Title of invention: In-building pedestrian flow estimation system and estimation method
Technical field
[0001]
　The present invention relates to an in-building human flow estimation system and an estimation method, and is suitable for application to, for example, an in-building human flow simulation for estimating a human flow in a building.
Background technology
[0002]
　In order to operate the elevator properly and plan a renewal to improve the usability of the elevator, grasp the operating status and usage status of the elevator and predict the status when various improvement plans are applied. It is important to.
[0003]
　Here, a method of estimating the number of people getting on and off an elevator on each floor from a change in load detected by the elevator is disclosed (see Patent Document 1).
[0004]
　Further, there is disclosed a device that estimates in-building traffic data indicating from which floor a pedestrian in a building has moved to, from the number of people getting on and off an elevator on each floor (see Patent Document 2).
[0005]
　Also, by inputting the traffic data in the building, the layout data of each floor in the building, and the information about the installation conditions such as the installation position of the elevator, the stop floor, the number of passengers, and the speed, the transportation of people by the elevator is simulated. A device is disclosed (see Patent Document 3).
[0006]
　For example, by recording the change in load on each floor detected by the elevator as the elevator operation record data, it is possible to estimate the in-building traffic data using the techniques disclosed in Patent Document 1 and Patent Document 2. Further, by using the device disclosed in Patent Document 3, the traffic data in the building, the layout data of the building, and the information on the installation conditions of the elevator installed in the building are simulated as input values, thereby You can estimate the movement of pedestrians and the operation of elevators. As a result, it becomes possible to grasp the operation status of the elevator and to make a prediction when the installation conditions are changed.
Prior art documents
Patent literature
[0007]
Patent Document 1: Japanese Patent Publication No. 60-310 Publication
Patent Document 2: Japanese Patent Publication No. 62-36954 Publication
Patent Document 3: Japanese Patent Publication No. 2009-096612
Summary of the invention
Problems to be Solved by the Invention
[0008]
　Here, Patent Document 3 describes a function of outputting the elevator operation record, the number of people waiting in the hall, and the estimation result indicating the flow of people based on the simulation result. There is no description about the method to judge whether it can be reproduced. For example, the evaluation of the operation results of the elevator can be realized by comparing the log data of the operation results recorded in the elevator control device with the estimation result, but unless the expert is familiar with the operation of the elevator. Analysis of log data is difficult. Therefore, it is difficult to judge whether or not the estimation result can correctly reproduce the reality, only by using the device described in Patent Document 3.
[0009]
　The present invention has been made in consideration of the above points, and is intended to propose an in-building human flow estimation system and an estimation method that can evaluate the degree of reproducibility of an estimated human flow.
Means for solving the problem
[0010]
　In order to solve such a problem, in the present invention, in-building traffic data indicating which floor a person moves from to which floor using an elevator in a building, building layout data related to the layout of the building, and the elevator An in-building pedestrian flow estimation system having an in-building pedestrian flow simulation unit that estimates a pedestrian flow using the elevator parameter value as an input value, the first evaluation relating to the occupant flow estimated by the in-building pedestrian flow simulation unit. Based on the index and the second evaluation index related to the actual flow of people in the building, an evaluation value for evaluating the degree of reproduction of the flow of people estimated by the in-building flow simulation unit is calculated. An evaluation unit and an output unit that outputs the evaluation value calculated by the evaluation unit are provided.
[0011]
　Further, in the present invention, intra-building traffic data indicating which floor a person moves from to which floor using an elevator in the building, building layout data related to the layout of the building, and elevator parameter values ​​related to the elevator. An estimation method in an in-building pedestrian flow estimation system having an in-building pedestrian flow simulation unit for estimating a pedestrian flow as an input value, wherein: And the second evaluation index related to the actual flow of people in the building, an evaluation value for evaluating the degree of reproducibility of the flow of people estimated by the pedestrian flow simulation unit in the building. A first step for calculating and a second step for the output unit to output the evaluation value calculated in the first step are provided.
[0012]
　According to the above configuration, the evaluation value for evaluating the degree of reproducibility of the estimated human flow is output.
Effect of the invention
[0013]
　According to the present invention, the degree of reproduction of the estimation result can be evaluated.
Brief description of the drawings
[0014]
FIG. 1 is a diagram showing an example of the configuration of an in-building pedestrian flow estimation system according to a first embodiment.
FIG. 2 is a diagram for explaining an arrival time at an elevator hall according to the first embodiment.
FIG. 3 is a diagram showing an example of an elevator operation log according to the first embodiment.
FIG. 4 is a diagram showing an example of building layout data according to the first embodiment.
FIG. 5 is a diagram showing an example of arrival distribution data according to the first embodiment.
FIG. 6 is a diagram showing an example of intra-building traffic data according to the first embodiment.
FIG. 7 is a diagram showing an example of elevator parameter values ​​according to the first embodiment.
FIG. 8 is a diagram showing an example of an evaluation index according to the first embodiment.
FIG. 9 is a diagram showing an example of a processing procedure relating to processing executed by a parameter value calculation unit according to the first embodiment.
FIG. 10 is a diagram showing an example of a relationship between a traveling distance and a traveling time according to the first embodiment.
FIG. 11 is a diagram showing an example of a relationship between traveling time and speed according to the first embodiment.
FIG. 12 is a diagram showing an example of a processing procedure related to processing executed by the evaluation unit according to the first embodiment.
FIG. 13 is a diagram showing an example of a correlation table according to the first embodiment.
MODE FOR CARRYING OUT THE INVENTION
[0015]
　Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.
[0016]
(1) First Embodiment In
　FIG. 1, reference numeral 1 denotes the whole in-building pedestrian flow estimation system according to the first embodiment. The in-building pedestrian flow estimation system 1 is a system that estimates the movement of people and the operation of an elevator in a building by simulation, and includes one or more computers that can communicate with each other and software that runs on the computers. Composed.
[0017]
　The more accurately the in-building pedestrian flow estimation system 1 reproduces the current situation, the more accurately the effects of introducing the latest elevators and the effects of changing the operation of existing elevators can be obtained. Become. However, neither a method of judging whether or not the current situation is precisely reproduced, nor a method of precisely reproducing the current situation is disclosed. Therefore, these methods will be mainly described in the present embodiment.
[0018]
　The in-building pedestrian flow estimation system 1 has various functions (intra-building traffic estimation unit 200, in-building pedestrian flow simulation unit 300, parameter value calculation unit 400, elevator index calculation unit 500, pedestrian flow measurement unit 600, evaluation unit 700, output unit 800. Etc.).
[0019]
　Various functions of the in-building pedestrian flow estimation system 1 include one or a plurality of computers (not shown), more specifically, a CPU (Central Processing Unit) that stores a program stored in a ROM (Read Only Memory) in a RAM ( It is realized by reading out to Random Access Memory) and executing it (software). Part or all of the various functions of the in-building pedestrian flow estimation system 1 may be realized by hardware such as a circuit and a sensor, or may be realized by software and hardware.
[0020]
　Further, the in-building pedestrian flow estimation system 1 includes the elevator operation log 110, the building layout data 120, the arrival distribution data 130, the in-building traffic data 140, the elevator parameter value 150, the first evaluation index 160A, and the second evaluation index 160B. A storage device (not shown) capable of recording and reading data as data is provided. Note that when it is not necessary to distinguish between the first evaluation index 160A and the second evaluation index 160B, it will be referred to as the evaluation index 160 as appropriate.
[0021]
　Although the details will be described later, the elevator operation log 110 is data indicating the operation record of the elevator, and FIG. 3 shows an example of the elevator operation log 110. The building layout data 120 is data relating to the layout of the building, and FIG. 4 shows an example of the building layout data 120. The arrival distribution data 130 is data indicating how quickly people arrive in the elevator hall with respect to the time of getting on the elevator, and FIG. 5 shows an example of the arrival distribution data 130. The in-building traffic data 140 is data indicating from which floor a person moves to which floor using an elevator in the building, and FIG. 6 shows an example of the in-building traffic data 140. The elevator parameter value 150 is a parameter value related to the elevator, and FIG. 7 shows an example of the elevator parameter value 150. The evaluation index 160 is an index related to a person and an elevator (an index related to the flow of people), and an example of the evaluation index 160 is shown in FIG.
[0022]
The in-
　building traffic estimation unit 200 uses the elevator operation log 110 including the transition of the load of the elevator and the arrival distribution data 130 as input data, and estimates (calculates) in-building traffic data 140 and outputs it.
[0023]
　The amount of movement of a person between floors can be estimated by combining the techniques disclosed in Patent Document 1 and Patent Document 2, for example. In Patent Document 2, the time when the movement amount of each floor occurs is the time when the elevator car arrives at each floor. When estimating the movement of a person existing in an elevator, it is sufficient to estimate the movement of the person from the time when the car arrives. With the in-building pedestrian flow estimation system 1, not only the inside of the elevator car but also the elevator hall, etc. , Estimate the movement of people outside the elevator.
[0024]
　For example, as shown in FIG. 2, considering the movement of pedestrians 221 in the elevator hall, each pedestrian 221 who gets on the car 231 generally has the elevator hall between the call button pressing time 211 and the car arrival time 212. Arrive at the car arrival time 212 and board the car 231 together. Therefore, the pedestrian 221 arrives at the elevator hall a certain time before the car arrival time 212. Therefore, in the in-building pedestrian flow estimation system 1, the pedestrian 221 arrives at the elevator hall before the car arrival time 212 of the elevator by the time according to the arrival distribution data 130, using the arrival distribution data 130 in which the above-described fixed time is defined by distribution. Assuming that the vehicle has arrived, the in-building traffic data 140 is estimated using the time of arrival at the elevator hall as the departure time of the pedestrian 221.
[0025]
　When estimating the movement of people not only in the elevator hall but also in a wide area such as a passage in the building, not only the time according to the arrival distribution data 130 but also the elevator hall from the point where the person in the building starts moving. In consideration of the travel time up to, the in-building traffic data 140 is estimated with the time preceding the total value of the time according to the arrival distribution data 130 and the travel time as the departure time of the person. The travel time may also be set as a distribution like the arrival distribution data 130, or the travel distribution time may be added in advance to create the arrival distribution data 130. In addition, when there are a plurality of points where a person starts moving, the arrival time distribution data 130 may be created by adding the moving time or the moving time for each combination of the elevator hall and the point where the person starts moving.
[0026]
　With this configuration, it is possible to estimate the movement of a person outside the elevator, such as the elevator hall, as well as the inside of the elevator car.
[0027]
　The in-building flow simulation unit 300 estimates (simulates) the flow of people using the building layout data 120, the in-building traffic data 140, and the elevator parameter value 150 as input values, or based on the simulation result (simulation result). The evaluation index 160A of 1 is output. More specifically, the in-building pedestrian simulation unit 300 includes the building layout data 120, which is structured data including installation positions of equipment such as elevators, and the in-building traffic data estimated by the in-building traffic estimation unit 200. 140 and an elevator parameter value 150 including values ​​such as elevator stop floor, number of passengers, rated speed, rated acceleration, etc. are input, a function of simulating elevator operation and movement of people, and a simulation result are aggregated first It has a function of outputting the evaluation index 160A.
[0028]
　The in-building pedestrian flow simulation unit 300 also includes, as internal parameter values, the walking speed of a person, the boarding rate for determining whether or not to board a car based on the number of passengers in the car of the elevator and the number of people already boarded.
[0029]
　For example, the boarding rate is set as a threshold value indicating the upper limit of the ratio calculated from the number of passengers in the elevator car and the number of passengers in the elevator car (the ratio of the passengers who are judged to be full to the passengers to the passengers). .. When the boarding rate exceeds the boarding rate, by performing processing that does not allow additional boarding in the simulation, it is possible to reproduce the actual behavior in which the additional boarding does not occur when the elevator is crowded even if the elevator is not full. As the boarding ratio, for example, a ratio of the maximum value of the in-car load 1116 recorded in the elevator operation log 110 to the capacity is used.
[0030]
　Although the boarding rate and the like have been described as internal parameter values, the present invention is not limited to this and may be calculated by the parameter value calculation unit 400 (may be included in the elevator parameter value 150).
[0031]
　In addition, the simulation in the in-building pedestrian flow simulation unit 300 can be realized, for example, by using the technique disclosed in Patent Document 3.
[0032]
　The parameter value calculation unit 400 receives the elevator operation log 110 and calculates the elevator parameter value 150 (rated speed, rated acceleration, etc.). The elevator parameter value 150 can be determined from the product specifications of the elevator, but normally, some error occurs during operation due to the installed conditions, deterioration over time, and the like. Therefore, the parameter value calculation unit 400 more accurately calculates the elevator parameter value 150 using the elevator operation log 110. According to such a configuration, the in-building pedestrian flow estimation system 1 can more accurately reproduce the current situation. However, the parameter value calculation unit 400 may be omitted and the product specifications may be used as they are as the elevator parameter values ​​150.
[0033]
　The elevator index calculation unit 500 aggregates the elevator operation logs 110, calculates and outputs a second evaluation index 160B for evaluating the operation status of the elevator. The elevator index calculation unit 500 may be omitted and only the index output by the pedestrian flow measurement unit 600 may be used as the second evaluation index 160B.
[0034]
　The pedestrian flow measurement unit 600 measures (measures) a congestion index indicating the number of people staying in the elevator hall, the number of people in line, the degree of congestion, and the like. The pedestrian flow measurement unit 600 includes a sensor unit that performs measurement, and an analysis unit that calculates and outputs a congestion index based on information output by the sensor unit.
[0035]
　For example, the pedestrian flow measurement unit 600 uses a camera as a sensor unit to photograph the inside of an elevator hall, and an analysis device (computer or the like) having image analysis software capable of analyzing the number of people within a predetermined range in the camera image is specified as the analysis unit. The number of people staying within the range is output. Sensors such as cameras may be installed in the elevator halls on all floors, or may be installed only on some floors. It should be noted that the human flow measuring unit 600 may be omitted and only the index output by the elevator index calculating unit 500 may be used as the second evaluation index 160B.
[0036]
　The human flow measuring unit 600 is not limited to the above configuration. For example, the pedestrian flow measurement unit 600 may include a sensor unit, and the in-building pedestrian flow estimation system 1 may include an analysis unit. Further, for example, the in-building pedestrian flow estimation system 1 may include a sensor unit and an analysis unit, and the pedestrian flow measurement unit 600 may acquire the congestion index output from the analysis unit.
[0037]
　The evaluation unit 700 uses the first evaluation index 160A related to the flow of people estimated by the in-building flow simulation unit 300 and the second evaluation index 160B related to the actual flow of people in the building, based on the building. An evaluation value (for example, an error evaluation value described later) for evaluating the degree of reproducibility of the human flow estimated by the inner flow simulation unit 300 is calculated.
[0038]
　More specifically, the evaluation unit 700 includes a first evaluation index 160A calculated from the movement of people and the operation of an elevator estimated by the in-building pedestrian flow simulation unit 300, an elevator index calculation unit 500, and a pedestrian flow measurement unit 600. The second evaluation index 160B output from at least one of the two is input, and the error evaluation value calculated from the difference between the first evaluation index 160A and the second evaluation index 160B is calculated. The evaluation unit 700 also has a function of calculating arrival distribution data 130, an elevator parameter value 150, and an internal parameter value of the in-building pedestrian flow simulation unit 300 such that the error evaluation value is equal to or less than a predetermined threshold value. Further, the evaluation unit 700 has a function of outputting the error evaluation value, the arrival distribution data 130, the elevator parameter value 150, and the internal parameter value of the in-building pedestrian flow simulation unit 300, which satisfy a predetermined threshold value, to the output unit 800.
[0039]
　The output unit 800 outputs the evaluation value calculated by the evaluation unit 700. More specifically, the output unit 800 includes, in addition to the simulation result by the in-building human flow simulation unit 300, an evaluation index 160 calculated based on the simulation result, an error evaluation value calculated by the evaluation unit 700, and an error evaluation value. Has a function of outputting arrival distribution data 130, elevator parameter values ​​150, internal parameter values ​​of the in-building pedestrian flow simulation unit 300 and the like used when calculating
[0040]

　Next, data used in the in-building human flow estimation system 1 will be described.
[0041]
　FIG. 3 is a diagram showing an example of the elevator operation log 110. The elevator operation log 110 is data recording the state of each elevator and the state of the call button in the elevator hall. For example, it is configured to include an elevator status log 111 that records the status of the elevator and a call button log 112 that records the status of the call button. The elevator operation log 110 is usually recorded by an elevator control device and used as input information for the in-building pedestrian flow estimation system 1.
[0042]
　Next, the elevator status log 111 will be described. The elevator status log 111 indicates, for example, that the load in the car is the maximum when the door is opened, when the direction of travel of the elevator is changed, when the door is opened or closed, when the state of the call button is changed. This data is recorded when it becomes the minimum.
[0043]
　Next, each item of the elevator status log 111 will be described. Information (numbers, character strings, etc.) that can identify each record is stored (recorded) in the ID 1111. The elevator ID 1112 stores information (numbers, character strings, etc.) that can identify the elevator. At time 1113, information indicating the time when the record is recorded is stored. Information (“up”, “down”, etc.) that can identify the traveling direction of the elevator is stored in the direction 1114. In the position 1115, information that can specify the vertical position of the elevator (for example, information indicating the height of the first floor of the building as a reference) is stored. The in-car load 1116 stores information indicating the total weight of passengers and luggage in the car. The door state 1117 stores information indicating the open/closed state of the door (“open”, “closed”, etc.).
[0044]
　Next, the call button log 112 will be described. The call button log 112 is data recorded when the call button is pressed or when the elevator car arrives. The call button log 112 is recorded separately for each elevator group controlled by the same call button group (one or a plurality of elevators controlled in association with one or a plurality of call buttons).
[0045]
　The call button is not limited to a physical button. For example, in an elevator control system in which the system automatically dispatches a vehicle, the dispatching command of the system may be treated as pressing a call button.
[0046]
　Next, each item of the call button log 112 will be described. The ID 1121 stores information (number, character string, etc.) that can identify each record. At time 1122, information indicating the time when the record is recorded is stored. The floor 1123 stores information (numbers, character strings, etc.) capable of specifying the floor where the call button is pressed or the floor where the car arrives. The direction 1124 stores information (“up”, “down”, etc.) that can specify the direction of the pressed call button or the traveling direction of the arrived car. The state 1125 stores information ("call", "arrival", etc.) for separately recording when the call button is pressed to issue a dispatch request and when the dispatched car arrives. ..
[0047]
　FIG. 4 is a diagram showing an example of the building layout data 120. The building layout data 120 is structured data having attribute information such as passability of the floor layout in the building and passing conditions, and is shown as a plan view. For example, the building layout data 120 is configured to include an entrance/exit 121, a passable area 122 indicating an elevator hall and passages, elevator installation positions 123 to 126, and the like. Although FIG. 4 shows only one floor, the building layout data 120 is provided such that the floor layout of the floor to be evaluated in the building is associated with the floor.
[0048]
　FIG. 5 is a diagram showing an example of the arrival distribution data 130. The arrival distribution data 130 is data holding a distribution indicating how quickly passengers arrive at the elevator hall with respect to the boarding time. In FIG. 5, the distribution is shown by using the relative frequency 1302 of the stay time in the hall for each floor 1301. For example, on the floor B1, 1% of passengers have arrived (entered) in the elevator hall by 10 seconds before the elevator boarding time, and 10% of passengers arrive by 20 seconds before the elevator boarding time. Indicates that you have arrived at the elevator hall.
[0049]
　The distribution may be defined by a distribution such as a normal distribution instead of the relative frequency, and the type of distribution to be used and the parameter values ​​such as average and variance may be recorded. The arrival distribution data 130 may be set for each time zone. For example, the first arrival distribution data may be used from 7:00 to 7:30, and the second arrival distribution data may be used from 7:30 to 8:00.
[0050]
　Here, since it is generally considered that the person who gets on the elevator arrives between the time when the call button is pressed and the time when the elevator arrives, the range of the difference between the time when the call button is pressed and the time when the elevator arrives. Set the distribution so that the frequency inside is large. If the elevator moving in the same direction immediately before is fully occupied and the call button is pressed within the predetermined time range, it is determined that there is a person who could not get on the elevator just before and the call button The distribution may be set so that the frequency of arrival times in a range exceeding the difference between the pressing time and the arrival time of the elevator is large.
[0051]
　FIG. 6 is a diagram showing an example of intra-building traffic data 140. The intra-building traffic data 140 is data indicating the movement demand between floors in the building. The in-building traffic data 140 is, for example, data in which the amount of movement of people between floors in the building is aggregated for each time zone, and the number of people moving between the departure floor 1402 and the destination floor 1403 is calculated for each time zone 1401. Show.
[0052]
　The in-building pedestrian simulation unit 300 virtually generates a person by a passenger agent (an example of a program) and simulates a passenger's movement (people flow simulation) in order to simulate the movement of the person according to the in-building traffic data 140. .. For example, in the example of FIG. 6, the in-building pedestrian flow simulation unit 300 includes 15 passengers heading from B1 to 1F, 3 passengers heading from B1 to 3F, and the like from “7:00:00” to “7:05:00”. To simulate the flow of people. The time of occurrence of the person by the passenger agent may be concentrated at the start time of the time zone 1401, or may be determined according to a constant distribution within the time zone such as a uniform distribution or Poisson distribution.
[0053]
　FIG. 7 is a diagram showing an example of the elevator parameter value 150. The elevator parameter value 150 is configured to include elevator specification data 151 having information such as elevator size and capacity, and service floor data 152 having information regarding the stop floor of the elevator.
[0054]
　Next, each item of the elevator specification data 151 will be described. The elevator ID 1511 stores information (numerical value, character string, etc.) that can identify the elevator. The type 1512 stores information that can identify the type of elevator. For example, in addition to a normal elevator with one car traveling in one hoistway, a multi-car type elevator with multiple cars traveling in one hoistway, and a double deck type car with two stories You can set the elevator etc.
[0055]
　The capacity 1513 stores information indicating the capacity of the elevator. The number of passengers 1513 is set so that the number of passengers cannot exceed the capacity. Information indicating the door width of the elevator is stored in the door width 1514. The door width 1514 is a parameter that affects the number of people who can get on and off at the same time. The rated speed 1515 stores information indicating the rated speed of the elevator. Information indicating the rated acceleration of the elevator is stored in the rated acceleration 1516.
[0056]
　The door opening/closing time 1517 stores information indicating the time required to open/close the door of the elevator. For example, the door opening/closing time 1517 stores the time from when the elevator car arrives at the floor until the door is completely opened, or the time from when the door starts to close until the elevator car starts moving. When the time required to open the door differs from the time required to close the door, the door opening/closing time 1517 may be stored separately for the time of opening the door and the time of closing the door.
[0057]
　Next, each item of the service floor data 152 will be described. The floor 1521 stores information (numerical value, character string, etc.) capable of specifying the floor. The floor height 1522 stores information indicating the height from the floor to the floor directly above. The service floor 1523 stores information indicating the floor on which each elevator can be stopped. For example, in FIG. 7, the elevator identified by "EV1" stops on all floors of "B1", "1F", "2F", and "3F", and the elevator identified by "EV3" is , "B1", "1F", "2F", and "3F", only stop at "1F".
[0058]
　FIG. 8 is a diagram showing an example of the evaluation index 160. The evaluation index 160 is configured to include at least one of the elevator evaluation index 161 and the pedestrian flow evaluation index 162.
[0059]
　Subsequently, the elevator evaluation index 161 will be described. The elevator evaluation index 161 is an evaluation index related to the operation of the elevator calculated by the in-building pedestrian flow simulation unit 300 and the elevator index calculation unit 500. As the elevator evaluation index 161, in order to evaluate both the operating status of the elevator and the movement of people, an index affected by both is used. For example, the items of the elevator evaluation index 161 include items such as floors 1611, average call duration 1612, call button press count 1613, average number of passengers 1614, and average door opening time 1615.
[0060]
　The floor 1611 stores information (numbers, character strings, etc.) that can identify the floor. The average call duration 1612 stores information indicating the average value of the time from the pressing of the elevator call button to the arrival of the elevator car. The call button press count 1613 stores information indicating the cumulative value of the number of times the elevator call button is pressed. The average number of passengers 1614 stores information indicating the average value of the number of passengers estimated from the change in the load inside the elevator car. The average door opening time 1615 stores information indicating the average value of the time from the opening of the elevator door to the closing of the elevator door.
[0061]
　Note that the elevator evaluation index 161 may be recorded for each of the up and down directions of the call button. Further, the information may be recorded separately for each elevator group controlled by the same call button group. Further, the entire time zone to be evaluated may be calculated and recorded, or the time zone may be divided and calculated and recorded.
[0062]
　Next, the pedestrian flow evaluation index 162 will be described. The pedestrian flow evaluation index 162 is an valuation index related to the pedestrian flow calculated by the in-building pedestrian flow simulation unit 300 and the pedestrian flow measurement unit 600. For example, the pedestrian flow evaluation index 162 is configured to include items such as the time 1621 and the waiting number 1622.
[0063]
　At time 1621, information indicating the time when the record is recorded is stored. The waiting person number 1622 stores information indicating the number of waiting persons in the elevator hall on each floor. The pedestrian flow evaluation index 162 is recorded, for example, when the door of the elevator is opened or closed. Further, the traffic flow evaluation index 162 may be recorded periodically, or the maximum value, the average value, etc. of the waiting persons in each cycle may be used as the evaluation index. The pedestrian flow evaluation index 162 may be recorded for each elevator hall or for each elevator when there are elevator halls or a plurality of elevators on each floor.
[0064]
　Each of the first evaluation index 160A and the second evaluation index 160B includes, as evaluation items, an average call duration 1612, a call button press count 1613, an average number of passengers 1614, and an average door opening time 1615. , And at least one of the waiting persons 1622 is included in common.
[0065]

　Next, the process of each unit will be described. It should be noted that publicly-known technologies may be used for the in-building traffic estimation unit 200, the in-building pedestrian simulation unit 300, the elevator index calculation unit 500, the pedestrian flow measurement unit 600, and the output unit 800, and thus the description of the processing will be omitted. ..
[0066]
　First, an example of a processing procedure related to processing executed by the parameter value calculation unit 400 will be described with reference to FIG. 9.
[0067]
　In step S2001, the parameter value calculation unit 400 totals (acquires) the elevator operation log 110.
[0068]
　Subsequently, the parameter value calculation unit 400 calculates the traveling time and the traveling distance between the floors of the elevator from the elevator operation log 110, and creates traveling result data that is information that combines the traveling time and the traveling distance (step). S2002). For example, the parameter value calculation unit 400 obtains, from the elevator status log 111, a record at the start of traveling of a record in which the door state 1117 is “closed”, and immediately after that, a record in which the door state 1117 of the same elevator is “open”. The data when the traveling is stopped, the difference between the time 1113 between the traveling stop and the traveling start is calculated as the traveling time, the difference between the position 1115 between the traveling stop and the traveling start is calculated as the traveling distance, and the combination thereof is calculated. Is the travel performance data.
[0069]
　Subsequently, the parameter value calculation unit 400 obtains the rated acceleration and the rated speed by regression analysis of the relationship between the traveling time and the traveling distance (step S2003). At this time, as shown in FIG. 10, when the travel record data is shown in a scatter diagram of the travel time and the travel distance, it is divided into a non-linear range 2101 where the rated speed is not reached and a linear range 2102 where the rated speed is reached. Can be divided.
[0070]
　In the range 2101 where the rated speed is not reached, the speed of the elevator changes as shown by the graph 2103 in FIG. Since the travel distance x is a value obtained by time integration of the speed v, the relationship between the travel distance x and the travel time t can be expressed by the following equation (1) with the rated acceleration a.
[Number 1]

[0071]
　Similarly, in the range 2102 in which the rated speed is reached, the speed of the elevator changes as shown by a graph 2104, and the relationship between the travel time t and the traveled distance x is expressed by the following equation, where the rated acceleration is a and the rated speed is V. It can be represented as (2).
[Number 2]

[0072]
　Therefore, the parameter value calculation unit 400 defines the set of travel record data included in the range that does not reach the rated speed as N and the set of travel record data included in the range that reaches the rated speed as R in the following formula (3). Find a and V that minimize the sum of error squares shown.
[Number 3]

[0073]
　Since it is considered that a and V do not greatly deviate from the specification value of the elevator, for example, the search for the minimum value can be obtained by searching all values ​​within a certain range from the specification value. Alternatively, it may be obtained using an optimization method such as a genetic algorithm.
[0074]
　Subsequently, the parameter value calculation unit 400 outputs a and V for which the sum of squared errors obtained in step S2003 is the minimum as the rated acceleration and the rated speed, respectively (step S2004). The output rated acceleration and rated speed are used as the elevator parameter value 150.
[0075]
　If the elevator operation log 110 is recorded during traveling of the elevator, the rated acceleration and the rated speed may be calculated from the relationship between the traveling time and the position.
[0076]
　Further, the parameter value calculation unit 400 is characterized by automatically calculating and setting the elevator parameter value 150 that can be calculated from the elevator operation log 110, and the parameter value to be calculated is limited to the rated speed and the rated acceleration. It is not something that can be done. For example, when the time when the elevator arrives at each floor and the time when the door is completely opened immediately after that are recorded in the elevator operation log 110, the average value of the difference may be obtained as the door opening/closing time 1517.
[0077]
　Next, an example of a processing procedure related to processing executed by the evaluation unit 700 will be described with reference to FIG.
[0078]
　In step S2201, the evaluation unit 700 sets the first evaluation index 160A output by the in-building pedestrian flow simulation unit 300 and the second evaluation index 160B output by at least one of the elevator index calculation unit 500 and the pedestrian flow measurement unit 600. As an input, the error of each evaluation item (for each element) of the first evaluation index 160A and the second evaluation index 160B is calculated.
[0079]
　Here, each evaluation item is, for example, the average call duration 1612, which is the evaluation index of the elevator, the number of times the call button is pressed 1613, the average number of passengers 1614, the average door opening time 1615, and the number of waiting people 1622, which is an evaluation index of the flow of people. That is. As the error, for example, an absolute error of each evaluation item included in the first evaluation index 160A and the second evaluation index 160B is used. If it is possible to subdivide into smaller units such as time zone and floor, such as the number of people waiting, the absolute error of each subdivided element may be directly used. Values, sums, etc. may be used instead.
[0080]
　Subsequently, the evaluation unit 700 calculates an error evaluation value by totaling the errors of each evaluation item (step S2202). As the error evaluation value, for example, the sum of the absolute value of the error of each evaluation item and the weight of each evaluation item can be used. When the set of evaluation items is I, the error of the evaluation item i is E i , and the weight is W i , the error evaluation value E total can be expressed by the following formula (4).
[Number 4]

[0081]
　Here, the weight W i is a value used for the purpose of normalization for equalizing the size of the value of each evaluation item or weighting for evaluation. For example, it is possible to normalize by using the reciprocal of the average value of the values ​​of the respective evaluation items in the second evaluation index 160B.
[0082]
　As the error evaluation value, one or more types of evaluation items of the evaluation index may be used as necessary, and it is not always necessary to use all the evaluation items.
[0083]
　Subsequently, the evaluation unit 700 determines whether the calculated error evaluation value is equal to or less than a preset threshold value (step S2203). The evaluation unit 700 moves the process to step S2207 when it determines that the difference is less than or equal to the threshold, and moves the process to step S2204 when it determines that the difference is greater than the threshold.
[0084]
　In step S2204, the evaluation unit 700 determines a parameter to be adjusted (adjustment parameter, in other words, an input parameter for changing an input value). For example, the evaluation unit 700 adjusts based on the error of each evaluation item calculated in step S2202 and the correlation table 2400 in which the correlation coefficient indicating the correlation between the evaluation item 2401 and the adjustment candidate parameter 2402 illustrated in FIG. 13 is recorded. Determine (select) parameters.
[0085]
　For example, the evaluation unit 700 calculates, for each adjustment candidate parameter 2402, a parameter evaluation value that is the sum of values ​​obtained by multiplying the correlation coefficient by the error of the evaluation item, and the adjustment candidate with the maximum absolute value of the parameter evaluation value. The parameter 2402 is selected as the adjustment parameter. For the correlation coefficient recorded in the correlation table 2400, for example, a value calculated from past simulation results may be used, or an assumed value may be used.
[0086]
　As the adjustment candidate parameter 2402, a parameter of arrival distribution data 130 (for example, an average value of arrival distribution), a parameter of an elevator parameter value (for example, the number of moving people), an internal parameter value of a pedestrian flow simulation unit in a building (for example, boarding rate). Parameters and the like.
[0087]
　Subsequently, the evaluation unit 700 updates the parameter value (input value) (step S2205). For example, when the parameter evaluation value is positive, the evaluation unit 700 sets the value reduced by a predetermined value as the next parameter value, and when it is negative, the value increased by the predetermined value as the next parameter value. ..
[0088]
　Further, some values ​​may be used as candidates for the next parameter value and may be sequentially or in parallel proceeded to the processing from step S2206. For example, the evaluation unit 700 may use some of the following parameter values ​​that have changed a predetermined value for changing the parameter value as candidates, or may use both the case of increasing the parameter value and the case of decreasing the parameter value. .. In addition, for example, when the arrival distribution data 130 is changed, the evaluation unit 700 may update the average or variance of the arrival distribution as the parameter value.
[0089]
　In the evaluation unit 700, the first elevator whose traveling direction is the first direction has started without being full, and the second elevator whose traveling direction is the same as the first direction has arrived subsequently. If it is determined that the arrival distribution data 130 is changed, the arrival distribution data 130 is limited to the time range from the time when the call button for calling the second elevator is pressed to the time when the second elevator arrives. May be.
[0090]
　Subsequently, the evaluation unit 700 outputs the new condition with the changed parameter value to the in-building flow simulation unit 300 (step S2206). The in-building pedestrian flow simulation unit 300 calculates the first evaluation index 160A under the new condition. In addition, when changing the arrival distribution data 130 as the parameter value to be changed, the evaluation unit 700 outputs the updated arrival distribution data 130 to the in-building traffic estimation unit 200, and the in-building traffic estimation unit 200 outputs the in-building traffic data. After re-estimating 140, the re-estimated intra-building traffic data 140 is output to the intra-building pedestrian simulation unit 300, and the process proceeds to step S2201.
[0091]
　In step S2207, the evaluation unit 700 outputs an error evaluation value in addition to the pedestrian flow (simulation result) in the building estimated by the pedestrian flow simulation unit 300. In addition to the error evaluation value, the error of each evaluation item of the evaluation index, the used parameter, the input value, etc. may be output.
[0092]
　The processing of the evaluation unit 700 is not limited to the above. For example, the end condition of the parameter value changing process may be to update and evaluate the parameter value a predetermined number of times, not whether the error evaluation value is equal to or less than the threshold value. In this case, the parameter value with the smallest error evaluation value may be output.
[0093]
　Further, for example, the method for determining the adjustment parameter is not limited to the above. For example, a combination of evaluation items and adjustment candidate parameters may be created in advance as a table, and the adjustment candidate parameters corresponding to the evaluation items having the largest error in the evaluation items may be used as the adjustment parameters. That is, the evaluation unit 700 determines the adjustment parameter based on the error that is the difference between the evaluation items included in the first evaluation index 160A and the second evaluation index 160B.
[0094]
　With the above processing, the in-building pedestrian flow estimation system 1 can output an error evaluation value as an evaluation value of the degree of reproduction between the operation of the elevator and the flow of people. Thereby, the degree of reproduction can be easily compared using the error evaluation value.
[0095]
　Further, since Patent Document 3 does not describe a configuration capable of acquiring the actual number of people waiting for a hall, even if the number of people waiting for a hall is estimated, it cannot be determined whether the estimation result reproduces the reality. In this regard, in the present embodiment, the pedestrian flow measurement unit 600 measures the pedestrian flow evaluation index 162, and therefore it is possible to determine whether the estimation result reproduces the reality.
[0096]
　Also, with the device described in Patent Document 3, it is difficult to find a simulation condition that can correctly reproduce the reality when it is determined that the reality is different. In this respect, in the present embodiment, it is possible to automatically calculate a highly evaluated simulation condition using the error evaluation value as an objective function and output the result. This makes it possible to obtain an estimation result with a better degree of reproduction.
[0097]
(2) Other Embodiments In the
　above-described embodiments, the case where the present invention is applied to the in-building pedestrian flow estimation system 1 has been described, but the present invention is not limited to this, and various other embodiments. It can be widely applied to the building flow estimation system.
[0098]
　Further, in the above-described embodiment, in step S2202, the error evaluation value is described as an example of the evaluation value, but the present invention is not limited to this, and the largest error among the errors of each evaluation item is the evaluation value. , The average value of the errors may be used as the evaluation value, and another value calculated based on the error may be used as the evaluation value.
[0099]
　Further, in the above-described embodiment, the case where the parameter value calculation unit 400 calculates the rated speed and the rated acceleration to be the elevator parameter value 150 has been described, but the present invention is not limited to this, and the parameter value calculation unit 400 is not limited to this. The elevator parameter value 150 may be at least one of the rated speed and the rated acceleration.
[0100]
　The configurations described above can be changed, combined, or omitted without departing from the scope of the present invention.
Explanation of symbols
[0101]
　1... In-building pedestrian flow estimation system, 200... In-building traffic estimation unit, 300... In-building pedestrian flow simulation unit, 400... Parameter value calculation unit, 500... Elevator index calculation unit, 600... People flow measurement unit, 700...Evaluation section, 800...Output section.
The scope of the claims
[Claim 1]
　Using the elevator in the building as an input value, the intra-building traffic data indicating which floor a person moves from to which floor, the building layout data related to the layout of the building, and the elevator parameter value related to the elevator are input. An in-building human flow estimation system having an in-building
　human flow simulation unit for estimating a flow, comprising: a first evaluation index relating to a human flow estimated by the in-building human flow simulation unit; and an actual human flow in the building. An evaluation unit that calculates an evaluation value for evaluating the degree of reproducibility of the flow of the person estimated by the in-building human flow simulation unit based on the second evaluation index related to the flow, and the calculation unit that calculates the evaluation value
　.
　An in-building pedestrian flow estimation system comprising: an output unit that outputs an evaluation value .
[Claim 2]
　Each of the first evaluation index and the second evaluation index has, as an evaluation item, an average call continuation indicating an average value of time from when the elevator call button is pressed to when the elevator car arrives. Time, the number of times the call button is pressed indicating the cumulative value of the number of times the elevator call button has been pressed, the average number of passengers indicating the average value of the number of passengers estimated from the change in the load in the car of the elevator, and The occupant
　flow in a building according to claim 1, wherein at least one of an average door opening time indicating an average value of time from opening to closing of the elevator door and the number of people waiting in the elevator hall is included. Estimation system.
[Claim 3]

　The building according to claim 2 　, wherein the evaluation unit calculates the evaluation value based on an error that is a difference between the evaluation items in the first evaluation index and the second evaluation index. Inward flow estimation system.
[Claim 4]
　The evaluation unit determines whether or not the evaluation value satisfies a predetermined threshold value, and when it is determined that the evaluation value does not satisfy the predetermined threshold value, the in-building traffic data, the elevator parameter value, and an internal parameter of the in-building pedestrian simulation unit. The
　in-building human flow estimation system according to claim 1, wherein at least one input value of the values ​​is changed, and the changed input value is output to the in-building human flow simulation unit .
[Claim 5]
　The evaluation unit is based on an error which is the difference of each evaluation item in the first evaluation index and second evaluation index, input parameters to determine, to change the input value
　claim 4, characterized in that The in-building pedestrian flow estimation system described in.
[Claim 6]
　A correlation coefficient indicating a correlation between each evaluation item in each of the first evaluation index and the second evaluation index and an input parameter of a candidate whose input value is to be changed is provided, and the
　evaluation unit is configured to based on the error and the correlation coefficient is a difference of each evaluation item in the evaluation index and the second evaluation index of 1, to determine the input parameters to change the input value,
　it in claim 5, wherein In-building pedestrian flow estimation system.
[Claim 7]
　Estimating the in-building traffic data based on an elevator operation log indicating the operation record of the elevator and arrival distribution data indicating how fast people arrive at the elevator hall with respect to the time of boarding the elevator with the inside of the traffic estimator buildings, for
　the evaluation unit, by change the arrival distribution data output to the in-building transportation estimation unit, changing the input value as the building within traffic data,
　characterized in that The human flow estimation system in a building according to claim 4.
[Claim 8]
　In the evaluation unit, the first elevator whose traveling direction is the first direction has started without being full, and the second elevator whose traveling direction is the same direction as the first direction has arrived subsequently. When it is determined, the arrival distribution data is changed by limiting the time range of the arrival distribution data to the range from the time when the call button for calling the second elevator is pressed to the time when the second elevator arrives. The in-
　building pedestrian flow estimation system according to claim 7, wherein
[Claim 9]
　The evaluation unit is configured as a change in intrinsic parameters, changes the rate boarded indicating an upper limit of a ratio calculated from a number of people riding in the capacity and the car of the elevator in the car of the elevator,
　characterized in that The human flow estimation system in a building according to claim 4.
[Claim 10]

　The building according to claim 1 　, further comprising: a pedestrian flow measurement unit that calculates a congestion index based on a result of measuring a person who uses the elevator, and uses the calculated congestion index as the second evaluation index. Inward flow estimation system.
[Claim 11]
　11.
　The building according to claim 10, wherein the pedestrian flow measurement unit includes a camera and an analysis device that analyzes the number of people included in a predetermined range in a camera image of the camera . People flow estimation system.
[Claim 12]
　The
　in-building pedestrian flow estimation system according to claim 10, wherein the pedestrian flow measurement unit uses the measured number of people waiting in the elevator hall as the congestion index .
[Claim 13]
　An elevator index calculation unit that calculates an evaluation index for evaluating the operation status of the elevator based on an elevator operation log indicating the operation record of the elevator, and uses the calculated evaluation index as the second evaluation index.
　The in-building pedestrian flow estimation system according to claim 1, wherein.
[Claim 14]
　At least one of the rated speed and the rated acceleration of the elevator is calculated as the elevator parameter value based on the data of the position of the elevator and the time corresponding to the position included in the elevator operation log indicating the operation record of the elevator. The
　in-building pedestrian flow estimation system according to claim 1 , further comprising a parameter calculation unit .
[Claim 15]
　Using the elevators in the building as the input values ​​of the intra-building traffic data indicating which floor a person moves from to which floor, the building layout data related to the layout of the building, and the elevator parameter value related to the elevator are input values. An estimation method in an in-building pedestrian flow estimation system having an in-building pedestrian flow simulation unit for estimating a flow
　, wherein the evaluation unit comprises: A first step of calculating an evaluation value for evaluating the degree of reproduction of the human flow estimated by the in-building human flow simulation unit based on the second evaluation index related to the actual human flow in the building And a
　second step in which the output unit outputs the evaluation value calculated in the first step,
　the estimation method.