) exceeds the allowable cost (Yes in S110), the combination unit 14 determines whether or not the value of the variable k is greater than 1 (step S116). When the value of the variable k is larger than 1 (Yes in Step S116), the combination unit 14 executes Step S117. When the value of the variable k is 1 or less (No in step S116), the combination unit 14 gives up the selection of the optimum route. This is because there is no entire route having a total route length equal to or less than the allowable cost c.

  Each Pareto optimal solution set for each child problem C (i, j) is composed of, for example, 10 candidate paths W (i, j), and the number of adjacent node i, j pairs is 10 in total. In this case, all combinations are 10 to the 10th power, and the amount of calculation becomes enormous as it is difficult to obtain a solution in practice. However, in practice, the amount of calculation is not so large. This is because, in many cases, the Pareto optimal solution set of each child problem C (i, j) is composed only of the shortest path Wmin (i, j), and even when there are other Pareto optimal solutions. This is because the number of elements of the Pareto optimal solution set is likely to be very small. The reason is as follows.

  In the case of a huge earthquake, the area of the same seismic intensity level is large. In a route passing through the same seismic intensity scale, approximately, the shorter the route length, the more advantageous (the cut probability is small). Only node i or node j, or a candidate route W (i, j) that can be detoured to a low seismic intensity scale region in the middle of the route can be a Pareto optimal solution for C (i, j). In the case of a local earthquake, there can be a Pareto optimal solution other than the shortest path Wmin (i, j) in the region near the epicenter, but if it is a little away from the region near the seismic region, it immediately becomes a low seismic intensity region. No failure occurs and only the shortest path Wmin (i, j) is the Pareto optimal solution. As a result, for example, for each adjacent node i, j, the Pareto optimal solution set of each child problem C (i, j) is composed of about two candidate paths W (i, j) on average. Become. In this case, if there are a total of 10 between adjacent nodes, the number of combinations is 2 to the 10th power, and the amount of calculation is a size that can be calculated at this time.

  Next, details of step S104 will be described. FIG. 7 is a flowchart for explaining an example of a processing procedure of a cutting probability calculation process for one earthquake on one candidate route. Since FIG. 7 shows the cutting probability calculation processing for one earthquake a of one candidate route W (i, j), in step S104, the processing of FIG. 7 is performed for each candidate route (i, j) and each earthquake. It may be executed for a∈A.

  In step S <b> 201, the cutting probability calculation unit 13 divides the candidate route W (i, j) into a plurality of parts (hereinafter referred to as “microsection u”).

  Subsequently, the cutting probability calculation unit 13 calculates the distance d from the focal region of the earthquake a for each minute section u (step S202). For example, the distance d from the point closest to the minute section u to the minute section u may be calculated for each minute section u within the range of the epicenter. Information indicating the geographical position of the epicenter of earthquake a is stored in the earthquake information storage unit 124.

  Subsequently, the cutting probability calculation unit 13 calculates the maximum earthquake speed on the engineering basis for each minute section u (step S203). In step S203, a calculation formula for the distance attenuation of the maximum earthquake speed, a correction method for the calculation formula, a conversion method from the maximum speed of the hard ground to the maximum speed of the engineering base, and the like are used. For example, http: //dil-opac.bosai.go can be used to calculate the distance attenuation of the maximum earthquake velocity, how to correct the equation, and how to convert the maximum velocity of the hard ground to the maximum velocity of the engineering foundation. It is shown in .jp / publication / nied_tech_note / pdf / n379_1.pdf (Materials for Disaster Prevention Science), and what is described in Non-Patent Document 1 may be referred to. Moreover, the calculation method described in http://www.giroj.or.jp/disclosure/q_kenkyu/No20_7.pdf may be referred to. Further, these calculation formulas may be stored in the earthquake information storage unit 124.

  Specifically, the distance d, the source magnitude and depth of the earthquake a (the depth of the source) are substituted into the formula for calculating the distance attenuation of the maximum velocity of the earthquake, so that Maximum speed is calculated. At this time, the calculation formula may be corrected by the calculation formula correction method. Subsequently, based on the conversion method from the maximum speed of the hard ground to the maximum speed of the engineering base, the maximum speed of the hard ground in the micro section u is converted to the maximum speed in the engineering base of the micro section u. . Such an operation is executed for each minute interval u.

  Subsequently, the cutting probability calculation unit 13 acquires a surface layer ground amplification factor corresponding to the position of the minute section u for each minute section u, and obtains the surface layer at the maximum speed of the engineering base calculated for the minute section u. By multiplying by the ground amplification factor, the maximum speed on the surface layer ground (ground surface) of the minute section u is obtained (step S204). The surface ground amplification factor refers to the amplification factor of the maximum speed from the engineering base (Vs = 400 m / s) to the ground surface, and the value varies depending on the location. The surface ground amplification factor may be downloaded from, for example, http://www.j-shis.bosai.go.jp/, or may be stored in the earthquake information storage unit 124. By using a different surface layer amplification factor for each location, the maximum velocity (and hence the assumed seismic intensity) can be calculated in consideration of the geology of each micro section u and the influence of the ground.

  Subsequently, the cutting probability calculation unit 13 calculates the assumed seismic intensity for each micro section u by applying the maximum speed on the surface ground calculated for the micro section u to the conversion formula from the maximum speed to the seismic intensity. (Step S205). For example, http://dil-opac.bosai.go.jp/publication/nied_tech_note/pdf/n379_1.pdf (Materials for Disaster Prevention Science) may be referred to for the conversion formula from maximum speed to seismic intensity. . Non-patent document 3 may be referred to.

  The earthquake information storage unit 124 may store in advance the assumed seismic intensity of each region for each earthquake a. In this case, as the assumed seismic intensity of each minute section u, a value stored in the earthquake information storage unit 124 for the area to which the minute section u belongs may be used.

  Subsequently, the cutting probability calculation unit 13 calculates the failure probability p (u) of each minute section when the earthquake a occurs (step S206). Specifically, the failure probability p (u) of each micro section u is calculated based on the assumed seismic intensity of each micro section u and the path failure rate for each seismic intensity stored in the facility information storage unit 123. The The path failure rate is a ratio of the length of the failure portion per reference length. For example, when there is one failure portion per 100 m, the path failure rate is 0.01 [1 / m]. For example, the cutting probability calculation unit 13 calculates the failure probability p (u) of the minute section u by multiplying the path failure rate corresponding to the assumed seismic intensity by the length of the minute section u.

  Note that the path failure rate may vary from region to region. Moreover, the attribute (for example, pipe line or aerial, a vinyl pipe or a steel pipe, a manufacture year etc.) corresponding to the micro area u may be considered. For example, a path failure rate for each seismic intensity may be stored in the facility information storage unit 123 for each attribute of the facility element. In this case, even if the failure probability p (u) of each micro section u is calculated using the attribute of the facility element corresponding to each micro section u and the path failure rate corresponding to the assumed seismic intensity of each micro section u. Good.

  Subsequently, the cutting probability calculation unit 13 calculates the cutting probability d_a (i, j) of the candidate route W (i, j) (step S207). Specifically, the cutting probability calculation unit 13 calculates the operation probability for each minute section u based on the following equation (7).

Operation probability of minute section u = 1−failure probability p (u) of the minute section u (7)
The disconnection probability calculating unit 13 also calculates the product of the operation probabilities (1- (1-p (u_1)) ×... × (1-p (1) of all the minute sections u related to the candidate route W (i, j). u_n)) However, u_k represents the kth minute section u)). The calculation result of the product is the operation probability of the candidate route W (i, j). Therefore, the cutting probability calculation unit 13 calculates the cutting probability d_a (i, j) based on the following equation (8).

Disconnection probability d_a (W (i, j)) = 1−Operation probability of candidate route W (i, j) (8)
Note that the length of the minute interval u is preferably determined so that the failure probability p (u) is 1.0 or less. That is, when the path failure rate is 0.01 and the reference length is 100, if the length of the minute section u exceeds 10 km, the failure probability p (u) exceeds 1.0. In order to prevent such a situation, the length of the minute interval u is preferably determined. For example, the length of each minute section u may be a value shorter than the reference length of the path failure rate.

  In addition, as an index value or scale indicating the magnitude of the earthquake's impact on the facilities that make up the network, the expected value based on the earthquake occurrence probability of the sum of the cut-off probability between arrival and departure points d_A (W) May be used. The expected value based on the earthquake occurrence probability of the sum of the disconnection probabilities between the departure and arrival points is given by Σ_ {a∈A, (u, v) ∈S} p_a · d_a (u, v). Here, p_a is the probability that an earthquake a will occur (within a certain period in the future), and is stored in the earthquake information storage unit 124. As p_a, for example, what is published for each earthquake at http://www.j-shis.bosai.go.jp/ may be used. When the expected value based on the earthquake occurrence probability is used instead of the worst-case disconnection point disconnection probability d_A (W), the worst-case disconnection point disconnection probability d_A (W) above is What is necessary is just to be replaced with the expected value based on the probability of earthquake occurrence in the total probability. Therefore, in this case, the route that minimizes the expected value based on the earthquake occurrence probability of the total probability of disconnection between the departure and arrival points is selected as the optimum route.

  As described above, according to the present embodiment, it is possible to support selection of a route that is relatively less affected by a disaster with respect to the construction destinations of facilities constituting the network. In the present embodiment, since the optimal route is selected in consideration of the disconnection probability between arbitrary departure and arrival points, a route that does not disrupt communication between the arrival and departure points by securing a plurality of routes is the optimum route. The possibility of being selected can be increased.

  Further, according to the present embodiment, after the parent problem is decomposed into child problems C (i, j) and the Pareto optimal solution set is obtained for each child problem C (i, j), the solution of the parent problem is Is obtained. In many cases, the Pareto optimal solution set obtained for each child problem C (i, j) is actually only the shortest path Wmin (i, j) or only the shortest path Wmin (i, j) + α. Consists of. As a result, the calculation amount can be greatly reduced. Therefore, even for a large-scale network with a large number of nodes, it is possible to select a path with a relatively low influence by a disaster with a realistic calculation amount or a limited calculation amount.

In the present embodiment, the multipurpose problem solving unit 12 is an example of an extracting unit. The combination unit 14 is an example of a selection unit. The path length is an example of a first index value. The allowable cost (c in Expression (3)) is an example of a first threshold value. C ′ in Expression 5 is an example of a second threshold value. The cutting probability is an example of a second index value that indicates the magnitude of the influence of the earthquake. The worst-case disconnection point disconnection probability d_A (W) is an example of a third index value for each non-adjacent node and for each earthquake. The expected value based on the earthquake occurrence probability of the sum of the disconnection probabilities between the departure and arrival points is also an example of a third index value using a weighted sum that weights the occurrence probability of each earthquake between nodes that are not adjacent and for each earthquake. C ′ in the case where the sum of the probability of disconnection between departure and arrival points is applied to Equation (5) is an example of a third threshold value.

  As mentioned above, although the Example of this invention was explained in full detail, this invention is not limited to such specific embodiment, In the range of the summary of this invention described in the claim, various deformation | transformation・ Change is possible.

DESCRIPTION OF SYMBOLS 10 Path selection apparatus 11 Problem decomposition part 12 Multi-objective problem solution part 13 Cutting probability calculation part 14 Combining part 100 Drive apparatus 101 Recording medium 102 Auxiliary storage apparatus 103 Memory apparatus 104 CPU
105 Interface Device 121 Connection Information Storage Unit 122 Geographic Information Storage Unit 123 Equipment Information Storage Unit 124 Earthquake Information Storage Unit B Bus

Claims (8)

For each adjacent node of the plurality of nodes constituting the network, for each candidate route included in the set of candidate routes between the adjacent nodes in the network , a route length that is a first index value and a predetermined predetermined value A second index value indicating the magnitude of the impact of the earthquake for each earthquake, and from the set based on the first index value and the second index value for each earthquake An extractor for extracting a subset ;
In the network, a sum of the first index values of each candidate route included in each combination is generated from a combination group of candidate routes generated by selecting one candidate route from each subset. A selection unit that selects one combination based on a third index value calculated using a candidate route included in each combination between the nodes for each node that is not necessarily adjacent;
A route selection device comprising: In the combination in which the sum of the first index values is equal to or less than a first threshold among the combination group, the selection unit includes the third index for each not necessarily adjacent node and for each earthquake. The combination having the smallest maximum value among the values is selected, or the maximum value among the third index values for each non-adjacent node and for each earthquake is selected from the combination group. A combination having a minimum sum of the first index values is selected from combinations that are equal to or less than a threshold;
The route selection device according to claim 1 . The selection unit includes the second index for each of the nodes that are not necessarily adjacent to each other and for each earthquake in a combination in which the sum of the first index values is equal to or less than a first threshold value. A combination having the smallest weighted sum with the occurrence probability of each earthquake as a weight is selected, or the weighted sum for each not necessarily adjacent node and for each earthquake is a third threshold value from the combination group In the following combinations, the combination that has the smallest sum of the first index values is selected.
The route selection apparatus according to claim 1 or 2 , wherein For each subset, the extraction unit is a Pareto optimal solution for an optimization problem aimed at minimizing the first index value of each candidate route and minimizing the second index value for each earthquake. Extracting the subset by obtaining a set;
The route selection device according to any one of claims 1 to 3, wherein Computer
For each adjacent node of the plurality of nodes constituting the network, for each candidate route included in the set of candidate routes between the adjacent nodes in the network , a route length that is a first index value and a predetermined predetermined value A second index value indicating the magnitude of the impact of the earthquake for each earthquake, and from the set based on the first index value and the second index value for each earthquake An extraction procedure to extract a subset ;
In the network, a sum of the first index values of each candidate route included in each combination is generated from a combination group of candidate routes generated by selecting one candidate route from each subset. A selection procedure for selecting one combination based on a third index value calculated using a candidate route included in each combination between the nodes for each node that is not necessarily adjacent;
The route selection method characterized by performing. In the selection procedure, the third index for each of the nodes that are not necessarily adjacent to each other and for each earthquake among the combinations in which the sum of the first index values is equal to or less than the first threshold value. The combination having the smallest maximum value among the values is selected, or the maximum value among the third index values for each non-adjacent node and for each earthquake is selected from the combination group. A combination having a minimum sum of the first index values is selected from combinations that are equal to or less than a threshold;
6. The route selection method according to claim 5 , wherein: In the selection procedure, the second index for each of the nodes that are not necessarily adjacent to each other and for each earthquake among the combinations in which the sum of the first index values is equal to or less than a first threshold value. A combination having the smallest weighted sum with the occurrence probability of each earthquake as a weight is selected, or the weighted sum for each not necessarily adjacent node and for each earthquake is a third threshold value from the combination group In the following combinations, the combination that has the smallest sum of the first index values is selected.
7. The route selection method according to claim 5 or 6 , wherein: The program for functioning a computer as each part as described in any one of Claims 1 thru | or 4 .

Images