JP2016173609A - Evacuation simulation device, evacuation simulation method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an evacuation simulation device, evacuation simulation method, and program capable of rapidly and appropriately formulating an evacuation plan in accordance with an actual disaster situation.SOLUTION: An evacuation simulation device 100 includes: a first simulation execution unit 110 for deriving, by a first simulation method, a plurality of optimal evacuation route candidates whose evacuation completion time is the shortest; and a second simulation execution unit 120 for selecting, by a second simulation method, an optimal evacuation route whose evacuation completion time is the shortest from the plurality of optimal evacuation route candidates.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an evacuation simulation apparatus, an evacuation simulation method, and a program, for example, a technique for generating an evacuation route according to a disaster situation.

  Various methods for predicting the behavior of evacuees when a disaster such as a fire occurs and formulating an appropriate evacuation plan are known. In the conventional evacuation safety verification method for buildings, the building evacuation safety performance is evaluated by calculating the evacuation completion time in the event of a disaster according to a simplified formula based on the floor plan and passage width at the building design stage. It has been broken. In recent years, verification methods for evacuation safety performance using computer simulations have been accepted instead of simplified formulas, and many previous studies and packaged software exist.

  For example, Patent Literature 1 describes a computer system that predicts the progress of an event and the behavior of a person existing on the route based on the structure of the route and an event (that is, a disaster) that occurs in the route.

  Patent Document 2 also tracks the evacuation situation at any point during evacuation by modeling the characteristics of individual refugees at the time of the disaster and by sequentially reproducing the state of each individual during the evacuation action, A computer system for identifying evacuation bottlenecks is described.

  According to the evacuation simulation technology using these computer systems, for example, when a new building is designed, a floor layout design with high evacuation safety performance can be performed. Specifically, the designer can repeat the evacuation plan using evacuation simulation technology and the floor layout design to improve the evacuation plan, thereby enabling more rapid and appropriate evacuation. Can be created. It is also possible to evaluate the robustness of the evacuation plan against changes in disaster scenarios (such as fire locations and refugee attributes). Furthermore, existing buildings can also be used for education for staff and residents by preparing an evacuation plan in advance using the same method.

Japanese Patent No. 4822812 Japanese Patent No. 5372421

  When formulating an evacuation plan using conventional evacuation simulation technology, simulation is usually carried out based on several disaster scenarios assumed in advance. However, the actual disaster situation does not necessarily match the disaster scenario assumed in advance. For example, in an actual disaster, the route through which the evacuees can pass changes every moment, and the optimum evacuation route and evacuation plan can change every moment according to such changes.

  Therefore, when applying the evacuation simulation technology to an actual disaster, it is desirable to reexamine the evacuation plan according to changes in the disaster situation. Moreover, it is required to make a quick and accurate decision within a limited time. However, the conventional evacuation simulation techniques as described in Patent Documents 1 and 2 firstly take into account the characteristics of each refugee in detail, and therefore the processing load is heavy and the calculation time tends to be large. is there. Secondly, it is necessary for the user to manually set a new disaster scenario according to the actual disaster situation and formulate a corrected evacuation route. That is, the user must correct the evacuation route by trial and error. For these reasons, it has been difficult to apply the conventional evacuation simulation technology to formulate an evacuation plan for an actual disaster.

  The present invention has been made to solve such problems, and provides an evacuation simulation apparatus, an evacuation simulation method, and a program capable of quickly and appropriately formulating an evacuation plan according to an actual disaster situation. With the goal.

  Other problems and novel features will become apparent from the description of the specification and the accompanying drawings.

  The evacuation simulation apparatus according to the present invention includes a first simulation execution unit for deriving a plurality of optimum evacuation route candidates having the shortest evacuation completion time by the first simulation method, and a plurality of the optimum by the second simulation method. A second simulation execution unit that selects an optimum evacuation route that minimizes the evacuation completion time from evacuation route candidates.

  The evacuation simulation method according to the present invention includes a first simulation step for deriving a plurality of optimum evacuation route candidates having the shortest evacuation completion time by a first simulation method, and a plurality of the optimum evacuation by a second simulation method. And a second simulation step of selecting an optimum evacuation route with the shortest evacuation completion time from the route candidates.

  The program according to the present invention is a program for causing a computer to execute the evacuation simulation method.

  According to the present invention, it is possible to provide an evacuation simulation apparatus, an evacuation simulation method, and a program capable of quickly and appropriately formulating an evacuation plan according to an actual disaster situation.

It is a figure which shows the structure of the evacuation simulation apparatus 100 concerning embodiment of this invention. It is a figure which shows operation | movement of the evacuation simulation apparatus 100 concerning embodiment of this invention. It is a figure which shows operation | movement of the 1st simulation execution part 110 concerning embodiment of this invention. It is a figure which shows operation | movement of the 1st simulation execution part 110 and the 2nd simulation execution part 120 concerning embodiment of this invention. It is a figure which shows operation | movement of the 2nd simulation execution part 120 concerning embodiment of this invention. It is a figure which shows operation | movement of the 2nd simulation execution part 120 concerning embodiment of this invention. It is a figure which shows the comparison with the 1st simulation execution part 110 and the 2nd simulation execution part 120. FIG.

<Embodiment 1>
Hereinafter, specific embodiments to which the present invention is applied will be described in detail with reference to the drawings.

  First, the configuration of the evacuation simulation apparatus 100 according to the first embodiment of the present invention will be described using the block diagram of FIG.

  The evacuation simulation apparatus 100 includes an input unit 102, an output unit 104, a first simulation execution unit 110, and a second simulation execution unit 120.

  The evacuation simulation apparatus 100 is typically an information processing apparatus such as a personal computer (PC). A typical evacuation simulation apparatus 100 includes a central processing unit (CPU), a storage device, an input / output device, and the like. The central processing unit reads the application program stored in the storage device and executes predetermined processing to logically connect the input unit 102, the output unit 104, the first simulation execution unit 110, and the second simulation execution unit 120. Realize.

  The input unit 102 acquires information necessary for the evacuation simulation, such as the spatial structure of the building, the number of evacuees, attributes, and initial arrangement. The input unit 102 acquires a basic evacuation route in the building. All or part of the information may be stored in advance in the storage device, or may be input from the outside via an input device such as a keyboard or a communication interface. The input unit 102 outputs the acquired information to the first simulation execution unit 110 and the second simulation execution unit 120.

  The first simulation execution unit 110 executes an evacuation simulation using a model that is faster than a second simulation execution unit 120 described later. In this embodiment, as an evacuation simulation method in the first simulation execution unit 110, an evacuation route of a building is expressed as a route network composed of nodes and a one-dimensional link, and the refugee is converted into a one-dimensional link of the route network. The evacuation simulation by the meso model that moves along is adopted (FIGS. 3 and 4).

  The first simulation execution unit 110 outputs a plurality of evacuation routes that satisfy a predetermined criterion as an evacuation simulation result. In other words, a plurality of optimum evacuation route candidates that reduce the overall evacuation time are output.

  The second simulation execution unit 120 executes an evacuation simulation for each of the plurality of optimum evacuation route candidates output by the first simulation execution unit 110 using a more accurate model than the first simulation execution unit 110. In the present embodiment, as an evacuation simulation method in the second simulation execution unit 120, a building is expressed as a two-dimensional space partitioned by a shield (wall), and the refugee freely moves in the space. The evacuation model simulation is adopted (Fig. 4).

  The second simulation execution unit 120 outputs the route having the shortest overall evacuation time among the plurality of optimum evacuation route candidates, that is, the optimum evacuation route as a processing result.

  The output unit 104 outputs the optimum evacuation route output by the second simulation execution unit 120 so that the user can recognize the screen by, for example, screen display.

  Next, the operation of the evacuation simulation apparatus 100 according to the first embodiment will be described using the flowchart of FIG.

S101: Basic Evacuation Route Input The input unit 102 acquires a basic evacuation route and outputs it to the first simulation execution unit 110. The basic evacuation route is an evacuation route used as an initial value when the first simulation execution unit 110 performs an evacuation simulation. Typically, the evacuation route assumed at the time of building design is used as the basic evacuation route.

  The basic evacuation route is data having a structure conforming to the simulation method performed by the first simulation execution unit 110. For example, in the present embodiment, the first simulation execution unit 110 expresses a route in a building by a route network composed of nodes and one-dimensional links, and evacuees follow the one-dimensional links of the route network. An evacuation simulation using a meso model is performed (Fig. 3). The basic evacuation route in this case is data composed of a combination of a plurality of nodes and a one-dimensional directed link connecting these nodes, as shown as evacuation route X in FIG.

  In addition, the input unit 102 acquires information necessary for the evacuation simulation, such as the spatial structure of the building, the number of evacuees, attributes, and initial arrangement. In the present embodiment, the building evacuation route is defined as a network structure including a plurality of nodes and a one-dimensional link connecting them as described above. The plurality of nodes include nodes serving as evacuation destinations (FIG. 3, destinations A and B). In addition, a predetermined number of evacuees are assigned to predetermined links as initial values (initial arrangement).

  The basic evacuation route and other information described above may be stored in advance in a storage device (not shown), may be input by a user via an input device such as a keyboard, or from an external device or the like via a communication interface or the like. It may be entered.

S102: Extraction of Optimal Evacuation Route Candidates The first simulation execution unit 110 extracts an optimal evacuation route candidate group using an evacuation simulation and a local search using the meso model shown in FIG. The basic framework of local search is shown below.

Step 1: Input an initial solution X (basic evacuation route).
The basic evacuation route input in S101 is set as an initial solution X.

Step 2: A current solution X is randomly modified to generate a solution Y.
The first simulation execution unit 110 randomly deletes one link from the evacuation route indicated by the solution X, randomly adds another link, and generates the solution Y. For example, as shown in FIG. 3, the link (17, 18) is deleted from the solution X, and the link (16, 10) is newly added. The new evacuation route thus created is defined as solution Y.

  Here, if the solution Y is not a tree structure, the first simulation execution unit 110 preferably corrects the evacuation route to a tree structure by inverting the direction of a specific link included in the solution Y. For example, in the example of FIG. 3, the solution Y can be corrected to a tree structure having the destination A as the root node by inverting the links (16, 17).

  The reason why the evacuation route is preferably a tree structure will be described. When an actual disaster occurs, it is generally desirable that the direction of evacuation guidance be uniform. For example, evacuation guidance that directs 50% of evacuees to the left and the remaining 50% to the right evacuation route is difficult in practice. In other words, it is desirable for the evacuees to follow the evacuation route so that they can reach the evacuation site as the destination without facing the fork. In the middle of the evacuation route, there may be a merger with another evacuation route for evacuation guidance.

  In the evacuation simulation method using a known meso model, an evacuation route expression method considering such restrictions and an algorithm for calculating an evacuation route based on such an expression have not been proposed. The conventional evacuation simulation method does not pay attention to the difficulty of evacuation guidance as described above, and adopts a model on the assumption that each refugee evacuates based on autonomous judgment. In other words, an evacuation route is represented as a network composed of nodes and links, and a given branching rate is set at the branch point on the route. Based on a flow model described in advance, the evacuation route is the shortest route. The time was being calculated.

Step 3: If the evacuation completion time of the solution Y is shorter than the solution X, the solution X is replaced with the solution Y.
The first simulation execution unit 110 calculates an evacuation completion time for each of the solution X and the solution Y. Specifically, based on the preset flow coefficient of each link (the number of evacuees that can pass through the link per unit time), the density-speed relational expression, and the number of evacuees at each node, Calculate the time to reach the destination. In addition, since the concrete numerical formula etc. concerning the said calculation are well-known, detailed description is omitted here.

  Note that it is desirable to adjust the flow coefficient and the density-speed relational expression described above so that the evacuation simulation result based on the meso model and the evacuation simulation result based on the micro model described later match as much as possible.

  Thereafter, if the evacuation completion time of the solution Y is shorter than the solution X, the first simulation executing unit 110 replaces the solution X with the solution Y.

Step 4: Repeat steps 2 and 3 until the abort condition is met.
The first simulation execution unit 110 repeatedly executes steps 2 and 3 until a predetermined termination condition is satisfied. By repeating this trial, a more appropriate evacuation route can be found.

  This iterative calculation can be terminated, for example, when the number of repetitions reaches a preset value. Alternatively, it may be terminated when it is determined that the evacuation completion time has converged using a known convergence determination method.

  Thereafter, the first simulation execution unit 110 stores the solution X as an evacuation route (optimum evacuation route candidate) estimated to have the shortest evacuation completion time.

  The first simulation execution unit 110 repeatedly executes the processes relating to the above-described steps 1 to 4 a predetermined number of times. Thereby, a plurality of optimum evacuation route candidates can be obtained.

S103: Identifying the optimal evacuation route The second simulation execution unit 120 expresses the building as a two-dimensional space partitioned by a shield (wall), and the refugee moves freely in the space. Perform a simulation. Here, the second simulation execution unit 120 evaluates the plurality of optimum evacuation route candidates obtained in S102 by the evacuation simulation of this micro model, and identifies the optimum evacuation route with the shortest evacuation time.

  By the way, in the evacuation simulation method using a known micro model, it is assumed that the evacuation is the shortest distance from the current location to the destination. It was. That is, as shown in FIG. 5, in the known micro model, the shortest distance (potential value) to the destination is given to the space where the refugee moves, and the refugee moves from the current location in a direction where the potential value decreases. . Therefore, it is difficult to intentionally form an evacuation route in which an evacuee moves along a link, which is considered in the above meso model.

  On the other hand, in this Embodiment, as shown in FIG. 6, the space inside a building is divided | segmented into the area | region corresponding to the link in the above-mentioned meso model. Then, the potential value in each corresponding region is calculated so as to spread from the start node to the end node of the link in the direction toward the destination of the link corresponding to the region. As a result, even when the micro model is used, the refugee can be moved in the direction along the shortest evacuation route candidate obtained by the first simulation execution unit 110. That is, the shortest evacuation route candidate obtained by the meso model can be evacuated by the micro model.

  The second simulation execution unit 120 calculates, for each of the plurality of optimum evacuation route candidates, the evacuation completion time when evacuating along the evacuation route using a micro model. As the micro model, for example, a multi-agent method, a cellular automaton method, or the like can be employed. In this embodiment, when evacuees evacuate in a direction with a small potential value while colliding and avoiding each other, the time until all the evacuees reach the destination is calculated. In addition, since the specific method of the said calculation is well-known, detailed description is omitted here.

  Here, FIG. 7 shows the difference between the meso model executed by the first simulation execution unit 110 and the micro model executed by the second simulation execution unit 120. Although the calculation speed of the micro model is slower than that of the meso model, the evacuation completion time can be simulated more precisely because it can take into account congestion at the confluence, walking speed change according to density, and stopping due to interference with others. It is possible.

  Finally, the second simulation execution unit 120 outputs the evacuation route (optimum evacuation route) having the shortest evacuation completion time to the output unit 104 as a result of the above calculation. Typically, a tree structure composed of nodes and links corresponding to the micro model with the shortest evacuation completion time is output as the shortest evacuation route.

S104: Output of optimum evacuation route The output unit 104 presents the optimum evacuation route to the user. For example, a link indicating the shortest evacuation route is superimposed on the plan view of the building and displayed on the display.

  According to the present embodiment, first, first simulation execution unit 110 extracts a plurality of optimum evacuation route candidates by a simple and high-speed simulation method. Thereby, it is possible to shorten the calculation time until the optimum evacuation route is derived as much as possible.

  Further, according to the present embodiment, the second simulation execution unit 120 performs a precise simulation on the plurality of optimum evacuation route candidates extracted by the first simulation execution unit 110 to determine the optimum evacuation route. Thereby, a more appropriate evacuation route can be derived even within a limited time.

  Further, in the present embodiment, the first simulation execution unit 110 generates a large number of routes in which basic evacuation routes are partially changed, performs evacuation simulations of various evacuation routes, and shortens the overall evacuation completion time. The optimal evacuation route candidate to be extracted is extracted. As a result, it is possible to automate the evacuation route examination work conventionally performed by trial and error of manual work, and to extract optimum evacuation route candidates at high speed even when an actual disaster occurs.

<Embodiment 2>
The evacuation simulation apparatus 100 according to the second embodiment has a feature in the evacuation route recombination strategy in steps 2 and 3 of S102. Other configurations and operations are the same as those in the first embodiment unless otherwise specified.

  In S102 of the first embodiment (optimum evacuation route candidate extraction processing) and step 2 (processing of randomly correcting the current solution X and generating solution Y), the links to be deleted and added are randomly selected. Because it was selected, it was sometimes inefficient and too time consuming, making it unsuitable for use in the event of an actual disaster. In order to solve this problem, Embodiment 2 presents two types of link recombination strategies.

Strategy A: Pre-exclusion of link patterns that may be altered The first simulation execution unit 110 deletes one link at random from the evacuation route indicated by the solution X in Step 2 and randomly one other link pattern. To generate a solution Y. Subsequently, in step 3, the evacuation completion time is calculated for each of the solution X and the solution Y.

  Here, if the evacuation completion time of the solution Y is later than the solution X, that is, if the evacuation route is deteriorated due to addition and deletion of the link, the first simulation execution unit 110 is added and deleted at this time Keep the link as a bad link pattern. In the subsequent trials, in step 2, the links included in the corrupted link pattern are not added or deleted.

  In addition, when the evacuation completion time of the solution Y becomes earlier than the solution X after that, the first simulation execution unit 110 preferably clears the corrupted link pattern. This is because the corrupted link pattern may change due to the replacement of the solution.

Strategy B: Link selection strategy switching in the middle of repetitive processing During the repetitive processing of steps 1 to 4, in the first half of repetitive processing, the first simulation execution unit 110 recombines links across evacuation routes of different destinations. Is acceptable. That is, as shown in FIG. 3, for example, a link belonging to a tree structure having destination B as a root node may be recombined so as to change the affiliation to a tree structure having destination A as a root node. To do. Thereby, it is possible to broadly and roughly search for an evacuation route that may shorten the evacuation time.

  On the other hand, in the second half of the iteration, the first simulation execution unit 110 performs recombination of links limited to the evacuation route of the same destination. That is, for example, in FIG. 3, if the delete link (8, 9) is selected from the tree structure having the destination A as the root node, the additional link (3, 9) is added to the tree structure having the same destination A as the root node. Only recombination that sets). Thereby, it is possible to perform a search for brushing up the evacuation route corresponding to each destination.

  According to the present embodiment, the first simulation execution unit 110 eliminates in advance the link recombination pattern that may cause the evacuation completion time to be worsened, so that the extraction process of the optimum evacuation route candidate can be made more efficient. it can.

  In addition, according to the present embodiment, the first simulation execution unit 110 switches the link selection strategy in the middle of the iterative process, so that it is possible to search widely for an evacuation route that may shorten the evacuation time, It is also possible to brush up the evacuation route corresponding to the destination.

  Note that the present invention is not limited to the above-described embodiment, and can be changed as appropriate without departing from the spirit of the present invention. For example, the number, attributes, and initial arrangement of evacuees can be input from a well-known personnel grasping system. A typical in-person personnel grasping system can estimate the number of people in each floor from the entrance / exit log using a security card or the like. In addition, the number of people in each floor and attribute data (walking speed, etc.) can be estimated by moving image recognition using a surveillance camera or the like.

  Further, the optimum evacuation route output by the above-described embodiment can be output to an existing evacuation guidance system. In this case, the information regarding the optimum evacuation route can be linked to an evacuation guidance indicator (digital signage or the like), in-house broadcasting, an emergency elevator operation system, or the like.

  In the above-described embodiment, the building targeted for the simulation of the evacuation route may be, for example, a high-rise building such as an office building or a condominium, a complex commercial facility, a stadium, a large hall, a large passenger ship, or the like.

  In the above-described embodiment, the present invention has been described as mainly configured by hardware. However, the present invention is not limited to this, and any processing can be performed by a computer program on a CPU (Central Processing Unit). It can also be realized by executing. In this case, the computer program can be stored and provided to the computer using various types of non-transitory computer readable media. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (for example, flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (for example, magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / W, semiconductor memory (for example, mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (Random Access Memory)). The program may also be supplied to the computer by various types of transitory computer readable media. Examples of transitory computer readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire and an optical fiber, or a wireless communication path.

DESCRIPTION OF SYMBOLS 100 Evacuation simulation apparatus 102 Input part 104 Output part 110 1st simulation execution part 120 2nd simulation execution part

Claims (8)

A first simulation execution unit for deriving a plurality of optimum evacuation route candidates with the shortest evacuation completion time by the first simulation method;
An evacuation simulation apparatus comprising: a second simulation execution unit that selects an optimum evacuation route that has the shortest evacuation completion time from the plurality of optimum evacuation route candidates by a second simulation method. The evacuation simulation apparatus according to claim 1, wherein a calculation time of the evacuation completion time by the first simulation execution unit is shorter than a calculation time of the evacuation completion time by the second simulation execution unit. The first simulation execution unit includes:
A first evacuation route including a plurality of nodes and links;
The evacuation simulation apparatus according to claim 1 or 2, wherein the optimum evacuation route candidate is searched by comparing the evacuation completion time with a second evacuation route obtained by randomly recombining the links of the first evacuation route. . The first simulation execution unit includes:
If the evacuation completion time of the second evacuation route is longer than the evacuation completion time of the first evacuation route, the recombined link is held as a specific link,
The evacuation simulation apparatus according to claim 3, wherein the specific link is excluded from the recombination targets and the search is repeatedly executed. The nodes and the links form a tree structure;
The first simulation execution unit includes:
A first phase allowing the recombination across a plurality of the tree structures;
The evacuation simulation apparatus according to claim 3, wherein the search is executed by a second phase that does not allow the recombination across a plurality of the tree structures. The second simulation execution unit
The evacuation simulation apparatus according to any one of claims 1 to 5, wherein an autonomous refugee selects the optimum evacuation route based on a simulation model in which an area corresponding to the optimum evacuation route candidate is moved. A first simulation step of deriving a plurality of optimum evacuation route candidates with the shortest evacuation completion time by the first simulation method;
An evacuation simulation method comprising: a second simulation step of selecting an optimum evacuation route having the shortest evacuation completion time from the plurality of optimum evacuation route candidates by a second simulation method. A program for causing a computer to execute the method according to claim 7.